Arbitrator's report on Nebraska-Kansas Republican River fight - KHGI-TV/KWNB-TV/KHGI-CD-Grand Island, Kearney, Hastings

Arbitrator's report on Nebraska-Kansas Republican River fight

Posted: Updated:

NON-BINDING ARBITRATION

Pursuant to Arbitration Agreement of October 23, 2008

IN ACCORDANCE WITH:

FINAL SETTLEMENT STIPULATION

Kansas v. Nebraska and Colorado

No. 126, Original, U.S. Supreme Court

Decree of May 19, 2003, 538 U.S. 720

ARBITRATOR'S FINAL DECISION

June 30, 2009

1

BACKGROUND

On December 15, 2002, the states of Kansas, Nebraska, and Colorado (the "States") executed the

Final Settlement Stipulation (the "FSS") "... to resolve the currently pending litigation in the

United States Supreme Court regarding the Republican River Compact by means of this

Stipulation and the Proposed Consent Judgment ... ." FSS, Volume 1 of 5, at 1. The FSS was

filed with the Special Master appointed by the U.S. Supreme Court (the "Court") in Kansas v.

Nebraska and Colorado, No. 126, Original, who recommended entry of the proposed consent

judgment which would approve the FSS. Second Report of the Special Master (Subject: Final

Settlement Stipulation) at 77. On May 19, 2003, the Court entered a consent decree approving

the FSS (the "Consent Decree").

By 2007, disputes arose between the States regarding compliance with the FSS and the

Republican River Compact (the "Compact"). The disputes were submitted to the Republican

River Compact Administration (the "RRCA") pursuant to the provision in the FSS for dispute

resolution. See FSS, Volume 1 of 5, § VII., at 34-40. The RRCA addressed the disputes, but no

resolution of certain disputes was reached. See Resolution of the RRCA dated May 16, 2008;

Exhibit 1 to Arbitration Agreement dated October 23, 2008. The RRCA submitted these

disputes to non-binding arbitration pursuant to the provisions of § VII. of the FSS, the States

executed the Arbitration Agreement on October 23, 2008 (the "Arbitration Agreement"), and I

was retained by the States to serve as the Arbitrator.

Exhibit 2 to the Arbitration Agreement sets forth the "Time Frame Designation" for the nonbinding

arbitration, Exhibit 3 to the Arbitration Agreement sets forth the disputed issues

identified by the State of Kansas to be arbitrated, and Exhibit 4 to the Arbitration Agreement sets

forth the disputed issues identified by the State of Nebraska to be arbitrated. The disputed issue

originally raised by the State of Colorado with the RRCA, which the RRCA submitted to nonbinding

arbitration pursuant to the provisions of § VII. of the FSS (See Attachment 3 to

Resolution of the RRCA dated May 16, 2008), has been withdrawn from this arbitration and is

not included in the Arbitration Agreement.

From the issues set forth in Exhibit 3 and Exhibit 4 to the Arbitration Agreement, the States

identified six legal issues to be decided by the Arbitrator by December 19, 2008, for the purpose

of narrowing discovery and the hearing on the merits. Based on a disagreement regarding the

appropriate scope of the arbitration, the Arbitrator identified a seventh legal issue during a

prehearing conference held telephonically on November 5, 2008. Each of the States filed

opening briefs on these seven legal issues with the Arbitrator on November 10, 2008. (The State

of Colorado briefed 3 arguments pertaining to only 4 of the legal issues.) Responsive briefs were

filed on November 24, 2008, and reply briefs were filed on December 5, 2008. Oral argument

on these legal issues was heard at the University of Denver, Strum College of Law, on December

10, 2008.

The Arbitrator treated the briefs filed by the States as being analogous to cross-motions for

summary judgment under Rule 56 of the Federal Rules of Civil Procedure. "A party claiming

relief may move, with or without supporting affidavits, for summary judgment on all or part of

the claim." Fed. R. Civ. P. 56(a). "The judgment sought should be rendered if the pleadings, the

2

discovery and disclosure materials on file, and any affidavits show that there is no genuine issue

as to any material fact and that the movant is entitled to judgment as a matter of law." Fed. R.

Civ. P. 56(c).

The Arbitrator issued his preliminary decision on these seven legal issues, including a summary

of his reasons for deciding each issue, on December 19, 2008. On January 22, 2009, the

Arbitrator issued his final decision on these seven legal issues. With minor corrections and the

addition of supporting analysis for each of the seven issues, the final decision is materially the

same as the preliminary decision issued on December 19, 2008. The Arbitrator's Final Decision

on Legal Issues is attached hereto1 and fully incorporated herein by reference.

The States submitted expert reports on the remaining issues to the Arbitrator in lieu of extensive

direct testimony on February 23, 2009. The Arbitrator subsequently conducted a hearing on

those issues at the Byron Rogers U. S. Courthouse in Denver, Colorado, beginning on March 9,

2009. The hearing was recessed on March 19, 2009, and reconvened and concluded on April 14,

2009. The Arbitrator has carefully considered the reports and testimony of the expert witnesses

for the States together with post-hearing briefs submitted by counsel for the States and issues the

following decision.

FINDINGS

Accounting Procedures - Estimating Computed Beneficial Consumptive Use for Groundwater

and Imported Water Supply

1. The Final Settlement Stipulation (the "FSS") executed by the States on December 15, 2002,

and approved by the U. S. Supreme Court on May 19, 2003, incorporates detailed

Accounting Procedures and Reporting Requirements ("Accounting Procedures"), which were

subsequently adopted and revised by the Republican River Compact Administration (the

"RRCA")2, as provided in § I.F. of the FSS. The adopted Accounting Procedures, as revised,

include procedures for estimating Computed Beneficial Consumptive Use ("CBCU") for

groundwater and determining the Imported Water Supply Credit ("IWS").

2. In their respective post-hearing briefs (each titled Post-Trial Brief),3 counsel for the states of

Colorado and Kansas assert that the issue of estimating CBCU of groundwater and

determining the IWS is not a proper subject for this arbitration because Nebraska's expert

1 The date in the first line of the attached Arbitrator's Final Decision on Legal Issues, dated January 22, 2009, has

been corrected to December 15, 2002.

2 Final Settlement Stipulation, Volume 1 of 5, Appendix C, as revised (July 2005) and adopted (August 10, 2006)

by the RRCA.

3 Counsel for Colorado, Kansas, and Nebraska signed and submitted briefs by FedEx sent on April 24, 2009.

3

report on this issue4 has not been submitted to the RRCA for its consideration,5 and therefore,

the Arbitrator should not consider the issue.

3. Exhibits 1, 3, and 4 of the Arbitration Agreement executed by each of the States on October

23, 2008, identify the procedures used to estimate CBCU of groundwater and determine the

IWS as a disputed issue "which may be taken to the next step in the dispute resolution

process"6 and an issue "to be Arbitrated."7

4. The difference between what Colorado and Kansas contend was submitted to the RRCA and

included in the Arbitration Agreement, as compared with what is before the Arbitrator, is the

weighting coefficients proposed by Nebraska to be applied to results from 8 differences

calculated using 16 runs of the RRCA Groundwater Model.8 Although the weighting

coefficients involved in the proposal currently before the Arbitrator are different than the

equal weighting coefficients resulting from averaging the 8 differences, which was the

approach presented to the RRCA in August of 2008,9 Nebraska's proposal to use

8 differences calculated using 16 runs of the RRCA Groundwater Model is essentially the

same as it was in August of 2008.

5. Prior to submitting their respective post-hearing briefs, neither Colorado nor Kansas asserted

that because Nebraska's expert report on this issue had not been submitted to the RRCA for

its consideration, the issue of estimating CBCU of groundwater and determining the IWS

was not a proper subject for this arbitration. Neither Colorado nor Kansas timely made this

assertion when they submitted their respective expert reports10, 11 in response to Nebraska's

expert report on this issue, and neither timely raised this assertion during the hearing

conducted from March 9 through March 19 and on April 14, 2009. Therefore, Nebraska's

4 Nebraska Exhibit 30, Expert Report of Dr. David P. Ahlfeld, Michael G. McDonald, and James C. Schneider,

Estimating Computed Beneficial Consumptive Use for Groundwater and Imported Water Supply under the

Republican River Compact, January 20, 2009.

5 State of Colorado's Post-Trial Brief at 30-33; Kansas' Post-Trial Brief at 65-66.

6 Exhibit 1 of the Arbitration Agreement, see Attachment 2: Commissioner Dunnigan's letter to Commissioners

Barfield and Wolfe dated April 15, 2008.

7 Exhibit 3 and Exhibit 4 of the Arbitration Agreement.

8 State of Colorado's Post-Trial Brief at 32; Kansas' Post-Trial Brief at 65; State of Nebraska's Post-Hearing

Brief at 43 and 49.

9 Id.

10 Colorado Exhibit 7, Expert Report of Willem A. Schreüder, Ph.D., Report in Response to: Estimating Computed

Beneficial Consumptive Use for Groundwater and Imported Water Supply under the Republican River Compact,

Ahfed [sic] et al. (January 20, 2009), February 16, 2009.

11 Kansas Exhibit 28, Expert Report of David W. Barfield, Steven P. Larson, and Dale E. Book, Kansas's Expert

Response to Nebraska's Expert Report, "Estimating Computed Beneficial Use for Groundwater and Imported

Water Supply under the Republican River Compact," February 17, 2009.

4

issue of estimating CBCU of groundwater and determining the IWS, as presented in its

expert report,4 is properly included as an issue in this arbitration.

6. Subsection III.A.1. of the Accounting Procedures specifies how the annual Virgin Water

Supply for each sub-basin is to be determined as follows:

The annual Virgin Water Supply for each Sub-basin will be calculated by adding: a) the

annual stream flow in that Sub-basin at the Sub-basin stream gage designated in Section

II., b) the annual Computed Beneficial Consumptive Use above that gaging station, and

c) the Change in Federal Reservoir Storage in the Sub-basin; and from that total subtract

any Imported Water Supply Credit. The Computed Beneficial Consumptive Use will be

calculated as described in Subsection III. D.

7. Subsection III.A.2. of the Accounting procedures specifies how the annual Virgin Water

Supply for main stem is to be calculated as follows:

The annual Virgin Water Supply for the Main Stem will be calculated by adding: a) the

flow at the Hardy gage minus the flows from the Sub-basin gages listed in Section II, b)

the annual Computed Beneficial consumptive Use in the Main Stem, and c) the Change

in Federal Reservoir Storage from Swanson Lake and Harlan County Lake; and from that

total subtract any Imported Water Supply Credit for the Main Stem.

8. Section II. of the Accounting Procedures define the terms Virgin Water Supply, Computed

Beneficial Consumptive Use, and Imported Water Supply Credit as follows:

Virgin Water Supply: the Water Supply within the Basin undepleted by the activities of

man;

Computed Beneficial Consumptive Use: for purposes of Compact accounting, the

stream flow depletion resulting from the following activities of man:

Irrigation of lands in excess of two acres;

Any non-irrigation diversion of more than 50 Acre-feet per year;

Multiple diversions of 50 Acre-feet or less that are connected or

otherwise combined to serve a single project will be considered as a

single diversion for accounting purposes if they total more than 50

Acre-feet;

Net evaporation from Federal Reservoirs;

Net evaporation form Non-federal Reservoirs within the surface

boundaries of the Basin;

Any other activities that may be included by amendment of these

formulas by the RRCA;

Imported Water Supply Credit: the accretions to stream flow due to water imports

from outside of the Basin as computed by the RRCA Groundwater Model. The Imported

Water Supply Credit of a State shall not be included in the Virgin Water Supply and shall

be counted as a credit/offset against the Computed Beneficial Consumptive Use of water

allocated to that State ...

5

9. Subsection III.D.1. of the Accounting Procedures specifies how Computed Beneficial

Consumptive Use of groundwater is to be determined for an accounting year as follows:

Computed Beneficial Consumptive Use of groundwater shall be determined by use of the

RRCA Groundwater Model. The Computed Beneficial Consumptive Use of groundwater

for each State shall be determined as the difference in streamflows using two runs of the

model:

The "base" run shall be the run with all groundwater pumping, groundwater pumping

recharge, and surface water recharge within the model study boundary for the period

1940 to the current accounting year "on".

The "no State pumping" run shall be the run with the same model inputs as the base run

with the exception that all groundwater pumping and pumping recharge of that State shall

be turned "off."

10. Subsection III.A.3. of the Accounting Procedures specifies how the Imported Water Supply

Credit is to be determined for an accounting year as follows:

The amount of Imported Water Supply Credit shall be determined by the RRCA

Groundwater Model. The Imported Water Supply Credit of a State shall not be included

in the Virgin Water Supply and shall be counted as a credit/offset against the Computed

Beneficial Consumptive Use of water allocated to that State. Currently, the Imported

Water Supply Credits shall be determined using two runs of the RRCA Groundwater

Model:

a. The "base" run shall be the run with all groundwater pumping, groundwater

pumping recharge, and surface water recharge within the model study

boundary for the period 1940 to the current accounting year turned "on."

This will be the same "base" run used to determine groundwater Computed

Beneficial Consumptive Uses.

b. The "no NE import" run shall be the run with the same model inputs as the

base run with the exception that surface water recharge associated with

Nebraska's Imported Water Supply shall be turned "off."

The Imported Water Supply Credit shall be the difference in stream flows between these

two model runs.

11. Nebraska has proposed essentially three changes in the Accounting Procedures adopted by

the RRCA involving computation of CBCU for groundwater and IWS that would modify

(1) the annual calculation of Virgin Water Supply ("VWS") in each Sub-basin and the Main

Stem; (2) the annual determination of CBCU in each Sub-basin and the Main Stem; and

(3) the annual determination of the IWS in each Sub-basin and the Main Stem.4 None of

these changes have been adopted by the RRCA, as provided in § I.F. of the FSS , and are at

issue in this arbitration pursuant to § VII.A., ¶ 1. and ¶ 7., of the FSS.

12. The calculation of annual VWS for any Sub-basin, as specified in § III.A.1. of the

Accounting Procedures and described in Finding 6 is:

6

VWS = Gage + CBCU + S - IWS.

Alternatively, this relationship can be written:

VWS = Gage + CBCUS + CBCUG + S - IWS

or

VWS = Gage + CBCUS + (CBCUC + CBCUK + CBCUN) + S - IWS

In these relationships, "Gage" is the annual streamflow in that Sub-basin measured at the

stream gage designated in § II. of the Accounting Procedures, CBCU is the computed

depletion of streamflow in that Sub-basin from all Beneficial Consumptive Use, and S is

the Change in Federal Reservoir Storage. Using the notation of Nebraska,4 CBCUS is the

computed depletion of streamflow in that Sub-basin from all Beneficial Consumptive Use of

surface water, CBCUG is the computed depletion of streamflow in that Sub-basin from all

Beneficial Consumptive Use of groundwater, CBCUC is the computed depletion of

streamflow in that Sub-basin from all Beneficial Consumptive Use of groundwater by

Colorado, CBCUK is the computed depletion of streamflow in that Sub-basin from all

Beneficial Consumptive Use of groundwater by Kansas, and CBCUN is the computed

depletion of streamflow in that Sub-basin from all Beneficial Consumptive Use of

groundwater by Nebraska.

13. The calculation of annual VWS for the Main Stem, as specified in § III.A.2. of the

Accounting Procedures and described in Finding 7 is the same as shown in Finding 12 except

the from the "Gage" (which for the Main Stem is the annual streamflow measured at the

Hardy gage), the sum of the annual streamflows measured at all Sub-basin gages upstream of

the Hardy gage is subtracted.

14. The first change proposed by Nebraska in the Accounting Procedures pertaining to CBCUG

and IWS would modify the determination VWS in Finding 12 to:

VWS = VWSS + VWSG

where

VWSG = (θ - CKMN).

In these relationships, again using the notation of Nebraska,4 VWSS is the surface-waterrelated

portion of VWS, VWSG is the groundwater-related portion of VWS, θ is the annual

base flow in a Sub-basin or the Main Stem determined from running the RRCA Groundwater

Model with all groundwater pumping, groundwater pumping recharge, and surface water

recharge within the model study boundary for the period 1940 to a particular accounting year

"off," and CKMN, is the base flow in a Sub-basin or the Main Stem determined from running

the RRCA Groundwater Model with all Colorado groundwater pumping and recharge (C),

7

Kansas groundwater pumping and recharge (K), all surface water recharge from Imported

Water Supply (M), and all Nebraska groundwater pumping and recharge (N) within the

model study boundary for the period 1940 to a particular accounting year "on."

15. The reason stated by Nebraska for the proposed change in determining VWS is: "This

independently-computed value of VWSG is the best estimate of the impact of all

groundwater-related human activity on streamflow and should be viewed as the true value of

this property."12

16. While the independently-computed value of VWSG (θ - CKMN) may be the best estimate of

base flow discharged from the groundwater system to surface water sources "undepleted by

the activities of man" over the period 1940 to a particular accounting year, it is an estimated

value derived from running the RRCA groundwater model and should not be viewed as the

"true value" as suggested by Nebraska. Although the RRCA Groundwater Model has

presumably been properly designed and calibrated and can provide reliable estimates of base

flow, the RRCA groundwater model is still an idealization of a complex hydrogeologic

system, and the results derived from running the model are not necessarily the true values.

17. The second and third changes proposed by Nebraska in the Accounting Procedures pertaining

to CBCUG and IWS would modify the determination of CBCUC, CBCUK, and CBCUN

specified in § III.D.1. of the Accounting Procedures as described in Finding 9 and the

determination of IWS specified in § III.A.3. of the Accounting Procedures described in

Finding 10 such that:

CBCUC + CBCUK + CBCUN - IWS = (θ - CKMN) = VWSG

under all conditions.

18. As described in Findings 9 and 10, the current Accounting Procedures require differencing

the results from two runs of the RRCA Groundwater Model (requiring 5 runs of the RRCA

Groundwater Model) to determine each of the four man-caused stresses to the groundwater

system; i.e., Colorado groundwater consumptive use (CBCUC), Kansas groundwater

consumptive use (CBCUK), Nebraska groundwater use (CBCUN), and recharge from

imported surface water (IWS). Nebraska proposes differencing the results from 16 runs of

the RRCA Groundwater Model (8 differences) for each of the four man-caused stresses to the

groundwater system and summing the 8 differences using weighting factors, which weighting

factors sum to one, for each of the four man-caused stresses such that the relationship in

Finding 17 is satisfied.13

12 Nebraska Exhibit 30, Expert Report of Dr. David P. Ahlfeld, Michael G. McDonald, and James C. Schneider,

Estimating Computed Beneficial Consumptive Use for Groundwater and Imported Water Supply under the

Republican River Compact, January 20, 2009, p. 9.

13 Id., p. 48. Also, see Nebraska Exhibit 33.

8

19. The reasons stated by Nebraska for the proposed changes in determining CBCUC, CBCUK,

CBCUN, and IWS include:

... the current Accounting Procedures assume that VWSG can be computed using the

individually-computed impacts in a sub-basin (CBCUC, CBCUK, CBCUN and IWS) as

VWSG = CBCUC + CBCUK + CBCUN - IWS14

... under some stream drying conditions, the current Accounting Procedures do not

produce values that combine to the independently-computed value of VWSG. This leads

to the conclusion that the values of CBCUC, CBCUK, CBCUN and IWS computed using

the current Accounting Procedures are in error.15

The deviation from additivity can be substantial and is of critical importance since this

additivity is assumed to hold under the current Accounting Procedures.16

The selection of the additional model runs to be used is based on the idea that using a

base condition with any one human activity either on or off may bias the results for or

against one state. ... As a result, analysis should be performed using all possible base

conditions in which human activities are either on or off.17

The proposed method provides values for impact that satisfy the expectation that

individual impacts will sum to the total impact of human activity for a given sub-basin.18

20. In the context of the changes proposed by Nebraska, "additivity" means that the relationship

described in Finding 17 is valid under all conditions. The "error" or "deviation from

additivity" asserted by Nebraska occurs when modeled groundwater use by any of the three

States, individually or in combination, fully depletes streamflow. That is, so long as

groundwater-caused depletions to a flowing stream do not cause streamflow to approach

zero, an increase or decrease in the use of groundwater that is hydraulically connected to the

stream will result in a decrease or increase in streamflow, respectively, that essentially is

linearly proportionate19 to the increase or decrease in groundwater use. The modeled

response of the stream is basically linear and the condition of "additivity" holds when

CBCUC, CBCUK, CBCUN, and IWS are determined in accordance with the current

Accounting Procedures as described in Findings 9 and 10. However, when modeled

groundwater use is increased such that groundwater-caused depletions result in stream drying

and a break in the hydraulic connection between the groundwater system and the stream,

14 Id., p. 9.

15 Id.

16 Id., p. 12.

17 Id., p. 47.

18 Id., p. 51.

19 Ignoring minor nonlinearities from unrelated factors.

9

there is no remaining streamflow to deplete. Under such conditions, the modeled response of

the stream becomes nonlinear, and the condition of "additivity" no longer holds when

CBCUC, CBCUK, CBCUN, and IWS are determined in accordance with the current

Accounting Procedures.

21. As described in Finding 19, Nebraska contends that the current Accounting Procedures

assume that VWSG, defined by Nebraska as (θ - CKMN), can be computed using the

individually-computed impacts in a sub-basin. That is: CBCUC + CBCUK + CBCUN - IWS

would equal (θ - CKMN) under all conditions. However, careful readings of the Accounting

Procedures20 and the Final Report of the Special Master,21 which includes a detailed

description of the significant attributes of the RRCA Groundwater Model and use of the

Model output, do not reveal that the assumption of "additivity" to (θ - CKMN) under all

conditions was made by either the representatives of the States that developed the

Accounting Procedures or the representatives of the States that developed the RRCA

Groundwater Model.

22. One of the co-authors of Nebraska's expert report on estimating CBCU for groundwater and

IWS, Michael McDonald, was a member of the Technical Groundwater Modeling

Committee that developed the RRCA Groundwater Model.22 However, Nebraska did not

offer any testimony during the hearing on this issue that would corroborate the assertion that

the Technical Groundwater Modeling Committee intended that CBCUC + CBCUK + CBCUN

- IWS would equal (θ - CKMN) under all conditions. The fact that this "additivity" holds

when streamflow response to groundwater depletions is linear does not establish that the

representatives of the States that developed the RRCA Groundwater Model and the

Accounting Procedures assumed or intended that this condition of additivity would hold

when streamflow response to groundwater depletions is nonlinear.

23. The description of the significant attributes of the RRCA Groundwater Model and use of the

Model output contained in the Final Report of the Special Master specifically includes a

description of how the Model is used to calculate CBCUC, CBCUK, CBCUN, and IWS,23

which is the same as specified in the Accounting Procedures as described in Findings 9

and 10.

24. The fact that "[t]he ‘base' run is the simulation with all groundwater pumping, groundwater

pumping recharge, and surface water recharge within the model study boundary for the

20 Final Settlement Stipulation, Volume 1 of 5, Appendix C, as revised (July 2005) and adopted (August 10, 2006)

by the RRCA.

21 Final Report of the Special Master With Certificate of Adoption of RRCA Groundwater Model, Kansas v.

Nebraska and Colorado, No. 126, Original, September 17, 2003.

22 See Kansas Exhibit 72.

23 See Final Report of the Special Master With Certificate of Adoption of RRCA Groundwater Model, Kansas v.

Nebraska and Colorado, No. 126, Original, September 17, 2003, pp. 49-50.

10

period 1918 to the current accounting year ‘on',"24 and that this base run would likely

simulate stream drying at some locations during certain years, resulting in nonlinear

response, suggests that such an outcome was anticipated by the Technical Groundwater

Modeling Committee that developed the RRCA Groundwater Model. This is supported by

the testimony of both Kansas' expert witness on this issue, Mr. Steve Larson,25 and

Colorado's expert witness on this issue, Dr. Willem Schreüder,26 both of whom served on the

Technical Groundwater Modeling Committee that developed the RRCA Groundwater Model.

25. Using flows in Beaver Creek in 2003 as an example, Nebraska correctly points out that:

... increasing pumping by either Kansas or Nebraska alone or both states together causes

baseflow at the Beaver Creek accounting point to drop to zero after a threshold is reached.

Baseflow remains zero beyond this threshold as pumping is further increased. Clearly,

increasing pumping beyond this point by either state must have some impact on the

groundwater/stream system. Where in the system is this impact felt?27

24 Id.

25 MR. DRAPER: Was it clear to you that the model, the groundwater model, has nonlinear features

related to stream depletions?

MR. LARSON: Yes, it was. There were several nonlinear features in the model that were, in my

view, pretty obvious. And one of them -- that is, the changes in saturated thickness with changes in water

levels -- there were some idealizations made, primarily for computational stability reasons, to at least

linearize that feature; but there were other nonlinear features that were pretty obvious. Evapotranspiration,

function is a method of piecewise linear; but, overall, similiarly [sic] the rain is nonlinear, similarly the

stream-drying-sort-of feature, if you will, is a piecewise linear feature as well.

Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1233:23-1234:13.

26 DR. SCHREÜDER: The first point is that Nebraska is using 2003 as an example of how the

modeling is not behaving in an appropriate way.

That is not correct.

In the first place, 2003 is a fairly extreme year; but, nevertheless, none of

the behavior that we observe in 2003 -- wasn't known to the committee at the time that the model

was put together. ...

But we looked in great detail at the period prior to 2000 and this similar

kind of behavior did, in fact, occur and was well known to many members.

MR. AMPE: Doctor, when did you first become aware of the nonlinearity of the model?

DR. SCHREÜDER: About 15 minutes after I saw it the first time.

Transcript of Arbitration Proceedings, March 18, 2009, Volume VIII at 1388:13-1389:3.

27 Nebraska Exhibit 30, Expert Report of Dr. David P. Ahlfeld, Michael G. McDonald, and James C. Schneider,

Estimating Computed Beneficial Consumptive Use for Groundwater and Imported Water Supply under the

Republican River Compact, January 20, 2009, p. 22.

11

Increasing groundwater consumption by either Kansas or Nebraska after base flow drops to

zero will result in additional reductions in groundwater storage than would have occurred had

the base flow not been fully depleted, unless streamflow other than from base flow is

available for depletion by the increased groundwater consumption. Obviously, once the

consumptive use of groundwater from a groundwater system that is hydraulically connected

to a stream has fully depleted the flow in that stream, any additional consumption of

groundwater from that system cannot be supplied from depletions to streamflow, but has to

be supplied from other sources including much larger increases in withdrawals from

groundwater storage.

26. While Nebraska's experts clearly understand the response described in Finding 25,28 its

proposed changes to calculate CBCUC, CBCUK, CBCUN, and IWS are based on depletions to

streamflow that cannot occur once streamflow has been fully depleted. Using Beaver Creek

in 2003 as an example, differencing results from the RRCA Groundwater Model as described

in Finding 9 produces an estimate of the base flow in 2003 subject to depletion by

consumptive groundwater use in Kansas of 323 acre-feet, with full groundwater use in

Nebraska. Because of consumptive groundwater use in Nebraska during the period 1940

through 2003, the estimated 323 acre-feet is the most amount of base flow that consumptive

groundwater use in Kansas could deplete from Beaver Creek. Once flows in Beaver Creek

are depleted, the consumptive use of groundwater in Kansas that would cause additional

depletions to streamflow in Beaver Creek, if such flow existed, must be satisfied with

groundwater from other sources, primarily groundwater storage. Similarly, with full

groundwater use in Kansas the estimated base flow in 2003 subject to depletion by

consumptive groundwater use in Nebraska is 727 acre-feet. Because of consumptive

groundwater use in Kansas during the period 1940 through 2003, the estimated 727 acre-feet

is the most base flow that consumptive groundwater use in Nebraska could deplete from

Beaver Creek. As for Kansas, the consumptive use of groundwater in Nebraska that would

cause additional depletions to streamflow in Beaver Creek, if such flow existed, must be

satisfied with groundwater from other sources, primarily groundwater storage. The estimated

streamflow in 2003 that can be depleted by Kansas with full groundwater use in Nebraska

added to the estimated streamflow in 2003 that can be depleted by Nebraska with full

groundwater use in Kansas is 1,050 acre-ft.

Nebraska contends that the "true total impact" is 6,445 acre-feet, calculated as (θ - KN),29

and that "[t]he difference between the true total impact, 6,445 ac-ft, and the total impact

estimated by summing individual impacts is 5,395 acre-feet." Nebraska further contends that

"[t]his amount of streamflow depletion is occurring but not being accounted for in the current

procedure."30 Nebraska's contention is flawed because although the consumptive beneficial

28 Id., p. 22-24.

29 Historically, there have not been any effects on streamflow in Beaver Creek other than from consumptive use of

groundwater in Kansas (K) and in Nebraska (N).

30 Id., p. 19.

12

use of groundwater in Kansas and Nebraska during 2003 must have been significantly greater

than 1,050 acre-feet, the sum of CBCUK and CBCUN, there could not have been 6,445 acrefeet

of base flow from groundwater discharge that could have been depleted from Beaver

Creek in 2003. The additional consumptive beneficial use of groundwater by Kansas and

Nebraska beyond what would deplete streamflow to zero had to have consumed groundwater

from other sources, primarily groundwater storage. Historically, there have obviously been

significant groundwater consumptive uses in both Kansas and Nebraska that have reduced

groundwater storage, lowered groundwater levels, and largely depleted the base flow that

was available in 2003. The Beaver Creek base flow in 2003 estimated by Nebraska to have

been 6,445 acre-feet would be a viable estimate only if there had never been consumptive

groundwater use in Kansas or Nebraska, which obviously is not what has actually occurred.

27. Nebraska terms the difference between VSWG, calculated as (θ - CKMN), and the sum of

CBCUC, CBCUK, and CBCUN, less IWS, a residual.31 As described in Finding 17,

Nebraska's proposed changes to the procedures for calculating CBCUC, CBCUK, CBCUN,

and IWS, result in the sum of CBCUC, CBCUK, and CBCUN, less IWS, equaling

(θ - CKMN), and a residual of zero.

28. One result from the analysis in Finding 26 is that Nebraska's proposed procedure for

determining VWS, whereby

VWS = VWSS + VWSG

and

VWSG = (θ - CKMN), also referred to by Kansas as the "virgin water supply metric,"32

is more consistent with the definition of VWS established in the Compact and adopted in the

Accounting Procedures (see Finding 8) than is summing CBCUC, CBCUK, and CBCUN, less

IWS, each determined in accordance with the existing Accounting Procedures, to compute

what Nebraska terms VWSG.

29. While Nebraska's proposal for determining what it terms VWSG, or what Kansas terms the

virgin water supply metric, is more consistent with the definition of VWS established in the

Compact and adopted in the Accounting Procedures, than is the definition implied by

summing CBCUC, CBCUK, and CBCUN, less IWS, Nebraska's proposed changes to

calculate CBCUC, CBCUK, CBCUN, and IWS are problematic. Again using flows in Beaver

Creek in 2003 as an example, Nebraska's proposed methodology results in a value for

CBCUK of 3,021 acre-feet and a value for CBCUN of 3,425 acre-feet for a total VWSG of

31 Id. at 46.

32 Nebraska Exhibit 36, Kansas' Review of Nebraska's Request for Change in Accounting Procedure, September 18,

2007, p. 2.

13

6,445 acre-feet.33 These values are equivalent to adding one-half of the residual (one-half of

5,395 acre-feet) to CBCUK (323 acre-feet) and one-half of the residual to CBCUN (727 acrefeet),

when CBCUK and CBCUN are calculated using the methodology prescribed in the

existing Accounting Procedures as described in Finding 9.34 The residual of 5,395 acre-feet

is essentially the amount of groundwater consumptive use beyond the sum of 323 acre-feet

and 727 acre-feet from streamflow depletion that must come from other groundwater

sources, primarily groundwater storage, and is equally divided between Kansas and Nebraska

using Nebraska's proposed methodology.35

30. Equally dividing what are primarily additional withdrawals from groundwater storage

between Kansas and Nebraska, when streamflow is depleted and there is no longer a

hydraulic connection with the groundwater system, to determine CBCUK and CBCUN

without regard to the decrease in groundwater storage caused by groundwater use in each

state is not appropriate. Similarly, equally dividing what are primarily additional

withdrawals from groundwater storage between Colorado and Nebraska in the case of

Frenchman Creek, when streamflow is depleted and there is no longer a hydraulic connection

with the groundwater system, to determine CBCUC and CBCUN without regard to the

decrease in groundwater storage caused by groundwater use in each state is problematic

given that "the majority of the Frenchman Basin is in Nebraska and Nebraska pumping can

be expected to have the largest influence."36

31. Using the examples of Beaver Creek and Frenchman Creek, equally dividing what are

primarily additional withdrawals from groundwater storage between two states when

streamflow is depleted and there is no longer a hydraulic connection with the groundwater

system to determine CBCU, without regard to the decrease in groundwater storage caused by

groundwater use in each state, is also inconsistent with there being "very little propagation of

head change across statelines."37

32. When the groundwater being consumptively used involves all three states, or when there is

significant IWS, the residual described in Finding 27 is divided in "a more complicated

way"38 but the residual must still be related to changes in groundwater storage.

33 Nebraska Exhibit 30, Expert Report of Dr. David P. Ahlfeld, Michael G. McDonald, and James C. Schneider,

Estimating Computed Beneficial Consumptive Use for Groundwater and Imported Water Supply under the

Republican River Compact, January 20, 2009, p. 50.

34 Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1148:19-1149:4 (Ahlfeld).

35 Transcript of Arbitration Proceedings, March 19, 2009, Volume IX at 1466:9-1470:8 (Ahlfeld).

36 Nebraska Exhibit 30, Expert Report of Dr. David P. Ahlfeld, Michael G. McDonald, and James C. Schneider,

Estimating Computed Beneficial Consumptive Use for Groundwater and Imported Water Supply under the

Republican River Compact, January 20, 2009, p. 30.

37 Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1173:8-9 (Ahlfeld).

38 Id.at 1149:7 (Ahlfeld).

14

33. Groundwater consumptively used from groundwater storage is not streamflow depletion, and

inclusion of the consumptive use of groundwater storage in the calculation of CBCUC,

CBCUK, and CBCUN is inconsistent with the definition of CBCU as set forth in § II. of the

Accounting Procedures. Similarly, including the base flow in VWSG that would be

discharged from groundwater as though groundwater storage had not been reduced by

consumptive groundwater use, or θ, results in overstating the Computed Water Supply (the

"CWS") that is available to be allocated to each state in any drainage basin during a year

where simulated stream drying in that basin occurs and there is no hydraulic connection

between the groundwater system and the stream.

34. Nebraska's proposed procedure for determining IWS has a related problem. Half of the

model runs and differences, and half of the weighting, proposed for determining IWS do not

include any simulated groundwater use by Nebraska. This means that for half of the model

runs, groundwater storage is undepleted by Nebraska groundwater use and simulated

groundwater levels are higher than historical levels. As a result, IWS determined as

proposed by Nebraska will generally be greater than IWS determined using the existing

procedure specified in § III.A.3. of the Accounting Procedures as described in Finding 10.39

In fact, the Main Stem IWS and the total IWS determined using Nebraska's proposed method

is greater than the corresponding IWS determined using the existing procedure described in

Finding 10 for all years from 1981 through 2006, except for 1993.40 The reason for the

anomaly in the 1993 IWS is unknown, but may be the result of computational error.

35. Colorado's expert on this issue, Dr. Willem A. Schreüder, identified another concern with

Nebraska's proposed changes. In his report, Dr. Schreüder states that: "The method

proposed by Nebraska, on the other hand, does included the consumption of imported

water."41 Dr. Schreüder shows that CBCUN calculated "... for the Swanson-Harlan reach are

greater with imported water than without imported water"42 and further states that: "As

shown in Figure 10, any simulation where surface water imports are on will include

consumption of imported water."43 Thus, the current Accounting Procedures for calculating

CBCUC, CBCUK, CBCUN, as described in Finding 9, may also include consumption of

imported water, since both the "base" run and the "no State pumping" run include surface

39 See testimony of Mr. Steve Larson, Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at

1240:25-1241:5.

40 See Tables 1a through 1z in Colorado Exhibit 7, Expert Report of Willem A. Schreüder, Ph.D., Report in

Response to: Estimating Computed Beneficial Consumptive Use for Groundwater and Imported Water Supply

under the Republican River Compact, Ahfed [sic] et al. (January 20, 2009), February 16, 2009.

41 Id. at 18.

42 Id.

43 Id. at 19.

15

water imports.44 Including the consumption of imported water in the calculation of CBCU is

not consistent with § IV.F. of the FSS, which specifically provides that: "Beneficial

Consumptive Use of Imported Water Supply shall not count as Computed Beneficial

Consumptive Use or Virgin Water Supply Credit."45

36. Although Nebraska's proposed changes to calculate CBCUC, CBCUK, CBCUN, and IWS are

problematic, the RRCA should consider reconvening the Technical Groundwater Modeling

Committee to thoroughly re-evaluate the nonlinear response of the RRCA Groundwater

Model when simulated stream drying occurs, re-evaluate the existing procedures for

determining CBCU and IWS described in Findings 9 and 10, and document its conclusions

and any recommendations in a report to the RRCA.

Accounting Procedures - Haigler Canal

37. Nebraska has proposed three changes in the Accounting Procedures adopted by the RRCA

involving the Haigler Canal that would modify (1) the annual determination of water diverted

from the North Fork Republican River in Colorado into the Haigler Canal46 for irrigation in

Nebraska; (2) the annual apportionment of return flows from irrigation in Nebraska between

the Main Stem, measured at the USGS stream gage near Hardy, Nebraska, station 06853500

(the "Hardy Gage"), and the Arikaree River, measured at the USGS stream gage at Haigler,

Nebraska, station 06821500 (the "Arikaree Gage); and (3) the annual calculation of VWS for

the North Fork of Republican River in Colorado and the Arikaree River.

38. Under the current Accounting Procedures, the Nebraska CBCU attributable to the annual

diversions from the North Fork Republican River to the Haigler Canal for irrigation in

Nebraska is based on using the total amounts of water diverted as measured at the Haigler

Canal Stateline Gage, station 00061400.47 The first change to the Accounting Procedures

involving the Haigler Canal proposed by Nebraska would reduce the amount of these annual

diversions from the North Fork Republican River by an amount equal to the annual

discharges from the Haigler Canal to the Arikaree River, as measured by Nebraska at the

Haigler Canal Spillback gage, station 00061500, which is located approximately one-half

mile west of the point of discharge to the Arikaree River,48 less some adjustments for

44 Colorado's expert, Willem A. Schreüder, proposed alternative methodology using differences between 5 runs of

the RRCA Groundwater Model to calculate CBCUC, CBCUK, CBCUN, and IWS, which do not include imported

water in the calculation of CBCUC, CBCUK, and CBCUN, Id., p. 7. However, there is no evidence that this

alternative methodology has been presented to the RRCA as required by the FSS.

45 Final Settlement Stipulation, Volume 1 of 5, p. 25.

46 The Pioneer Canal in Article V, Republican River Compact.

47 Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (on title page, revised August 10, 2006), § IV.B.3. [sic], p. 26.

48 Transcript of Arbitration Proceedings March 17, 2009, Volume VII at 1226:23-1227:1 (Williams).

16

precipitation inflow to the canal.49 Nebraska has maintained the Haigler Canal Spillback

gage and recorded the flow in the canal at this location for approximately the last 20 years.50

39. Nebraska's proposed change to subtract the amount of water measured annually at the

Haigler Canal Spillback gage from the amount of water measured annually at the Haigler

Canal Stateline Gage to determine the amount of water diverted from the North Fork of the

Republican River for irrigation in Nebraska assumes that much if not all of the water

measured at the Haigler Canal Spillback gage is discharged from the Haigler Canal to the

Arikaree River and is surface water in the Arikaree River that can be measured at the

Arikaree Gage.51

40. Nebraska's expert witness on this issue, Mr. James Williams, testified that "... we have seen

much of the [Haigler Canal Spillback] water, if not all, in past six or seven years showing up

at the Arikaree gage ... ."52 Beginning in about 2001, streamflows measured at the Arikaree

Gage decreased significantly. During the years 2002, 2003, 2004, and 2005, the annual

amounts of water measured at the Haigler Canal Spillback gage exceeded the actual annual

amounts of water measured at the Arikaree Gage by 58 acre-feet (20 percent of spillback),

610 acre-feet (37 percent of spillback), 314 acre-feet (48 percent of spillback), and 187 acrefeet

(14 percent of spillback), respectively.53 Thus contrary to Mr. Williams' testimony,

significant portions of the Haigler Canal Spillback water did not reach the Arikaree Gage

during the years 2002 through 2005.

41. When asked whether analyses of losses and gains had been made between the Haigler Canal

Spillback gage and the point of discharge to the Arikaree River and between the point of

discharge and the Arikaree Gage, Mr. Williams testified: "No, we did not."54

42. In its post-hearing brief, Nebraska asserts:

There is no dispute that the Arikaree is now frequently dry and that spillback/return water

may not get to the Arikaree gage - but that doesn't change the fact that North Fork water

49 Id. at 1206:23-1207:11 (Williams).

50 Id. at 1193:3-5 (Williams).

51 Id. at 1193:8-14; 1222:23-1223:3.

52 Id.

53 Nebraska Exhibit 31, Expert Report of James C. Schneider and James R. Williams, Expert Report on Accounting

Issues: Haigler Canal and Groundwater Model Accounting Points, January 20, 2009, Table 1 (p. 4) and Table 2

(p. 7); Kansas Exhibit 29, Expert Report of David Barfield and Scott Ross, Kansas's Responsive Expert Report

Concerning Haigler Canal and Groundwater Modeling Accounting Points, February 17, 2009, Table 1 (Arikaree

gage value).

54 Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1208:4-13.

17

is nevertheless discharged into the Arikaree River and thereby directly or indirectly

inflates the VWS.55

The calculation for the Arikaree River VWS specified in the Accounting Procedures is:

VWS = Arikaree Gage at Haigler Stn. No. 06821500 + CBCUc + CBCUk + CBCUn - IWS.56

For VWS for the Arikaree River to increase, flows at the Arikaree Gage must increase and/or

CBCU must increase. As described in Finding 40, during four of the six years from 2001

through 2006, significant portions of the flows from the Haigler Canal Spillback did not

reach the Arikaree River Gage and could not have increased VWS. Also, there is no

evidence that CBCU has increased as a result of the Haigler Canal Spillback. Therefore,

Nebraska's assertion is flawed.

43. In its post-hearing brief, Nebraska also asserts:

The diminished streamflows [at the Arikaree Gage] could be the result of many different

human activities but it is clear that any discharge [from the Haigler Canal Spillback] into

the stream, is a direct credit to that stream whether it is lost to seepage or not.57

This assertion would hold if the amount of the Haigler Canal Spillback lost to seepage

resulted in an equivalent amount of groundwater discharge to the Arikaree River. However

as described in Findings 55 and 56, the prevalent direction of groundwater flow, at least on

the north side of the Arikaree River, is to the north towards the Main Stem, not towards the

Arikaree River, which is consistent with Finding 40 that during recent years significant

portions of the Haigler Canal Spillback water did not reach the Arikaree Gage.

44. Based on the available information, a significant portion of the water measured at the Haigler

Canal Spillback gage, at least during the years since about 2001, does not remain in the

Arikaree River as measurable surface water at the Arikaree Gage. While some of the water

measured at the Haigler Canal Spillback gage undoubtedly reaches the Arikaree Gage under

certain conditions, there is insufficient information to justify changing the Accounting

Procedures to reduce the diversions from the North Fork Republican River into the Haigler

Canal by the amount of water measured at the Haigler Canal Spillback gage.

45. As a result, the changes proposed by Nebraska to the Accounting Procedures involving VWS

calculations for the North Fork of Republican River in Colorado and the Arikaree River are

not justified.

55 State of Nebraska's Post-Hearing Brief at 54.

56 Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (on title page, revised August 10, 2006), § IV.B.4. [sic], p. 26.

57 State of Nebraska's Post-Hearing Brief at 54.

18

46. Under the current Accounting Procedures, the Nebraska CBCU attributable to the annual

diversions from the North Fork Republican River to the Haigler Canal for irrigation in

Nebraska is calculated as 60 percent of the total amounts of water diverted as measured at the

Haigler Canal Stateline Gage.58 The remaining 40 percent of the total amounts of water

diverted is return flow,59 which is accounted for as returning to the Main Stem in the

calculation of VWS.60 The second change to the Accounting Procedures involving the

Haigler Canal proposed by Nebraska would apportion the return flows from irrigation in

Nebraska between the Main Stem, calculated at the Hardy Gage, and the Arikaree River,

calculated at the Arikaree Gage, in proportion to the acreage irrigated using water from the

Haigler Canal in the Main Stem drainage (51 percent) and the Arikaree River drainage (49

percent).61

47. Nebraska proposes the change described in Finding 46 to implement the directive in

§ IV.B.3. [sic]62 of the Accounting Procedures which states:

The RRCA will investigate whether return flows from the Haigler Canal diversion in

Colorado may return to the Arikaree River, not the North Fork of the Republican River,

as indicated in the formulas. If there are return flows from the Haigler Canal to the

Arikaree River, these formulas will be changed to recognize those returns.

48. The term "return flow" is not defined in the Accounting Procedures but as commonly used,

return flow is that part of a diverted flow that is not consumptively used and is returned to its

original source or another source of water.63 In the context of the Accounting Procedures,

return flow is that part of a diverted flow returned to the Main Stem and its tributaries as

surface water by overland flow or through groundwater discharge.

49. Nebraska's proposal to apportion return flows returned to the Main Stem and the Arikaree

River from irrigation in Nebraska , in proportion to the acreage irrigated using water from the

Haigler Canal in the Main Stem drainage (51 percent) and the Arikaree River drainage (49

percent) is appropriate for that portion of the return flows comprised by overland flow, since

overland flow would remain within the drainage where the associated irrigation occurred.

50. Nebraska's proposal to apportion return flows returned to the Main Stem and the Arikaree

River in proportion to the acreage irrigated using water from the Haigler Canal in the Main

58 Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (revised date on title page: August 10, 2006), § IV.B.3. [sic], p. 26.

59 Id. at § IV.A.2.a)., p. 20.

60 Id. at § IV.B.3. [sic], p. 26; § IV.B.15 [sic], p 36.

61 Nebraska Exhibit 31, Expert Report of James C. Schneider and James R. Williams, Expert Report on Accounting

Issues: Haigler Canal and Groundwater Model Accounting Points, January 20, 2009, p. 5-6.

62 § IV.B.1. in Final Settlement Stipulation, Volume 1 of 5, Appendix C.

63 See USGS Water Science Glossary of Terms, http://ga.water.usgs.gov/edu/dictionary.html#main.

19

Stem drainage and the Arikaree River drainage is not necessarily appropriate for that portion

of the return flows comprised by groundwater discharge, since groundwater flow is not

constrained to the drainage where the associated irrigation occurs because groundwater level

gradients do not necessarily conform to the overlying topographical gradients.

51. Nebraska's expert witness on this issue, Mr. James Williams, did not provide any testimony

or other evidence regarding the portion of return flows from irrigation in Nebraska returning

to the Main Stem or the Arikaree River as overland flow.

52. Mr. Williams did testify that the soils in the Arikaree drainage near Haigler "tend to be

somewhat sandy."64 Colorado's expert on this issue, Mr. James Slattery, testified that the

soils in the Arikaree drainage near Haigler are "extremely sandy" and that because "the

majority of this land has been converted over to center pivot sprinklers ... there is just very

little surface water runoff ... ."65 This suggests that there may be minimal return flow to the

Arikaree River comprised by overland flow.

53. During the period of years from 1995 through 2006, the annual amounts of water returning to

the Arikaree River from irrigation using water from the Haigler Canal, as estimated in

accordance with only this change to the Accounting Procedures as proposed by Nebraska,66

exceeded the actual annual amounts of water measured at the Arikaree Gage by 515 acre-feet

(48 percent of the proposed return flow), 767 acre-feet (77 percent of the proposed return

flow), 70 acre-feet (6 percent of the proposed return flow), and 385 acre-feet (53 percent of

the proposed return flow) for the years 2001, 2002, 2003, and 2004, respectively.53 Thus,

significant portions of the annual amounts of return flow estimated in accordance with

Nebraska's proposed change to the Accounting Procedures did not reach the Arikaree Gage

during the years 2001 through 2004.

54. When asked whether he knew the direction of groundwater flow in the Haigler area,

Mr. Williams testified: "No, I do not."67

55. Simulations using the RRCA Groundwater Model indicate that the prevalent direction of

groundwater flow under lands irrigated using water from the Haigler Canal in the Haigler

area (on the north side of the Arikaree River) is to the north towards the Main Stem, not the

Arikaree River.68

64 Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1210:20-1211:8.

65 Transcript of Arbitration Proceedings, March 18, 2009, Volume VIII at 1360:9-18.

66 Without reducing the amounts of water measured at the Haiglar Canal Stateline Gage by the amounts of water

from the Haiglar Canal Spillback.

67 Id. at 1210:1-3.

68 Id. at 1365:24-1366:7; Colorado Exhibit 11, Expert Report of James E. Slattery, State of Colorado's Response to

Nebraska's Expert Report on Accounting Issues: Haigler Canal and Groundwater Model Accounting Points,

February 16, 2009, p. 5.

20

56. In its post-hearing brief, Nebraska contends:

Such a determination [that the prevalent direction of groundwater flow is to the north

towards the Main Stem] seems doubtful given that the Groundwater Model uses one-mile

cells and the distance between the Haigler Canal and the Republican River is less than

one mile. If the Haigler Canal and Republican River are in the same model cell, or even

in adjacent cells, no gradient would likely be determined.69

However, it is not the location of Haigler Canal that is pertinent to the direction of

groundwater flow for that portion of return flows that return from groundwater discharge.

Rather, it is the location of the lands irrigated that is pertinent, and the lands irrigated with

water from the Haigler Canal are located from one to three miles south of the Republican

River. Thus, results from simulations using the RRCA Groundwater Model can be used to

estimate the prevalent direction of groundwater return flow under lands irrigated with water

from the Haigler Canal.

57. Based on the available information, most of the return flow comprised by groundwater

discharge from irrigation in Nebraska using water from the Haigler Canal returns to the Main

Stem, not the Arikaree River, at least during the years since 2001. While some of the water

measured at the Arikaree Gage may be comprised of return flow from groundwater discharge

under certain conditions, there is insufficient information to justify changing the Accounting

Procedures to apportion any of the return flow to the Arikaree River.

Accounting Procedures - Groundwater Model Accounting Points

58. Article II of the Republican River Compact defines the Republican River Basin as follows:

The Basin is all the area in Colorado, Kansas, and Nebraska, which is naturally drained

by the Republican River, and its tributaries, to its junction with the Smoky Hill River in

Kansas. The main stem of the Republican River extends from the junction near Haigler,

Nebraska, of its North Fork and the Arikaree River, to its junction with Smoky Hill River

near Junction City Kansas.70

59. The "equitable division" or "allocation" of the waters of the Republican River Basin between

the States is set forth in Article IV of the Compact, subject to the proportionate adjustment

required in Article III. Article IV of the Compact specifies the amounts of water allocated to

each state from each source of water in the Republican River Basin and identifies each

source of water from which an allocation is made as a named "drainage basin."

69 State of Nebraska's Post-Hearing Brief at 55.

70 Republican River Compact, Pub. Law No. 78-60, 57 Stat. 86 (1943); codified at § 82a-518, K.S.A. (2007);

App. § 1-106, 2A N.R.S. (1995); and § 37-67-101 C.R.S. (2008).

21

60. The term "drainage basin" is not defined in the Compact but as commonly used, a drainage

basin is a land area where precipitation runs off into streams, rivers, lakes, and reservoirs.63

A drainage basin ends where there is no longer an area from which precipitation runs off,

which corresponds to the lowest point in elevation above which a delineated area is drained.

The end of a drainage basin is also located at the point where the collected precipitation

runoff discharges into another surface water feature, which is termed the "confluence" when

one stream or river joins another stream or river.

61. The "equitable division" or "allocation" of the waters of the Republican River Basin set forth

in Article IV of the Compact for a named "drainage basin" is derived from the "computed

average annual virgin water supply"71 originating in that drainage basin, which ends at the

confluence of the stream draining that basin and the main stem of the Republican River,72 as

set forth in Article III of the Compact.

62. In § II. of the Accounting Procedures, the term "Designated Drainage Basins" is defined as

"the drainage basins of the specific tributaries and the Main Stem of the Republican River as

described in Article III of the Compact." The term "Sub-basin" is defined as:

[T]he Designated Drainage Basins, except for the Main Stem, identified in Article III of

the Compact. For purposes of Compact accounting the following Sub-basins will be

defined as described below:

North Fork of the Republican River in Colorado drainage basin is that drainage

area above USGS gaging station number 06823000, North Fork Republican

River at the Colorado-Nebraska State Line,

Arikaree River drainage basin is that drainage area above USGS gaging station

number 06821500, Arikaree River at Haigler, Nebraska,

Buffalo Creek drainage basin is that drainage area above USGS gaging station

number 06823500, Buffalo Creek near Haigler, Nebraska,

Rock Creek drainage basin is that drainage area above USGS gaging station

number 06824000, Rock Creek at Parks, Nebraska,

South Fork of the Republican River drainage basin is that drainage area above

USGS gaging station number 06827500, South Fork Republican River near

Benkelman, Nebraska,

71 Pursuant to the Accounting Procedures, the "computed average annual virgin water supply" is termed the

Computed Water Supply (the "CWS"), which equals the VWS reduced by changes in Federal reservoir storage

and flood flows. The CWS is used to calculate the allocations between the States (See Republican River Compact

Administration Accounting Procedures and Reporting Requirements, revised July 2005 [revised date on title page:

August 10, 2006], p. 10).

72 Or the North Fork of the Republican River in Nebraska for the drainage basins specified in the Compact as the

"North Fork of the Republican River drainage basin in Colorado" and the "Arikaree River drainage basin."

22

Frenchman Creek (River) drainage basin in Nebraska is that drainage area above

USGS gaging station number 06835500, Frenchman Creek in Culbertson,

Nebraska,

Driftwood Creek drainage basin is that drainage area above USGS gaging station

number 06836500, Driftwood Creek near McCook, Nebraska,

Red Willow Creek drainage basin is that drainage area above USGS gaging

station number 06838000, Red Willow Creek near Red Willow, Nebraska,

Medicine Creek drainage basin is that drainage area above the Medicine Creek

below Harry Strunk Lake, State of Nebraska gaging station number 06842500;

and the drainage area between the gage and the confluence with the Main Stem,

Sappa Creek drainage basin is that drainage area above USGS gaging station

number 06847500, Sappa Creek near Stamford, Nebraska and the drainage area

between the gage and the confluence with the Main Stem; and excluding the

Beaver Creek drainage basin area downstream from the State of Nebraska gaging

station number 06847000 Beaver Creek near Beaver City, Nebraska to the

confluence with Sappa Creek,

Beaver Creek drainage basin is that drainage area above State of Nebraska

gaging station number 06847000, Beaver Creek near Beaver City, Nebraska, and

the drainage area between the gage and the confluence with Sappa Creek,

Prairie Dog Creek drainage basin is that drainage area above USGS gaging

station number 06848500, Prairie Dog Creek near Woodruff, Kansas, and the

drainage area between the gage and the confluence with the Main Stem;

63. In § II. of the Accounting Procedures, the term "Main Stem" is defined as:

[T]he Designated Drainage Basin identified in Article III of the Compact as the North

Fork of the Republican River in Nebraska and the main stem of the Republican River

between the junction of the North Fork and the Arikaree River and the lowest crossing of

the river at the Nebraska-Kansas state line and the small tributaries thereof, and also

including the drainage basin Blackwood Creek;

This definition for "Main Stem" differs from the description of the main stem in Article II of

the Compact, as set forth in Finding 58, in that it includes the North Fork of the Republican

River in Nebraska and ends at "the lowest crossing of the river at the Nebraska-Kansas state

line" rather than at "its junction with the Smoky Hill River in Kansas." However, this

definition for "Main Stem" is wholly consistent with the designated drainage basin defined in

the next to the last full paragraph in Article III of the Compact.

64. The Accounting Procedures, § III.D.1., specify that CBCU of groundwater

... for each Sub-basin will include all depletions and accretions upstream of the

confluence with the Main Stem. The values for the Main Stem will include all depletions

and accretions in stream reaches not otherwise accounted for in a Sub-basin.

23

This is consistent with the allocations made by named drainage basin in Article IV of the

Compact as described in Finding 61.

65. In § III.D.2. of the Accounting Procedures, the procedure for determining CBCU of surface

water is specified as follows:

For Sub-basins where the gage designated in Section II. is near the confluence with the

Main Stem, each State's Sub-basin Computed Beneficial Consumptive Use of surface

water shall be the State's Computed Beneficial Consumptive Use of surface water above

the Sub-basin gage. For Medicine Creek, Sappa Creek, Beaver Creek and Prairie Dog

Creek, where the gage is not near the confluence with the Main Stem, each State's

Computed Beneficial Consumptive Use of surface water shall be the sum of the State's

Computed Beneficial Consumptive Use of surface water above the gage, and its

Computed Beneficial Consumptive Use of surface water between the gage and the

confluence with the Main Stem.

This is consistent with the allocations made by named drainage basin in Article IV of the

Compact as described in Finding 61, assuming there is no significant CBCU of surface water

downstream from the Sub-basin gages, other than for Medicine Creek, Sappa Creek, Beaver

Creek, and Prairie Dog Creek, where CBCU of surface water downstream from each Subbasin

gage is added to the CBCU of surface water above each Sub-basin gage. However,

since the CBCU of surface water below the gage in each of these four sub-basins is already

included in the amount of water measured at the gage for each Sub-basin, the CBCU of

surface water below the gage for each Sub-basin is subtracted from the VWS for that Subbasin

and added to the VWS for the Main Stem,73 to avoid a double-accounting of water in

that Sub-basin.

66. Nebraska has identified four sub-basins where the stream gaging station designated in § II. of

the Accounting Procedures is located several miles upstream of the confluence with the Main

Stem, where the cell in the RRCA Groundwater Model is used to simulate base flow for

determining CBCU of groundwater (the "accounting point"): Frenchman Creek (River)

drainage basin in Nebraska, North Fork of the Republican River in Colorado drainage basin,

South Fork of the Republican River drainage basin, and Driftwood Creek drainage basin.

Nebraska contends that: "A discrepancy is introduced because VWS is calculated by adding

streamflow at one location to estimated groundwater impacts at a separate location."74

Nebraska further contends that this results in "... the potential for some of the surface water

passing that gage to then be consumed by the groundwater [pumping] and, in effect, a

double-accounting."75

73 Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (revised date on title page: August 10, 2006), § IV.B.11.-14. [sic], pp. 30-33.

74 Nebraska Exhibit 31, Expert Report of James C. Schneider and James R. Williams, Expert Report on Accounting

Issues: Haigler Canal and Groundwater Model Accounting Points, January 20, 2009, p. 9.

75 Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1220:7-9 (Williams).

24

67. Because stream gages must be sited where the hydraulic characteristics of a stream channel

are suitable for accurate measurements of streamflow in that channel, stream gages in the

named drainage basins for the Republican River are generally not located at their confluences

with the Main Stem.76

68. Nebraska notes that § II. of the Accounting Procedures defines the "Frenchman Creek

(River) drainage basin in Nebraska," "North Fork of the Republican River in Colorado

drainage basin," "South Fork of the Republican River drainage basin," and "Driftwood Creek

drainage basin," in each instance as being that drainage area above the corresponding gage

designated for each Sub-basin. Nebraska asserts that the "accounting points must be moved

to match the locations of the gages, and thus the Sub-basin definitions from Appendix C."77

69. As described in Findings 60 and 61, the allocations of water made to the States, as specified

by the Compact, are made for individual drainage basins, and each drainage basin implicitly

ends at the confluence between the stream associated with a particular drainage basin and the

Main Stem. The Accounting Procedures provided for by the FSS cannot change the

definitions of individual drainage basins implicit in the Compact.78 For the stated purposes

of Compact accounting, the sub-basins as defined in § II. of the Accounting Procedures are

appropriate provided adjustments are made such that the VWS is correctly estimated for the

drainage basin above the confluence between the stream associated with a particular drainage

basin and the Main Stem.

70. For the "Frenchman Creek (River) drainage basin in Nebraska," "South Fork of the

Republican River drainage basin," and "Driftwood Creek drainage basin," moving the

accounting points for determining the CBCU of groundwater to correspond to the locations

of the gages designated in § II. of the Accounting Procedures would result in the CBCU of

groundwater between a designated gage and the confluence of that Sub-basin's stream with

the Main Stem being included in the CBCU for the Main Stem rather than in the CBCU for

the tributary drainage basins. These changes would be inconsistent with the definitions of

these drainage basins implicit in Article III of the Compact and are not appropriate.

76 Colorado Exhibit 11, Expert Report of James E. Slattery, State of Colorado's Response to Nebraska's Expert

Report on Accounting Issues: Haigler Canal and Groundwater Model Accounting Points, February 16, 2009,

p. 7.

77 Nebraska Exhibit 31, Expert Report of James C. Schneider and James R. Williams, Expert Report on Accounting

Issues: Haigler Canal and Groundwater Model Accounting Points, January 20, 2009, p. 9.

78 See § I.D. of the FSS, which provides that:

The States agree that this Stipulation and the Proposed Consent Judgment are not intended to, nor

could they, change the States' respective rights and obligations under the Compact. The States

reserve their respective rights under the Compact to raise any issue of Compact interpretation and

enforcement in the future.

25

71. However, to the extent groundwater pumping causes depletions to streamflows downstream

of the gages designated in § II. of the Accounting Procedures for the "Frenchman Creek

(River) drainage basin in Nebraska," "South Fork of the Republican River drainage basin,"

and "Driftwood Creek drainage basin," and upstream of the confluence of each associated

stream with the Main Stem, the current Accounting Procedures for estimating VWS result in

a double-accounting of these depletions. The measured streamflow at each of these Subbasin

gages already includes the amount of the streamflow depletion between the gage for

each Sub-basin and the confluence of the stream for each Sub-basin with the Main Stem.

Adding the CBCU of groundwater between the gage for a particular Sub-basin and the

confluence of that Sub-basin's stream with the Main Stem to the measured streamflow at that

gage counts the same water twice in calculating VWS,79 and is not appropriate.

72. While it is not appropriate to move the accounting points as described in Finding 70, the

RRCA should modify the Accounting Procedures for the "Frenchman Creek (River) drainage

basin in Nebraska," "South Fork of the Republican River drainage basin," and "Driftwood

Creek drainage basin," to subtract the CBCU of groundwater below the designated gage for

each Sub-basin and above the confluence of that Sub-basin's stream with the Main Stem

from the VWS for that Sub-basin, to avoid double-accounting, and add that increment of

groundwater CBCU in the VWS for the Main Stem, such as is currently done in accounting

for the CBCU of surface water below the Sub-basin gages for Medicine Creek, Sappa Creek,

Beaver Creek, and Prairie Dog Creek.

73. At the hearing and in its post-trial brief, Colorado asserts that the Special Master appointed

by the Court in Kansas v. Nebraska and Colorado, No. 126, Original, made a specific finding

that the Republican River is formed at the junction of the Arikaree River and the North Fork

of the Republican River, near Haigler, Nebraska,80 which Colorado uses as the basis for its

contention that the current accounting point for the North Fork of the Republican River is at

the correct location. The statement made by the Special Master quoted by Colorado occurs

in the First Report of the Special Master (Subject: Nebraska's Motion to Dismiss) at the

beginning of § II. titled "BACKGROUND" (on page 6) and is simply a restatement of the

description of the Republican River Basin from Article II of the Compact, as partially set

forth in Finding 58. The Special Master's statement can not be a "finding" that the Main

Stem of the Republican River begins at the junction of the Arikaree River and the North Fork

of the Republican River for Compact accounting purposes pursuant to the FSS when Article

III of the Compact explicitly defines two separate drainage basins, from which allocations of

water are made in Article IV that include the North Fork: "North of the Republican River

drainage basin in Colorado" and "The North Fork of the Republican River in Nebraska and

the main stem of the Republican River between the junction of the North Fork and Arikaree

River and the lowest crossing of the river at the Nebraska-Kansas state line and the small

tributaries thereof ... ." The latter drainage basin is the Main Stem in § II. of the Accounting

79 Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (revised date on title page: August 10, 2006), § IV.B.7.-9. [sic], pp. 28-29.

80 Transcript of Arbitration Proceedings, March 17, 2009, Volume VII at 1205:2-22 (Williams); State of Colorado's

Post-Trial Brief at 54.

26

Procedures, which were incorporated in the FSS and as part of the FSS were found by the

Special Master to be "... in all respects compatible with the controlling provisions and

purposes of the Compact."81

74. The accounting point currently used to determine the CBCU of groundwater in the "North

Fork of the Republican River in Colorado drainage basin" is not located at the confluence

with the Main Stem, as the Main Stem is defined in Section II. of the Accounting Procedures

and set forth in Finding 63. This is inconsistent with the explicit meaning of the "North Fork

of the Republican River drainage basin in Colorado" in Article III of the Compact and results

in CBCU of groundwater in Kansas and Nebraska that should be included in the CBCU for

the Main Stem being included instead in the CBCU for the "North Fork of the Republican

River in Colorado drainage basin."

75. The accounting point used to determine the CBCU of groundwater in the "North Fork of the

Republican River in Colorado drainage basin" should be moved to the cell of the RRCA

Groundwater Model in which the North Fork of the Republican River crosses the Colorado-

Nebraska state line. This will result in reduced VWS for the "North Fork of the Republican

River in Colorado drainage basin" to the extent of "GWk" and "GWn" between the

Colorado-Nebraska state line and the confluence between the North Fork of the Republican

River in Nebraska and the Arikaree River.82 This will also result in increased VWS for the

Main Stem by the same amounts.

76. The changes to the Accounting Procedures described in Findings 72 and 75 should apply to

all years for which the accounting of water use has not been finalized and approved by the

RRCA. This is consistent with the positions of both Colorado and Nebraska83 (Kansas did

not address this issue). This is also consistent with the decision of the Special Master.84

Damages - Losses to Kansas Water Users from Overuse in Nebraska

77. Subsection V.B.2.a. of the FSS explicitly requires that:

a. During Water-Short Year Administration, Nebraska will limit its Computed

Beneficial Consumptive Use above Guide Rock to not more than Nebraska's

Allocation that is derived from sources above Guide Rock, and Nebraska's share of

81 Second Report of the Special Master (Subject: Final Settlement Stipulation), Kansas v. Nebraska and Colorado,

No. 126, Original, April 15, 2003, p. 3.

82 See Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (revised date on title page: August 10, 2006), § IV.B.3. [sic], p. 26.

83 State of Colorado's Post-Trial Brief at 56; State of Nebraska's Post-Hearing Brief at 57.

84 Second Report of the Special Master (Subject: Final Settlement Stipulation), Kansas v. Nebraska and Colorado,

No. 126, Original, April 15, 2003, p. 32.

27

any unused portion of Colorado's Allocation (no entitlement to Colorado's unused

Allocation is implied or expressly granted by this provision).85

Subsection V.B.2.e. of the FSS provides that:

e. For purposes of determining Nebraska's compliance with Subsection V.B.2.:

i. Virgin Water Supply, Computed Water Supply, Allocations and Computed

Beneficial Consumptive Use will be calculated on a two-year running average, as

computed above Guide Rock, with any Water-Short Year Administration year

treated as the second year of the two-year running average and using the prior

year as the first year;86

Subsection V.B.2.e. of the FSS does not explicitly address the amount of the violation when

Nebraska is not in compliance with § V.B.2. based on calculated two-year running averages

for Virgin Water Supply, Computed Water Supply, Allocations, and Computed Beneficial

Consumptive Use.

78. The States agreed "to implement the obligations and agreements in this Stipulation in

accordance with the schedule attached hereto as Appendix B."87 Appendix B of the FSS

unambiguously sets the "First year Water-Short Year Administration compliance" as 2006.88

79. Nebraska does not deny that it exceeded its basin-wide allocations in 2005 and 200689 and its

Water-Short Year allocations above Guide Rock in 2005 and 2006,90 based on the

Accounting Procedures currently approved by the RRCA, although Nebraska disagrees with

the amount of the violations estimated by Kansas for 2006.

80. Based on the accounting approved by the RRCA for 2005, Nebraska exceeded its 2005

Water-Short Year Administration allocation above Guide Rock by 42,860 acre-feet, when the

evaporation from Non-Federal Reservoirs below Harlan County Lake is included.91 Kansas'

estimate of the amount of Nebraska's exceedance of its 2006 Water-Short Year

Administration allocation above Guide Rock is 36,100 acre-feet, using data approved by the

85 Final Settlement Stipulation, Volume 1 of 5, p. 28.

86 Id., p. 30.

87 Id., p. 1.

88 Id., p. B1.

89 State of Nebraska's Post-Hearing Brief at 4.

90 Nebraska Exhibit 8, Expert Report of Marc Groff, Tom Riley, and David Kracman, Review of the 20 January

2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17, 2009, Table 2-2, p. 5.

91 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Attachment 1.

28

RRCA.92 The total of Nebraska's exceedance in 2005 and in 2006, as estimated by Kansas,

is 78,960 acre-feet.

81. The basin-wide exceedance by Nebraska in 2005, based on the accounting approved by the

RRCA for 2005, is 42,330 acre-feet.93 The two-year running average of Nebraska's

exceedance of its Water-Short Year Administration allocation above Guide Rock for 2006,

using the exceedance estimated by Kansas for 2006, is 39,480 acre-feet.94 The total of

Nebraska's basin-wide exceedance in 2005 and the two-year running average of Nebraska's

exceedance of its Water-Short Year Administration allocation above Guide Rock for 2006,

using the exceedance estimated by Kansas for 2006, is 81,810 acre-feet. This total amount is

greater than the sum of Nebraska's basin-wide exceedance in 2005 and Nebraska's

exceedance of its Water-Short Year Administration allocation above Guide Rock in 2006

only, as estimated by Kansas, by 3,380 acre-feet.95 The total amount of 81,810 acre-feet is

also greater than the sum of Nebraska's exceedance of its Water-Short Year Administration

allocation above Guide Rock in 2005 and in 2006, as estimated by Kansas, by 2,850 acrefeet.

96

82. Because § V.B.2.e. of the FSS explicitly provides for using two-year running averages for

Virgin Water Supply, Computed Water Supply, Allocations, and Computed Beneficial

Consumptive Use to determine whether Nebraska is in compliance with § V.B.2. but does

not explicitly address the amount of the violation when Nebraska is not in compliance with

§ V.B.2. and based on the comparisons in Finding 81, the two-year average of Nebraska's

exceedance of its Water-Short Year Administration allocation above Guide Rock for 2006

should not be used to determine the amount of Nebraska's violation for 2006. Rather, the

amount of Nebraska's violation for 2006 should be equal to Nebraska's exceedance of its

2006 Water-Short Year Administration allocation above Guide Rock. Similarly, the amount

of Nebraska's violation for 2005 should be equal to Nebraska's exceedance of its 2005

Water-Short Year Administration allocation above Guide Rock. Both Kansas and Nebraska

used Nebraska's exceedance of its Water-Short Year Administration allocation above Guide

Rock for both 2005 and 2006 to establish the amount Nebraska's violation during these

years,91, 90 although Kansas estimates the amount of the 2006 violation as being 36,100 acrefeet

whereas Nebraska estimates the amount of the 2006 violation as being 28,615 acre-feet,

a difference of 7,485 acre-feet.

92 Id.

93 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Attachment 2.

94 (42,860 acre-feet + 36,100 acre-feet) / 2.

95 81,810 acre-feet - (42,330 acre-feet + 36,100 acre-feet).

96 81,810 acre-feet - 78,960 acre-feet.

29

83. The primary reason for the difference of 7,485 acre-feet between Kansas' estimate of

Nebraska's 2006 violation and Nebraska's estimate is the assignment of evaporation from

Harlan County Lake. Kansas assigned evaporation to both Kansas and Nebraska,91 whereas

Nebraska assigned 100 percent of the Harlan County Lake evaporation to Kansas since only

KBID diverted water from Harlan County Lake in 2006.97

84. In the Arbitrator's Final Decision on Legal Issues, which is attached hereto, the Arbitrator

decided the following concerning Question 3:

The current Republican River Compact Administration Accounting Procedures allocate

evaporative losses from Harlan County Lake entirely to Kansas when the Kansas

Bostwick Irrigation District is the only entity actually diverting stored water from Harlan

County Lake for irrigation.98

This decision was based on the assumption that Nebraska did not "[choose] to substitute

supply for the Superior Canal from Nebraska's allocation below Guide Rock" in 2006

pursuant to § IV.A.e)(1) of the Accounting Procedures. The Arbitrator made this assumption

because in their respective briefs on legal issues, neither Kansas nor Nebraska identified

Nebraska's use of substitute supply for the Superior Canal from Nebraska's allocation below

Guide Rock in 2006.

85. On the last day of the arbitration hearing, Kansas introduced as its Exhibit 84 a copy of a

2006 letter from Nebraska which stated the following:

As identified in the Final Settlement Stipulation Section V.B.2.d., Nebraska is advising

you of the following measures Nebraska plans to take in anticipation of a Water Short

Year. The measures are cited by the corresponding Section in the Final Settlement

Stipulation:

V.B.2.a.i. - "supplementing water for Nebraska Bostwick Irrigation District by

providing alternate supplies from below Guide Rock or from outside the Basin".

Nebraska intends to enter into an agreement with the Nebraska Bostwick

Irrigation District whereby it is unlikely that Superior Canal will be diverting

surface water during 2006. ... Some irrigators in the Superior Canal surface

water delivery area will be using an alternate supply from ground water wells

located below Guide Rock Diversion Dam.99

This fact was not known by the Arbitrator when he decided Question 3.

97 Nebraska Exhibit 26, Electronic Data for Expert Report of Marc Groff, Tom Riley, and David Kracman, Review

of the 20 January 2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17,

2009, Excel Workbook NE 2006 Corrected, Tab Fed_Reservoir.

98 Arbitrator's Final Decision on Legal Issues at 10.

99 Kansas Exhibit 84, Letter from Ann Bleed, Acting Director, Nebraska Department of Natural Resources, to Hal

Simpson, Colorado State Engineer, David Pope, Kansas Chief Engineer, and Steve Raunshagen, Acting Area

Manager, Great Plains Region (USBR), May 1, 2006, p. 1.

30

86. In light of Finding 85 and given the explicit provision in § IV.A.e)(1) of the Accounting

Procedures pertaining to use of substitute supplies for the Superior Canal from Nebraska's

allocation below Guide Rock, a portion of the 2006 evaporation from Harlan County Lake

should be assigned to Nebraska.

87. The actual amount of groundwater diverted from wells below Guide Rock in 2006 is

unknown,100 which prevents a proportionate determination of the amount of Harlan County

Lake evaporation in 2006 that should be assigned to Nebraska. However, for 2005 the

allocation of net evaporation for Harlan County Lake between Kansas and Nebraska was

very nearly 50 percent for each state.101 Equally splitting the 2006 evaporation from Harlan

County Lake between Kansas and Nebraska using Kansas' 2006 net evaporation of 16,298

acre-feet102 or Nebraska's 2006 net evaporation of 16,182 acre-feet103 would increase

Nebraska's estimate of its Water-Short Year Administration exceedance above Guide Rock

in 2006 by about 8,100 acre-feet, for a total violation in 2006 of about 36,715 acre-feet. This

revised estimate of Nebraska's 2006 exceedance is sufficiently close to Kansas' estimate of

the 2006 violation of 36,100 acre-feet to justify acceptance of Kansas' estimate, which

allocated evaporation from Harlan County Lake "... based on long-term average uses."104

88. To provide a basis for estimating the direct economic impacts to Kansas caused by

Nebraska's exceedance of its Water-Short Year allocation above Guide Rock, the additional

amount of water that should have been available for use in Kansas was routed in accounting

simulations by the experts for Kansas and Nebraska to where the direct economic of impacts

of the shortages occurred: the farm headgates in KBID and downstream of KBID. To

perform these simulations the experts for both Kansas and Nebraska assumed that the

additional amount of water that should have been available for use in Kansas was regulated

through Harlan County Lake.105, 106

100 Kansas' Post-Trial Brief at 14.

101 Nebraska Exhibit 26, Electronic Data for Expert Report of Marc Groff, Tom Riley, and David Kracman, Review

of the 20 January 2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17,

2009, Excel Workbook NE 2005 With Comment, Tab Fed_Reservoir.

102 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Appendix A.

103 Nebraska Exhibit 26, Electronic Data for Expert Report of Marc Groff, Tom Riley, and David Kracman, Review

of the 20 January 2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17,

2009, Excel Workbook NE 2006 Corrected, Tab Fed_Reservoir.

104 Kansas' Post-Trial Brief at 14.

105 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, p. 2.

106 Nebraska Exhibit 8, Expert Report of Marc Groff, Tom Riley, and David Kracman, Review of the 20 January

2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17, 2009, p. 6.

31

89. Nebraska's experts used the same methods as Kansas' expert to estimate the additional net

evaporation from Harlan County Lake in 2005 and 2006 that would have resulted from the

additional supplies that should have been available for release from Harlan County Lake for

use in Kansas.107 Also, Nebraska's experts and Kansas' expert both assumed that the

conveyance losses between Harlan County Lake and the diversion to the Courtland Canal,

which conveys water to KBID, were insignificant in 2005 and 2006.108, 109

90. To estimate the conveyance losses between the Courtland Canal diversion and the Nebraska-

Kansas state line, Kansas' expert used the procedure for determining Courtland Canal losses

between the diversion and the state line chargeable to Kansas CBCU as specified in

§ IV.B.13. of the Accounting Procedures.110, 111 The Accounting Procedures specify that:

The allocation of transportation losses in the Courtland Canal above Lovewell between

Kansas and Nebraska shall be done by the Bureau of Reclamation and reported in their

"Courtland Canal Above Lovewell" spreadsheet. Deliveries and losses associated with

deliveries to both Nebraska and Kansas above Lovewell shall be reflected in the Bureau's

Monthly Water District reports. Losses associated with delivering water to Lovewell shall

be separately computed.

Amount of transportation loss of the Courtland Canal deliveries to Lovewell that does not

return to the river, charged to Kansas shall be 18% of the Bureau's estimate of losses

associated with these deliveries.112

The above provision sets the amounts of conveyance losses from Courtland Canal deliveries

to Lovewell Reservoir that do not "return to the river," which are chargeable to Kansas

CBCU, at 18 percent. The amounts of conveyance losses from Courtland Canal deliveries to

Kansas irrigators above Lovewell Reservoir that are chargeable to Kansas CBCU are to equal

"1-%BRF," where %BRF is defined as "Percent of Diversion from Bureau Canals that

returns to the stream."113

107 Id.

108 Id., p. 7.

109 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Appendix B (Note that the only

"Additional Transportation Losses" are for water diverted to the Upper Courtland unit and for water diverted for

delivery to Lovewell Reservoir).

110 Id., p. 2.

111 Republican River Compact Administration Accounting Procedures and Reporting Requirements, revised July

2005 (on title page, revised August 10, 2006), § IV.B.15. [sic], p. 33-34.

112 Id., p. 34.

113 Id., p. 25.

32

91. The losses from the Courtland Canal assigned to Kansas in 2005 and 2006 for deliveries to

Kansas irrigators and for deliveries to Lovewell Reservoir adopted by Kansas' expert114 are

the same as those reported for 2005 and 2006 in the RRCA Compact accounting spreadsheets

provided by Nebraska's experts,115 which reference the Bureau of Reclamation as the source.

For 2005 those losses total 8,651 acre-feet, and for 2006 the losses total 12,158 acre-feet.

92. The RRCA Compact accounting spreadsheets provided by Nebraska's experts confirm that

for 2005 and 2006, 18 percent of the conveyance losses from Courtland Canal deliveries to

Lovewell Reservoir were attributed to Kansas CBCU.116 The spreadsheets also show that for

2005 and 2006, 18 percent of the conveyance losses from Courtland Canal deliveries to

Kansas irrigators above Lovewell Reservoir, referred to as "Upper Courtland", "does not

recharge"117 as adopted by Kansas' expert118. Therefore, %BRF for both 2005 and 2006 was

82 percent.

93. Kansas' expert assumed that only the conveyance losses that do not recharge (i.e.,

consumptive losses) were lost from the Courtland Canal. As a result, Kansas' expert

estimated that the additional amount of water that would have been available at the

Nebraska-Kansas state line in 2005 for delivery to Kansas irrigators, but for Nebraska's

overuse, would equal the amount of Nebraska's exceedance (42,860 acre-feet), less the

additional net evaporation from Harlan County Lake (1,341 acre-feet), and less the average

of the conveyance losses "that do not recharge (18%)" as a percentage of Courland Canal

diversions over the period 1995 through 2006 (968 acre-feet), for an adjusted additional

supply of 40,551 acre-feet (rounded to 40,600 acre-feet).119 Using this same procedure for

2006, Kansas' expert estimated an adjusted additional supply of 32,605 acre-feet (rounded to

32,600 acre-feet). These are the additional amounts of water Kansas' expert assumed would

be available in the Courtland Canal at the Nebraska-Kansas state line for delivery to KBID in

2005 and 2006.120 This assumption is incorrect.

114 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Appendix B.

115 Nebraska Exhibit 26, Electronic Data for Expert Report of Marc Groff, Tom Riley, and David Kracman, Review

of the 20 January 2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17,

2009, Excel Workbooks NE 2005 With Comment and NE 2006 Corrected, Tab CourtlandAvLove.

116 Id., Tab MAINSTEM.

117 Id.

118 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Appendix B.

119 Id.

120 Id., Table 1.

33

94. As described in Finding 91, the total amounts lost from the Courtland Canal in Nebraska in

2005 and 2006 were 8,651 acre-feet and 12,158 acre-feet, respectively. Because these

amounts of water were lost from the Courtland Canal in Nebraska, these amounts of water

could not be in the Courtland Canal at the Nebraska-Kansas state line, even though only

18 percent of these losses (the consumptive losses) were allocated to Kansas CBCU.

Therefore, the actual amounts of water presumably determined by the Bureau of Reclamation

to be available in the Courtland Canal at the Nebraska-Kansas state line for delivery to KBID

in 2005 and 2006 were 40,086 acre-feet121 and 38,473 acre-feet,122 respectively, not the

amounts of 47,180 acre-feet and 48,442 acre-feet implied by the flawed assumption of

Kansas' expert.

95. Applying the computational methodology used by Kansas' expert to estimate the additional

amounts of water that would have been available in the Courtland Canal at the Nebraska-

Kansas state line in 2005 and 2006 for delivery to KBID, but using the average of the total

conveyance losses as a percentage of Courland Canal diversions over the period 1995

through 2006 instead of the average of the conveyance losses that do not recharge as a

percentage of Courland Canal diversions, results in adjusted additional supplies of 36,143

acre-feet123 and 29,060 acre-feet,124 respectively.

96. Some, if not all, of the amounts of water equal to the differences between the revised

estimates in Finding 95 and the estimates of Kansas' expert described in Finding 93 (i.e.,

non-consumptive losses of 4,408 acre-feet for 2005 and 3,545 acre-feet for 2006) would

reasonably be assumed to be available to Kansas as groundwater and as additional flow in the

Republican River. There is insufficient information in the record to allow a reasonably

reliable estimate of how this additional groundwater and flow in the Republican River might

have been used in Kansas. However, it is not reasonable to assume these amounts of water

would have been available to KBID at the Nebraska-Kansas state line from the Courtland

Canal. Kansas' expert has overstated the additional amounts of water that would have

available to KBID at the Nebraska-Kansas state line from the Courtland Canal, but for

Nebraska's overuse in 2005 and 2006, by at least approximately 12 percent.

97. Nebraska's experts use a different approach to estimate the additional amounts of water that

would have available to KBID at the Nebraska-Kansas state line from the Courtland Canal in

121 48,737 acre-feet less total losses of 8,651 acre-feet. This equals the quantity of water at Courtland Canal 15.1 in

Nebraska Exhibit 26, Electronic Data for Expert Report of Marc Groff, Tom Riley, and David Kracman, Review

of the 20 January 2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17,

2009, Excel Workbook NE 2005 With Comment, Tab CourtlandAvLove.

122 50,631 acre-feet less total losses of 12,158 acre-feet. This equals the quantity of water at Courtland Canal 15.1 in

Nebraska Exhibit 26, Electronic Data for Expert Report of Marc Groff, Tom Riley, and David Kracman, Review

of the 20 January 2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17,

2009, Excel Workbook NE 2006 Corrected, Tab CourtlandAvLove.

123 42,860 acre-feet, less additional net evaporation of 1,341 acre-feet, less total additional losses of 5,376 acre-feet.

124 36,100 acre-feet, less additional net evaporation of 2,717 acre-feet, less total additional losses of 4,323 acre-feet.

34

2005 and 2006, but for Nebraska's overuse in those years.125 While the methodology

employed by Nebraska's experts properly excluded all of the estimated canal losses from the

Courtland Canal in Nebraska, Nebraska's experts made no attempt to estimate the amounts of

canal losses that would have been available to Kansas as groundwater or as additional flow in

the Republican River. Nebraska's experts have understated the additional amounts of water

that would have available to Kansas below the Nebraska-Kansas state line in 2005 and 2006.

Damages - Direct Economic Impacts

98. To estimate the economic impacts (damages) incurred by irrigators within KBID and

downstream of KBID caused by overuse of water by Nebraska in 2005 and 2006, Kansas'

experts estimated the difference in irrigated and non-irrigated crop mix and yields between:

(1) the crop mix and yields Kansas' experts projected would have been realized, had overuse

not occurred in Nebraska and irrigators in Kansas received the full amount of water to which

they were entitled under the FSS; and (2) the reported crop mix and yields realized by

impacted Kansas farmers in 2005 and 2006. The crop prices used by Kansas' experts to

estimate the direct economic impacts as lost profits were the same for (1) and (2).126

99. To project irrigated crop yields that would have been realized, had overuse of water by

Nebraska not occurred, Kansas' experts utilized a crop-yield model called IPYsim, which is

named after irrigation and precipitation yield simulation.127 While based in part on cropyield-

water-response functions reported in Stone et al., 2006128 ("Stone's response

functions"),129 IPYsim differs from Stone's response functions in at least four respects that

are important. First, Stone's response functions were based on the response of crop yield to

precipitation and irrigation only,130 whereas the version of IPYsim employed by Kansas'

experts includes not only crop-yield response to precipitation and irrigation but also includes

125 Nebraska Exhibit 8, Expert Report of Marc Groff, Tom Riley, and David Kracman, Review of the 20 January

2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17, 2009, pp. 7-10.

126 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 178:24-179:4 (Kastens).

127 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 2.

128 Loyd Stone is a Professor of Agronomy at Kansas State University and was a rebuttal expert for Kansas in Kansas

v. Colorado, No. 105, Original. The Special Master appointed by the U. S. Supreme Court in this matter, Arthur

L. Littleworth, believed that "Professor Stone's testimony is entitled to great weight." See Third Report of Special

Master Littleworth, August 2000, p. 56.

129 Id.; Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 179:7-16 (Kastens).

130 See Kansas Exhibit 18, Water Supply: Yield Relationships Developed for Study of Water Management,

L. R. Stone, et al., Journal of Natural Resources & Life Sciences Education, Volume 35, 2006, p. 162.

35

crop-yield response to total usable nitrogen.131, 132 Second, Stone's response functions do not

include economic considerations,133 whereas IPYsim incorporates both nitrogen fertilizer

costs (average nitrogen fertilizer to crop price ratio by crop observed over the 1994-2000

time period) and water costs (after accounting for delivery efficiency).134 Third, Kansas'

experts adjusted the IPYsim response functions, as described in Finding 103, and did not

provide any information to verify the reasonableness of the resulting response functions that

were then used to assess impacts, whereas Stone's response functions were based on

empirical relationships; that is, relationships based on observations that can be verified or

disproved by observation or experiment.135 Fourth, Stone's response functions in Kansas'

Exhibit 18 were not developed or used to assess economic impacts. Rather Stone's response

functions were developed "for use in water resource education."136 While Stone's response

functions may be "similar in all material respects" to those used in Kansas v. Colorado, No.

105, Original, the IPYsim crop-yield response functions employed by Kansas' experts in this

arbitration proceeding are not,137 contrary to Kansas' assertion in its closing brief.138

100. The IPYsim response functions are quadratic and of the mathematical form: Y = A + BX -

CX2 where for a particular crop Y is the calculated yield, A, B, and C are positive numerical

constants, and X is the level of crop input.139 With this quadratic form, as X increases Y

131 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 2; Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 180:3-9 (Kastens); Kansas Exhibit 17,

Background for KSU-NPI_CropBudgets.xls, January 2009, p. 4 (referenced in FN 1 of Kansas Exhibit 5, p. 2).

132 When asked what effect the inclusion of phosphate would have on his analysis, as is done in a newer version of

IPYsim, Dr. Kastens testified:

Actually, I can't even answer the effect the nitrogen has on the analysis in terms of the , magnitude, say, of

the moneys owed. I have not done that. Too [sic] me - and I'm not even sure that I have the intuition,

without going back and studying it and analyzing it, what that would do.

Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 201:2-11.

133 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 173:11-16 (Kastens).

134 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 6.

135 Kansas Exhibit 18, Water Supply: Yield Relationships Developed for Study of Water Management, L. R. Stone, et

al., Journal of Natural Resources & Life Sciences Education, Volume 35, 2006.

136 Id., p. 162.

137 See Third Report of Special Master Littleworth, August 2000, p. 47-48.

138 Kansas' Post-Trial Brief at 21.

139 Kansas Exhibit 17, Background for KSU-NPI_CropBudgets.xls, January 2009, p. 4 (referenced in FN 1 of Kansas

Exhibit 5, p. 2).

36

increases at a diminishing rate until Y reaches its maximum value, after which Y begins to

decrease as X increases. The response functions have a horizontal slope when Y is at its

maximum value for a particular crop. Kansas' experts call this point "the maximum of the

quadratic plateau function that defines yield,"140 and the response function for a particular

crop is adjusted such that when Y is at its maximum value, it equals what Kansas' experts

term the "yield goal",141 which is defined as "the expected crop yield given that neither

nitrogen fertilizer nor water is limiting."142

101. The "yield goal" is determined using IPYsim by assuming that the economically optimal

yield for a particular crop, considering costs for nitrogen fertilizer and irrigation water,

equals what the Kansas' experts term "trend yield" for that crop.143 As a result of this

assumption, the "trend yield" for a particular crop must be less than or equal to the calculated

"yield goal" for that crop. The "trend yield" was determined by fitting a linear trend line

through the observed yields by year for each crop within KBID (excluding ensilage) for the

years 1962 through 2006, including or excluding yields during water-short years to derive the

maximum yield along the trend line for the year 2006. The resulting "trend yield" was used

for 2006 as well as 2005.144

102. The IPYsim response functions for each crop (excluding ensilage), adjusted such that the

"trend yield" equaled the economically optimal yield, as described in Finding 101, were then

used to simulate yields assuming KBID irrigators could have all of the irrigation water they

desired during 2005 and 2006 ("full irrigation") and to simulate yields for the actual water

available during 2005 and 2006.145 (It is not clear why Kansas' experts assumed KBID

irrigators could have all of the irrigation water they desired instead of assuming KBID

irrigators would have received the quantity of water to which they were entitled had there

been no overuse of water by Nebraska, although adjustments were subsequently made to

account for this difference.)146

103. For each crop in the areas above and below Lovewell Reservoir, the actual crop yields

reported for KBID were then multiplied by the ratio of the "full irrigation" yield simulated by

140 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 183:8-10 (Kastens).

141 Id.

142 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 6.

143 Id.

144 Id.

145 Id., p. 7.

146 Id., p. 9; Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 186:4-15 (Kastens).

37

IPYsim divided by the yield simulated for the actual amount of irrigation water received to

derive what Kansas' experts term the fully irrigated "expected yield."147 The effect of this

adjustment is to change the shape of the IPYsim response functions for each crop, assuming

the Y intercept of the function does not change, and to increase the "yield goal." For corn in

2005,148 for which the actual yield was 187 bushels/acre, this adjustment results in a fully

irrigated "expected yield" of 206 bushels/acre. If the relationship between fully irrigated

yield and "yield goal" remains proportionate or nearly proportionate, a fully irrigated

"expected yield" of 206 bushels/acre implies a "yield goal" of 212 bushels/acre. Both the

fully irrigated "expected yield" of 206 bushels/acre and the implied "yield goal" of 212

bushels/acre are close to the yield for maximum crop ET for corn from Stone et al., 2006,

14.0 megagrams/hectare or 222 bushels/acre.149

104. Kansas' experts did not use the adjustment procedure described in Finding 103 to derive the

fully irrigated "expected yield" for crops above Lovewell Reservoir in 2005 and instead

assumed the "expected yield" values above Lovewell Reservoir were the same as those

derived for crops below Lovewell Reservoir.150 Kansas' experts did not state why this

assumption was made, but applying the adjustment procedure described in Finding 103 for

corn in 2005 above Lovewell Reservoir would result in a fully irrigated "expected yield" of

258 bushels/acre, which is nearly 40 percent higher than the highest historical yield of 187

bushels/acre as of 2006 and more than 15 percent higher than the yield for maximum crop ET

for corn from Stone et al., 2006, which is clearly not reasonable.

105. The fully irrigated "expected yield" is associated with the expectation of irrigators in KBID

that all of the irrigation water "economically desired" would be available, which is more than

the amount of water KBID irrigators would have received had there been no overuse of water

in Nebraska.151 Therefore Kansas' experts revised the "expected yield" for each crop

downward to the yields simulated using the IPYsim crop response functions that would have

been realized for amounts of irrigation water equal to the actual amounts received plus the

147 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 7 and Table 10.

148 Kansas' experts identified corn as the most appropriate crop for this "base yield modeling framework ... since it is

the crop where yield-response-to-irrigation data are most prevalent and the crop most frequently managed in an

irrigation setting." Id., p. 7.

149 -11.55 + 0.416 x 61.3 = 14.0 megagrams/hectare, Kansas Exhibit 18, Water Supply: Yield Relationships

Developed for Study of Water Management, L. R. Stone, et al., Journal of Natural Resources & Life Sciences

Education, Volume 35, 2006, Table 2, p. 164.

150 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

Table 10.

151 Id., pp. 8-9.

38

additional amounts estimated by Kansas' experts152 that would have been received had there

been no overuse of water in Nebraska.153

106. Kansas' experts then used the revised crop-specific "expected yield" together with other

relevant factors for 2005 and 2006 with and without overuse of water in Nebraska including

actual crop yields (both irrigated and non-irrigated), growing season precipitation, acres

irrigated, irrigation technology and efficiency, irrigated crop mix, non-irrigated crop mix,

crop prices, and production costs to estimate the lost profit in KBID for 2005 and 2006 from

overuse of water in Nebraska. The estimated lost profits in KBID for 2005 and 2006 were

then divided by the amounts of farm-gate water shortages estimated from overuse of water in

Nebraska for 2005 and 2006, respectively, and the resulting value per acre-foot of water

shortage were multiplied by the estimated shortages caused by reductions in return flows

outside of KBID.154 The total direct economic impacts for each of 2005 and 2006 were

calculated as the sum of the estimated lost profit in KBID and the value of the estimated

shortages outside of KBID.155

107. The reasonableness of the estimates of total direct economic impacts in 2005 and 2006

proffered by Kansas' experts is dependent on the reasonableness of the many assumptions

made by Kansas' experts. Besides the estimated shortages in irrigation water resulting from

Nebraska's overuse of water in 2005 and 2006, the core of Kansas' estimates of total direct

economic impacts centers on the IPYsim crop response functions.

108. One of Kansas' experts, Dr. Terry Kastens, testified that although "IPYsim has not been

really academically reviewed, ... it has been very critically reviewed by many users who

continue to use it on a regular basis for making crop decisions."156 While IPYsim may have

been "critically reviewed by many users," Kansas did not provide or offer any evidence that

the adjusted IPYsim crop response functions used to estimate the fully irrigated "expected

yield" for crops in KBID, as described in Finding 103, have been peer-reviewed by anyone

other than the six authors of Kansas' expert report on this issue. While acknowledging that

the adjustments made to the IPYsim crop response functions described in Finding 103 were

152 Kansas Exhibit 1, Expert Report of Dale E. Book, Engineering Analysis of Losses to Kansas Water Users

Resulting from Overuse of Republican River Supply in Nebraska 2005 and 2006, January 20, 2009, p. 6.

153 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 9; Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 186:4-11 (Kastens).

154 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 8-9.

155 Apparently, the total direct economic impacts were not reduced to account for Federal income tax that would have

been paid on increased farm net income, as was done in Kansas v. Colorado. See Third Report of Special Master

Littleworth, August 2000, p. 72.

156 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 180:25-181:3.

39

"not suggested by Stone,"157, 158 Kansas did not provide or offer any empirical data

demonstrating that the adjusted IPYsim crop response functions and the estimates of fully

irrigated "expected yield" are consistent with actual observations.

109. The experts for Colorado and Nebraska on this issue were both critical of the adjustment of

the IPYsim crop response functions to estimate the crop-specific fully irrigated "expected

yield." In his report, Colorado's expert, Dr. James Pritchett stated the following:

In my opinion, the IPYsim model is accurate in suggesting the predicted yield under

actual irrigation is 90% of the predicted model yield under full irrigation. However, I do

not find documentation that the percentage difference [10%] may be applied to higher

yield levels with accuracy.

More specifically, the IPYsim model predicts that if the crop receives 6.12 fewer inches

of water than is necessary, a yield loss of 15.4 bushels (165.9 bu. - 150.5 bu.) results.

When scaled up, the EIA [Kansas Exhibit 5, Economic Impacts on Kansas of Diminished

Surface Water Supplies ...] reports that if the crop receives 6.12 fewer inches of water a

yield loss of 19.1 bushels (206.1 bu. - 187.0 bu.) results. Implicitly, at [sic] higher base

yield generates increasingly larger incremental yields with additional water. I believe

this to be inaccurate as the accepted relationship between applied water and crop yield is

one of diminishing returns.159

In his direct testimony, Dr. Pritchett testified:

What I do note is that in terms of its yield prediction, those seem to fit trend yields and

also the National Ag Statistic Service yields. And so I felt comfortable in that sense, that

the yields [Model Yield in Table 10, Kansas Exhibit 5] were representative.

Later, the Kansas experts boot-strapped those yields to a higher level [fully irrigated

Expected Yield in Table 10, Kansas Exhibit 5] and I'm not sure I'm comfortable with

that.160

Nebraska's expert, Dr. David Sunding, testified in his direct testimony:

So now the next step in what they describe as their calibration procedure, we have Stone

down here. We have the quote/unquote, calibrated IPYsim to hit their assumptions about

the 2005 trend yield.

157 It is unknown why Kansas did not utilize Professor Loyd Stone of Kansas State University as an expert witness on

this issue, given that his testimony in Kansas v. Colorado, No. 105, Original, was given great weight.

See FN 128.

158 Transcript of Arbitration Proceedings, March 11, 2009, Volume III at 498:7-10 (Kastens).

159 Colorado Exhibit 2, Expert Report of Dr. James Pritchett, Reviewing the Assumptions, Methods and Results of:

Economic Impacts on Kansas of Diminished Surface Water Supplies to the Lower Republican River Basin Caused

by Nebraska in 2005 and 2006, February 16, 2008 [sic], p. 6.

160 Transcript of Arbitration Proceedings, March 10, 2009, Volume II at 287:6-13.

40

Well, as you just pointed out, actual yield was somewhere up here, again off the front tier

[sic].

So how do we deal with that?

And the way they deal with that is simply by taking the ratio between these two points

and applying it up here. So whatever this vertical distance is, they take the actual

observed yield and boost it up by that amount. That was what Dr. Pritchett referred to as

this boot-strapping procedure.

So this is the 187. And this is, I believe, 206, which is, as Dr. Kastens described, 10

percent higher than the highest observed yield ever; and I think, frankly, lacking

credibility.161

...

Now, why does that matter? That matters because the heart of their valuation analysis or

their damage analysis is to answer the question: What would have been the extra yield

and, hence, the extra profit earned from a few extra units of water, few extra inches of

water per acre?

So this slope matters a lot for their damage analysis. It's not derived from Stone. It is, I

would submit, totally made up to fit this particular trend yield and, therefore, I think

inadequate as a basis for a damage calculation.162

110. Kansas' expert report on economic impacts states that: "IPYsim was developed using

expected yield response to water data reported in Stone et al., 2006, which were the same

data underlying KSU's Crop Water Allocator (KSU-CWA)."163 Stone et al. states that:

"Crop-water production relationships are altered by variations in soil and climate and have

not been well defined for most crops in most areas (internal citations omitted)."164 However,

Kansas' experts did not address variations in soil types and climate between western Kansas,

for which Stone's response functions were developed, and north-central Kansas several

hundred miles to the northeast, where KBID and the other impacted areas in Kansas are

located, other than in Dr. Kasten's testimony when he stated:

And though it's said that, you know, it makes a point, for example, about soil types

mattering, we don't believe that the difference in the silt loam soils of western Kansas

161 Id., at 322:4-20.

162 Id., at 323:16-324:1.

163 Kansas Exhibit 5, Economic Impacts on Kansas of Diminished Surface Water Supplies to the Lower Republican

River Basin Caused by Nebraska in 2005 and 2006, Dr. Bill Golden et al., January 20, 2009, p. 2.

164 Kansas Exhibit 18, Water Supply: Yield Relationships Developed for Study of Water Management, L. R. Stone,

et al., Journal of Natural Resources & Life Sciences Education, Volume 35, 2006, p. 161.

41

and those of the KBID area, for example, are sufficiently large that they would diminish

our efforts of using this model specifically for KBID.165

Kansas did not provide or offer any empirical data confirming Dr. Kasten's testimony and

did not address the significance of any climate variations.

111. Since the assumed lack of significance of soil and climate variations and the methodology

applied by Kansas' experts for the purposes of estimating lost profits and establishing

damages have not been shown to be reasonable, the assumptions and methodology should be

validated by peer review or by empirical data before being accepted for the purposes of

estimating lost profits and establishing damages. Even if validated, the estimates of lost

profits can not be adopted because Kansas has overstated the additional amounts of water

that would have available to KBID, but for Nebraska's overuse in 2005 and 2006, as

described in Finding 96.166 The preponderance of evidence at this juncture does not support

the assumed lack of significance of soil and climate variations, the methodology used, or the

estimates of the total direct economic impacts in 2005 and 2006 made by Kansas' experts

with reasonable certainty.

112. The alternative estimates of total direct economic impacts developed by Nebraska's expert,

Dr. David Sunding, based on the difference between the rental rates paid by farmers to rent

irrigated land in 2005 and 2006 and the rental rates paid for non-irrigated land are not

sufficiently reliable. Dr. Sunding relied on land prices and cash rental rates for 2005 and

2006 published by the Kansas State University Agricultural Experiment Station and

Cooperative Extension Service.167 The introduction for this published data contains the

following qualifier:

These data are useful to farm managers in determining cash rental rates, to farmland

appraisers in calculating indexes for making time adjustments to land prices, and to

landowners and investors who base expectations on historical price and return levels for

farmland. The average prices in this guide encompass parcels of land that vary widely in

165 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 182:16-22.

166 When asked what the effect would be if the estimated amounts of additional water that should have been available

to KBID were reduced, the following exchange occurred:

DR. KASTENS: I can't say exactly. I can say that the dollars per acre-foot likely would go up. The total

dollars likely would go down, but I can't say to what magnitude.

MR. WILMOTH: Thank you.

ARBITRATOR DREHER: So Mr. Wilmoth, just so I understand. It's not a linear relationship then?

DR. KASTENS: That's correct.

Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 216:4-12.

167 Nebraska Exhibit 6, Expert Report of Dr. David Sunding, Analysis of Kansas' Economic Losses Caused by

Nebraska's Overuse of Water in the Republican River Basin in 2005 and 2006, February 17, 2009, p. 14.

42

productivity. Thus, these data are more appropriate for analyzing trends than for

establishing market value or rental rates for specific tracts of farmland.168

The limited applicability of the data relied on by Dr. Sunding was further confirmed by the

following testimony of Dr. Kastens, who was co-publisher of the data:

I don't like to say we don't trust the data, but we don't. And I can say that because

anybody that has ever heard me speaking in Kansas have heard us say this for years and

for hundreds of presentations, the irrigated rent data in Kansas, we don't believe them.

That's all I can say.

We have plenty of anecdotal evidence to suggest otherwise, but we don't believe the data

and so we don't use them for anything.169

113. In its closing brief, Nebraska argues that: "When checked against reality, it is clear Kansas

suffered relatively little economic harm from any loss of Republican River water she

sustained."170 Nebraska further concludes that: "In sum, the actual, direct economic harm

suffered by Kansas as a result of Nebraska's overuse is somewhere between ‘nearly zero' and

$930,630.00."171 Yet in 2006, Nebraska172 may have spent as much as $3.5 million173 to

lease a total of 23,518 acre-feet of surface water in Nebraska from the Frenchman Valley

Irrigation District, Riverside Irrigation Company, and Bostwick Irrigation District in

Nebraska.174 The leased surface water was relinquished by Nebraska for diversion by KBID

at the Guide Rock Diversion Dam.175 Nebraska would not have paid $ 3.0 or $3.5 million to

lease 23,518 acre-feet of surface water, for an average volume-weighted unit cost as high as

$149/acre-foot,176 if the additional water that would have been available to KBID but for

overuse by Nebraska had an economic value of nearly zero.

168 Id., p. 1 of attachment marked MF-1100 in upper right-hand corner.

169 Transcript of Arbitration Proceedings, March 11, 2009, Volume III at 518:19-519:2.

170 State of Nebraska's Post-Hearing Brief at 17.

171 Id. at 22.

172 The Middle Republican Natural Resources District paid $50,000 of the total. Kansas Exhibit 44, p. 1; Kansas

Exhibit 51, p. 2.

173 Kansas Exhibit 44 shows $3.0 million paid to Bostwick Irrigation District in Nebraska whereas Kansas Exhibit 52

shows $2.5 million plus $64,500 was paid to the District.

174 Kansas Exhibit 44, Memorandum to Jeanne Glenn from Ann Bleed, March 5, 2007, p. 1.

175 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, p. 12.

176 $3,500,000 / 23,518 acre-feet.

43

114. Other than the leasing transactions by the state of Nebraska described in Finding 113, there is

no evidence in the record of an active water market in or adjacent to south-central Nebraska.

Therefore, the unit cost that Nebraska paid to lease water in its attempt to comply with the

FSS in 2006 is not the same as the unit value of water to Kansas from lost profits due to

overuse by Nebraska in 2006. As Nebraska's expert correctly noted regarding Nebraska's

lease payments:

So you have basically a monopolist, on one side, as opposed to what you would have in a

land rental market, where you have many participants on either side of the transaction.177

115. The alternative estimates of total direct economic impacts in 2005 and 2006 developed by

Colorado's expert, Dr. James Pritchett, based on modifications to the methodology used by

Kansas' experts are also not sufficiently reliable. Dr. Pritchett used the IPYsim crop

response functions to predict yield under actual irrigation and under full irrigation and did not

perform the adjustment described in Finding 103 to adjust the response functions upward to

the fully irrigated "expected yield."178 However, Dr. Pritchett used crop production costs

from northwest Kansas, which is predominantly irrigated using groundwater from the

Ogallala Aquifer,179 and did not investigate whether these costs were comparable to the crop

production costs in the KBID, which is predominantly irrigated using surface water.180

Because the production costs associated with using groundwater from the Ogallala Aquifer in

northwest Kansas include pumping costs to lift water from wells that are 250 ft to 300 ft

deep,181 as compared to the pumping costs to operate "relatively small centrifugal [booster]

pumps" to deliver surface water to center pivots in KBID,182 the farm production costs used

by Dr. Pritchett are not representative of the farm production costs in KBID. Since the

alternative estimates of total direct economic impacts in 2005 and 2006 developed by

Dr. Pritchett necessarily incorporate his estimates of farm production costs, his estimates of

lost profits in 2005 and 2006 are not sufficiently reliable.

116. There presently is not a sufficiently reliable basis to form an appropriate recommendation for

awarding damages to Kansas for overuse of water by Nebraska in 2005 and 2006. Clearly

Kansas incurred damages and those damages may well be in the range of one to several

million dollars. However, until such time Kansas can demonstrate with a preponderance of

evidence that its assumptions and methodology for estimating lost profits, including its

estimate of the amount of water that would have been available at the headgates of Kansas

177 Transcript of Arbitration Proceedings, March 10, 2009, Volume II at 374:22-25 (Sunding).

178 Colorado Exhibit 2, Expert Report of Dr. James Pritchett, Reviewing the Assumptions, Methods and Results of:

Economic Impacts on Kansas of Diminished Surface Water Supplies to the Lower Republican River Basin Caused

by Nebraska in 2005 and 2006, February 16, 2008 [sic], p. 6.

179 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 125:25-126:3 (Ross).

180 Id. at 121:13-5; Transcript of Arbitration Proceedings, March 10, 2009, Volume II at 292:7-293:25.

181 Transcript of Arbitration Proceedings, March 9, 2009, Volume I at 125:18-126:3 (Ross).

182 Id. at 124:3-17.

44

irrigators, and establishing actual damages is reasonably reliable (either through independent

peer review or with empirical data), during subsequent arbitration or before the Court, only

an award of nominal damages should be made.

Damages - Indirect Economic Impacts

117. Kansas' experts estimated indirect economic impacts from their estimates of reduced farm

income resulting from Nebraska's overuse of water in 2005 and 2006 by modeling the

Kansas state economy using an input-output accounting system termed "Social Accounting

Matrix" ("SAM"). The SAM system used by Kansas' experts was the Micro-IMPLAN

(Impact analysis for PLANing) system, which was also used to estimate indirect or secondary

impacts in Kansas v. Colorado, No. 105, Original.183

118. The indirect economic impacts, or "Value Added Impact" or "Indirect Value Added Loss"

estimated by Kansas' experts for both 2005 and 2006 are listed in Table 16 of their report184

and total 44 percent of the direct economic impacts (gross income loss), meaning that total

economic impacts were estimated to be 1.44 times the estimated direct economic impacts.185

119. In his report, Colorado's expert stated that:

While I have not been able to independently verify the SAM used in the EIA [Kansas

Exhibit 5, Economic Impacts on Kansas of Diminished Surface Water Supplies ...], the

multiplier [1.44] is consistent with my own research in the regional economic activity

generated by irrigated agriculture.186

120. Nebraska's expert stated in his report that:

While the method is standard, the use of IMPLAN to assess indirect impacts resulting

from changes in water availability is fraught with problems relating to the generally poor

quality of the input purchase and consumer expenditure data, including information on

"export" coefficients, for rural area in the United States.187

183 Kansas Exhibit 5, Expert Report of Dr. Bill Golden et al., Economic Impacts on Kansas of Diminished Surface

Water Supplies to the Lower Republican River Basin Caused by Nebraska in 2005 and 2006, January 20, 2009,

p. 9-10.

184 Id., p. 21.

185 Id., Table 16 and Table 17, p. 21.

186 Colorado Exhibit 2, Expert Report of Dr. James Pritchett, Reviewing the Assumptions, Methods and Results of:

Economic Impacts on Kansas of Diminished Surface Water Supplies to the Lower Republican River Basin Caused

by Nebraska in 2005 and 2006, February 16, 2008 [sic], p. 13.

187 Nebraska Exhibit 6, Expert Report of Dr. David Sunding, Analysis of Kansas' Economic Losses Caused by

Nebraska's Overuse of Water in the Republican River Basin in 2005 and 2006, February 17, 2009, p. 4. Also, see

Transcript of Arbitration Proceedings, March 10, 2009, Volume II at 363:15-364:17.

45

When asked whether a multiplier of "1.44 would be appropriate for indirect effects or do you

think it's too high or too low?"188 Nebraska's expert responded:

I think it's a - well, it's hard to know for sure if it's too high or too low without getting in

supplemental information specific to Kansas that I discussed; but within the confines of

the analysis that Kansas has proffered, I think the multiplier would be the same for both

years. 1.44, I think, is not out of the realm of what I have seen in other contacts [sic], so

that particular part of their analysis didn't stick out particularly.189

121. Nebraska's expert also stated in his report that:

More importantly ... indirect impacts are not a legitimate consideration in a proceeding

of this type ... because any damage payment from Nebraska to Kansas will generate its

own multiplier effects, and a damage payment that compensates for direct losses should

result in indirect benefits that compensate for indirect losses.190

122. In response, Kansas' expert, Dr. John Leatherman, testified that:

[T]heoretically, there could, in fact, be offsetting impacts, positive impacts associated

with the payments versus the damage occurred by the loss of family income. But, once

again, that would be under a very narrow set of circumstances. You would essentially

have to replicate as closely as possible in terms of the amount of damage, as well as the

timing of those payments, as well as what ultimately happened to stimulate economic

activity. And, here again, it's simply not feasible. Indeed, the State of Kansas, perhaps,

would take any - any type of moneys awarded to them and they would - they would do

something with that; but exactly what, I really don't know. And so that is something that

would be very speculative on my part to try to estimate any kind of offsetting damages,

absent there being specific information with regard to how they would spend the

money.191

123. During cross, Nebraska's expert testified that:

There are indirect impacts and I have never challenged that in this case. I do challenge

their relevance to the proceeding going on here, both because I have questions about the

reliability of the results and the Kansas analysis failed to consider the indirect benefits

that result from Nebraska's payments.192

188 Id. at 371:1-2.

189 Id. at 371:3-11.

190 Nebraska Exhibit 6, Expert Report of Dr. David Sunding, Analysis of Kansas' Economic Losses Caused by

Nebraska's Overuse of Water in the Republican River Basin in 2005 and 2006, February 17, 2009, pp. 4, 2.

191 Transcript of Arbitration Proceedings, March 10, 2009, Volume II at 264:14-265:8.

192 Transcript of Arbitration Proceedings, March 10, 2009, Volume II at 364:18-23 (Sunding).

46

124. Even though the indirect benefits resulting from Nebraska's payments may be "speculative,"

they are nonetheless real, and Kansas' experts should have attempted to reasonably quantify

them.

125. In Kansas v. Colorado, No. 105, Original, the Court accepted the use of the IMPLAN model

to assess secondary impacts to the economy of Kansas, and did not consider the indirect

benefits that result from Colorado's payment of money damages.193 However, based on the

testimony of different experts for Kansas in that case, the Court found that "[s]econdary

economic impacts are also affected by a concept known among economists as ‘opportunity

costs'"194 and that "[o]nly 20 percent of the total secondary impacts were counted as net

gains or losses."195

126. There is no evidence in the record for this proceeding whether opportunity costs offsetting or

reducing gross secondary impacts were considered by Kansas' experts or whether such

offsets are even relevant.

127. Since an award of only nominal damages for direct economic impacts is recommended in this

proceeding, no award of damages for indirect economic impacts should be made.

128. If Kansas seeks to demonstrate with a preponderance of evidence that its assumptions and

methodology for estimating lost profits and establishing actual damages is reasonably

reliable during subsequent arbitration or before the Court, Kansas should also attempt to

reasonably quantify indirect benefits resulting from Nebraska's payment for actual damages

and should also include any offsetting opportunity costs if relevant.

Future Compliance

129. To ensure future compliance with the FSS, "Kansas has proposed that Nebraska reduce its

groundwater-irrigated acreage in the Basin by approximately 515,000 acres of approximately

1.2 million acres which receive groundwater irrigation in the Nebraska portion of the

Basin."196 This would represent a reduction of 43 percent from the approximately 1.2 million

acres in the Nebraska portion of the Republican River Basin estimated by Kansas as being

193 Third Report of Special Master Littleworth, August 2000, p. 65-71.

194 Id., p. 68.

195 Id., p. 69.

196 Kansas Exhibit 6, Expert Report of David W. Barfield, Ensuring Future Compliance by Nebraska, January 20,

2009, § III. Remedies.

47

irrigated with groundwater, which Kansas's experts estimate would reduce consumptive

groundwater withdrawals by an average of 619,000 acre-feet per year.197

130. To derive the amount of reduction in groundwater-irrigated acreage proposed by Kansas, one

of Kansas' experts on this issue, Mr. Dale Book, first estimated the reduction in the Nebraska

groundwater CBCU that would have been necessary for compliance with the FSS on a 5-year

average basis for the years 2002 through 2006 as follows:

... I reviewed and utilized the Compact Administration, RRCA, the accounting data for

the five years. I compared the results of the beneficial consumptive use in the state of

Nebraska with the Nebraska allocation and computed the difference and determined what

the resulting required reduction in beneficial consumptive use would be to achieve a

balance between the allocation and consumptive use for the five years. I then made an

estimate of the amount of reduced consumptive use resulting from reducing groundwater

pumping that would be resulting in increased surface water use within the state of

Nebraska [45 percent of the reduction in groundwater CBCU] and adjusted for that in the

calculation. The result of the analysis was a recommendation for a level of groundwater

consumptive use that would balance with the allocations for this five-year period.198

...

The imported water supply credit ... was obtained from the RRCA Groundwater Model

results with the - this level of pumping and that was averaging 30,000 acre-feet per year.

The result is a balance for the five-year period.199

The result of this analysis is an ongoing, year-to-year, estimated limitation on groundwater

CBCU in the Nebraska portion of the Republican River Basin of 175,000 acre-feet.200

131. Assuming that 45 percent of the reduction in groundwater CBCU would approximately equal

the amount of increased streamflow resulting from curtailment of groundwater irrigation that

would then be consumptively used by surface water irrigators in Nebraska188 has the effect of

increasing the amount of the reduction in groundwater CBCU that must be achieved to

comply with the FSS. While reducing groundwater CBCU in Nebraska would clearly

increase streamflows in Nebraska, a portion of which would undoubtedly be diverted and

consumed by surface water irrigators, there is presently insufficient evidence to support the

assumption that the increased surface water CBCU in Nebraska would equal 45 percent of

the reduction in groundwater CBCU.

132. The RRCA Groundwater Model was then used:

197 Kansas Exhibit 3, Expert Report of Samuel P. Perkins and Steven P. Larson, Attachment 5: RRCA groundwater

model analysis (revised) Impact of Nebraska pumping and proposed remedy, January 4, 2008, p. 4.

198 Transcript of Arbitration Proceedings, March 11, 2009, Volume III at 533:9-534:1.

199 Id. at 539:3-7.

200 Kansas Exhibit 2, Expert Report of Dale E. Book, Requirements for Nebraska's Compliance with the Republican

River Compact, January 20, 2009, p. 3-4 and Table 1.

48

... in a trial-and-error process ... [to look] at various levels of curtailment of pumping,

again focusing on, in part, looking at what we call quick response areas, or areas near the

stream system that would respond relatively quickly to reductions in groundwater

irrigation and upland areas that respond more slowly, looking at combinations of those to

determine how much reduction would be necessary in order to achieve the level of

groundwater consumptive use that Mr. Book had determined.

Ultimately, what we determined was that if we -- if we curtailed pumping within about

2 ½ miles of the stream system and if we also held the pumping outside that -- that

corridor along the stream system to the amount of acreage that was in place in the year

2000, that the combination of those two things would produce a reduction in groundwater

beneficial consumptive use that would, over the long haul, stay below the level that

Mr. Book had determined.201

In the simulated reductions of groundwater consumption using the RRCA Groundwater

Model, the amount of irrigated acreage using comingled groundwater and surface water

supplies was "held at 2006 levels at all distances from stream cells within the Republican

River basin in Nebraska."202 The result of this analysis was a reduction of "350,970 acres

within the no-pumping zone and 163,640 acres outside the no-pumping zone."203

133. In performing the simulations described in Finding 132:

Model datasets for historical years 1990-2006 were used to construct future scenarios.

These years were chosen initially because of the higher quality of Kansas water use

reporting data beginning in 1990. The sequence of historical years 1990-2006, beginning

with year 1990, was repeated three times to represent future scenarios for years 2007-

2057. Median annual precipitation for years 1990-2006, spatially averaged over the

groundwater model domain, is 19.58 inches/year. Compared against the model's years of

record 1918-2006, this corresponds to a probability of 54.5 percentile, which is slightly

above median rainfall of 19.28 in/yr for years 1918-2006. This indicates that the

sequence is a reasonable projection, at least with respect to the historical record.

Additionally, the sequence consists of a relatively wet period (1990-1999) followed by a

relatively dry period (2000-2006).204

Nebraska's experts on this issue reported that the annual precipitation for the years 1990 -

2006 was at the 60th percentile, meaning that the annual precipitation for this period of years

201 Transcript of Arbitration Proceedings, March 11, 2009, Volume III at 554:20-555:14 (Larson).

202 Kansas Exhibit 3, Expert Report of Samuel P. Perkins and Steven P. Larson, Attachment 5: RRCA groundwater

model analysis (revised) Impact of Nebraska pumping and proposed remedy, January 4, 2008, p. 1.

203 Id.

204 Id., pp. 1-2.

49

was above average and equaled or exceeded 60 percent of the measurements of annual

precipitation over the longer term of 1918 through 2006.205

134. Because of the nonlinear response of the RRCA Groundwater Model when stream-drying

occurs,206 introducing streamflow to de-watered streams in the RRCA Groundwater Model

increases the simulated streamflows that can be depleted by groundwater consumption,

which increases groundwater CBCU. For example, 1993 was a year with unusually high

amounts of precipitation,207 and 1993 was used to represent the years 2010, 2027, and

2044208 in Kansas' simulations using the RRCA Groundwater Model described in Finding

132. For each of the three years during the simulations, when the dataset for 1993 is

introduced (i.e., 2010, 2027, and 2044), computed impacts from pumping in Nebraska

increase significantly, except for the simulation of Kansas' proposed remedy.209 The reason

why simulated impacts from pumping in Nebraska do not increase significantly in 2010,

2027, and 2044 for the simulation of Kansas' proposed remedy may result from the reduction

in the acreage irrigated with groundwater being so significant that simulated de-watering of

streams is relatively limited and the response of the Groundwater Model is for the most part

linear.

135. Kansas has adequately demonstrated that its proposed remedy would result in Nebraska's

compliance with the FSS, even during dry-year conditions similar to what occurred during

the period 2002 through 2006.210 However, given the magnitude of the assumed increase in

205 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, p. 16.

206 See Finding 20.

207 MR. DRAPER: Dr. Schneider, you've mentioned several times that 1993 was the wettest year on

record?.

DR. SCHNEIDER: I may not be completely accurate on that. I believe I'm referring to the

rainfall precipitation gages within the model that are located in Nebraska and looking at the --

that's generally what I'm looking at. And if it's not the wettest year, it's second or third, but it's

my -- it's my recollection that it's the wettest year in terms of precipitation in Nebraska.

MR. DRAPER: In fact, I have no quarrel with that. I think it's often referred to as the "Great

Flood of 1993," isn't it?

Transcript of Arbitration Proceedings, March 13, 2009, Volume V at 940:10-23.

208 See Finding 133.

209 See Kansas Exhibit 65, Comparison of Nebraska pumping impact under baseline conditions, Kansas proposed

remedy, and NRD Pumping Alternatives, 3/16/2009.

210 For this decision, the period of years 2002 through 2006 is considered a period of dry years, even though the

probability of non-exceedance over the period of record (1918 - 2007) for precipitation in the Nebraska portion of

the Republican River Basin during 2004 through 2006 was more than 0.5 (See Kansas Exhibit 6, Expert Report of

David W. Barfield, Ensuring Future Compliance by Nebraska, January 20, 2009, Figure 7), since both 2005 and

2006 were years of Water-Short Year Administration.

50

surface water CBCU from reductions in groundwater CBCU described in Finding 131 and

the fact that Kansas' experts used datasets from years when precipitation was above average

overall as described in Finding 133, Kansas' experts likely have overstated the amount of

reduction in groundwater irrigated acreage that is necessary in Nebraska for Nebraska to

comply with the FSS. Therefore, Kansas has not adequately demonstrated that its proposed

remedy is the "minimum remedy necessary for compliance" as it has asserted.211 Based on

the testimony and evidence in the record for this proceeding, it is not possible to reasonably

assess the extent that Kansas' experts may have overestimated the reduction in groundwater

irrigated acreage in Nebraska that is necessary for Nebraska's compliance with the FSS.

136. Nebraska asserts that:

Following the signing of the FSS, Nebraska has implemented landmark changes to its

system of water regulation. The resulting integrated management planning process

mandates a cooperative effort between the Department [of Natural Resources]

(historically responsible for surface water administration), and the NRDs [Upper

Republican Natural Resources District, Middle Republican Natural Resources District,

and Lower Republican Natural Resources District] (historically responsible for

groundwater management). Taking into account all proposed future scenarios by Kansas

and Nebraska, and assuming there are no changes to the current RRCA Accounting

Procedures, Nebraska will under the worst case, have only a modest shortfall of 8,288

acre feet on average (less than 3.5%). Recently, through dry year leasing of surface water

supplies, Nebraska has shown the ability to make up substantially greater than this

amount annually. We are confident the IMPs [Integrated Management Plans] are more

than sufficient to maintain compliance with the Compact [and the FSS] through 2012,

when they will be reevaluated and modified to ensure compliance into the future.212

137. One of Nebraska's experts, Mr. Williams, testified that the Upper Republican Natural

Resources District (URNRD), Middle Republican Natural Resources District (MRNRD), and

Lower Republican Natural Resources District (LRNRD) account for 95 percent of the

depletions to surface water sources in the Republican River Basin caused by consumptive

groundwater withdrawals.213 The Nebraska Department of Water Resources and each of

these three NRDs jointly developed an individual Integrated Management Plan and

associated rules and regulations ("IMP") for each NRD.214 While there are differences

between each of the IMPs, the three IMPs are substantially similar. Each IMP, as revised in

211 Kansas Exhibit 6, Expert Report of David W. Barfield, Ensuring Future Compliance by Nebraska, January 20,

2009, § III.a.

212 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, p. 18.

213 Transcript of Arbitration Proceedings, March 13, 2009, Volume V at 829:7-9; 831:24-832:2.

214 Id. at 964:10-16 (Dunnigan).

51

late 2007 or early 2008,215 generally has three increasingly stringent requirements limiting

consumptive groundwater withdrawals, although the IMP for the LRNRD only has two

requirements. The first requirement is a limitation on the amount of groundwater that may be

withdrawn and applied to crops by individual irrigators. The second, and more stringent,

requirement is a limitation on the average annual volume of groundwater withdrawals for

each NRD, averaged over the period 2008 through 2012, which is 20 percent less than the

baseline average groundwater withdrawals for the years 1998 through 2006, excluding the

LRNRD in which the allotments for individual irrigators were further reduced with the intent

of achieving a 20 percent reduction from the 1998 through 2006 baseline.216 The average

annual groundwater withdrawals for the URNRD, MRNRD, and LRNRD during the period

of 1998 through 2006 are reported to be 531,763 acre-feet, 309,479 acre-feet, and 242,289

acre-feet, respectively, totaling just more than 1,083,530 acre-feet per year.217 The

limitations on the average annual volume of groundwater withdrawals for the URNRD and

MRNRD, averaged over the period 2008 through 2012, are 425,000 acre-feet and 247,580

acre-feet, respectively.218, 219 The intended limitation for the LRNRD is 193,830 acre-feet.220

The sum of the required limitations on the average annual volume of groundwater

withdrawals for the URNRD and MRNRD plus the intended limitation for the LRNRD total

866,410 acre-feet per year, a reduction of 217,120 acre-feet from the 1998 - 2006 average of

1,083,530 acre-feet per year.

The third and most stringent requirement, at least during dry years, is a limitation on either

the annual net groundwater depletions (URNRD and LRNRD) or the groundwater depletions

averaged over the period 2008 through 2012 (MRNRD). The net groundwater depletions for

the URNRD, MRNRD, and LRNRD are not to exceed 44 percent, 30 percent, and 26

percent, respectively, of Nebraska's allowable groundwater CBCU determined from using

the RRCA Groundwater Model.221, 222, 223 Although the limitations on net groundwater

215 For IMPs adopted for URNRD, MRNRD, and LRNRD, respectively, see Nebraska Exibits: 16; 17; and 15,

Appendix A.

216 Transcript of Arbitration Proceedings, March 13, 2009, Volume V at 893:7-13; 963:3-10 (Williams).

217 Nebraska Exhibit 16, Integrated Management Plan Jointly Developed by the Department of Natural Resources

and the Upper Republican Natural Resources District, p. 2.

218 Id., p. 7.

219 Nebraska Exhibit 17, Rules and Regulations and the Integrated Management Plan for the Middle Republican

Natural Resources District and the Nebraska Department of Natural Resources, February 8, 2008, p. 8 (Integrated

Management Plan revised January 8, 2008).

220 242,289 acre-feet x 0.80.

221 Nebraska Exhibit 16, Integrated Management Plan Jointly Developed by the Department of Natural Resources

and the Upper Republican Natural Resources District, p. 7.

222 Nebraska Exhibit 17, Rules and Regulations and the Integrated Management Plan for the Middle Republican

Natural Resources District and the Nebraska Department of Natural Resources, February 8, 2008, p. 8-9

(Integrated Management Plan revised January 8, 2008).

52

depletions for the URNRD and LRNRD are stated as annual requirements in the respective

IMPs, these are effectively average limitations, at least for a two-year period, since the

accounting is done after-the-fact during the following year. Consequently, whether or not

compliance with the FSS was achieved and whether further reductions in groundwater use

are needed is not known until the year following the year in which the groundwater

depletions actually occurred.

138. The IMPs for the URNRD, MRNRD, and LRNRD have considerable flexibility in that

average limitations are used, meaning that the limitations can be exceeded during any given

year. The IMPs also provide for variances, carryover of unused individual allocations,

pooling of individual allocations (URNRD and MRNRD), and bonus inches (MRNRD) when

compliance is achieved in a preceding year. Despite this flexibility, a careful reading of the

IMPs indicates that there are no exceptions to the overall limitations on the average annual

volume of groundwater withdrawals for the URNRD and MRNRD, as well as the overall

limitations on allowable net groundwater depletions for all three Republican River NRDs.

139. When asked whether the IMPs were enforceable, the Nebraska official responsible for

ensuring compliance with the Compact and the FSS, Mr. Brian Dunnigan224, answered:

"Absolutely."225 When asked "what happens if an NRD refuses to honor an IMP?"226

Mr. Dunnigan answered as follows:

Well, certainly the department would look at that; and if there was an issue with that, we

would certainly confer with the Attorney General's office to see if action would be taken

by the State against [the] Natural Resources District. The department could also look at

and the State could look at enforcement actions against individuals.227

When asked what if there is a failure of compliance, Mr. Dunnigan answered:

I would say it's both and, ultimately, it would come to the DNR and we would take

whatever measures we needed to take to make sure that we were in compliance.228

Mr. Dunnigan also testified that: "The State will do what is necessary to achieve Compact

compliance."229

223 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, Appendix A, p. 16.

224 Director, Nebraska Department of Natural Resources.

225 Transcript of Arbitration Proceedings, March 13, 2009, Volume V at 948:6.

226 Id., at 948:25-949:1.

227 Id., at 949:2-8.

228 Id. at 970:5-8.

53

140. Although Mr. Dunnigan was not appointed as the Director for the Nebraska Department of

Natural Resources ("DNR") until December 9, 2008,230 his statements set forth in Finding

139 that "we [DNR] would take whatever measures we needed to take to make sure that we

were in compliance" and "The State will do what is necessary to achieve Compact

compliance" are presumably accurate statements of Nebraska's intentions when it entered

into the FSS on December 15, 2002. Yet, in the very first year for Water-Short Year

Administration compliance (2006), Nebraska concedes it violated the FSS.231 Similarly, in

the very first normal compliance year (2007), Nebraska concedes it again violated the FSS.2, 32

141. In its attempts to ensure future compliance with the Compact and FSS, Nebraska first relies

on the 20 percent reduction in the average annual groundwater withdrawals within the

URNRD, MRNRD, and LRNRD, compared to the average withdrawals for 1998 through

2006, as described in Finding 137. Assuming the URNRD and MRNRD do not exceed their

average annual withdrawal limitations of 425,000 acre-feet and 247,580 acre-feet,

respectively, and assuming that the additional reductions in the allotments for individual

irrigators in the LRNRD results in a 20 percent reduction in LRNRD's average annual

groundwater withdrawal as compared to its average withdrawals for 1998 through 2006,

resulting in a reduced average annual LRNRD withdrawal of 193,830 acre-feet, the average

annual groundwater withdrawals in the NRDs for the period 2008 through 2012 will not total

more than 866,410 acre-feet per year, a reduction of 217,120 acre-feet from the 1998 - 2006

average of 1,083,530 acre-feet per year.233 For comparison, this amount of reduction in

average annual groundwater withdrawals is 35 percent of the average annual reduction of

619,000 acre-feet per year that Kansas estimates would result from its proposed remedy.234

142. Nebraska's experts simulated the performance of the IMPs, assuming 20 percent reductions

in the average annual groundwater withdrawals within the URNRD, MRNRD, and LRNRD,

compared to the average withdrawals for 1998 through 2006, under "average climatic

conditions" using the RRCA Groundwater Model and the Accounting Procedures.235 The

results from these simulations showed that Nebraska would be in compliance under normal

229 Id. at 980:15-16.

230 Id. at 946:22-24.

231 Nebraska Exhibit 8, Expert Report of Marc Groff, Tom Riley, and David Kracman, Review of the 20 January

2009 Report Prepared by Spronk Water Engineers, Inc for the State of Kansas, February 17, 2009, Table 2-2, p. 5.

232 State of Nebraska's Post-Hearing Brief at 4 (row in table for average 2003 - 2007).

233 See Finding 137.

234 See Finding 129.

235 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, p. 7.

54

year administration and under its allocation by an average amount of 18,950 acre-feet per

year over the 5-year simulation period.236

143. However, it is not during "average climatic conditions" that compliance with the Compact

and FSS are the most challenging for Nebraska and the Republican River NRDs. Rather, it is

during dry-year conditions that compliance with the Compact and FSS will be the most

difficult, and as correctly noted by Kansas' expert, Mr. David Barfield, it is under those

conditions in particular "when the Compact needs to work."237

144. Nebraska's experts also simulated the performance of the IMPs, assuming 20 percent

reductions in the average annual groundwater withdrawals within the URNRD, MRNRD,

and LRNRD, compared to the average withdrawals for 1998 through 2006, under an

"exceptionally (arguably unrealistic) scenario of repeated dry conditions" using the RRCA

Groundwater Model and the Accounting Procedures.235 The results from these simulations

showed that Nebraska would be over its allocation under normal year administration by an

average amount of 340 acre-feet per year over the 5-year simulation period238 and would be

over by 8,288 acre-feet per year under Water-Short Year Administration.239 However,

Nebraska's basin-wide allocation from these simulations averaged 231,360 acre-feet per year

over the 5-year simulation period,238 which is 20,000 acre-feet per year more than the

average basin-wide allocation of about 211,000 acre-feet per year that was determined by the

RRCA for the actual dry-year period of 2002 through 2006.240 Similarly, Nebraska's

allocation above Guide Rock from these simulations for Water-Short Year Administration

averaged 221,680 acre-feet per year over the 5-year simulation period,239 which is nearly

32,000 acre-feet per year more than the actual average allocation above Guide Rock of

189,820 acre-feet per year that was determined by the RRCA for the Water-Short Year

Administration in 2005 and 2006.241 These computed allocations that are larger than the

actual allocations for 2002 through 2006 likely primarily result from Nebraska's experts

using the average streamflows for the years 2000 through 2005, which totaled 195,250 acrefeet,

242 as compared to the actual average streamflows for 2002 through 2006, which were

236 Id., Appendix B to Appendix E, Table 3C.

237 Transcript of Arbitration Proceedings, March 16, 2009, Volume VI at 1049:15-16.

238 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, Appendix B to Appendix G, Table 3C.

239 Id., Table 5C.

240 Kansas Exhibit 2, Expert Report of Dale E. Book, Requirements for Nebraska's Compliance with the Republican

River Compact, January 20, 2009, Table 1.

241 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Attachment 1.

242 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, Appendix G, Table D, p. 4 (Total of entries in column titled "Dry conditions").

55

reported to total approximately 126,000 acre-feet per year.243 Consequently, Nebraska has

underestimated the amounts by which it is likely to exceed its allocations during dry-year

conditions by perhaps as much as 20,000 acre-feet to 30,000 acre-feet per year. As a result,

the 20 percent reductions in the average annual groundwater withdrawals within the

URNRD, MRNRD, and LRNRD, compared to the average withdrawals for 1998 through

2006, are likely inadequate to ensure compact compliance during prolonged dry-year

conditions, such as occurred from 2002 through 2006.

145. When a 20 percent reduction in the average annual groundwater withdrawals within the

URNRD, MRNRD, and LRNRD, compared to the average withdrawals for 1998 through

2006, is not sufficient to achieve compliance with the Compact and FSS, Nebraska then

relies on the provisions in the IMPs that limit the net groundwater depletions for the

URNRD, MRNRD, and LRNRD to 44 percent, 30 percent, and 26 percent, respectively, of

Nebraska's allowable groundwater CBCU determined from using the RRCA Groundwater

Model, as described in Finding 137. The difficulty in ensuring compliance with the Compact

and FSS through these provisions of the IMPs is what is termed the "lag effect." That is, just

as for groundwater withdrawals, where "there is [a] long time lag between the time when the

pumping actually occurs and the time when it manifests itself on streamflows,"244 depending

on the location of the wells from which consumptive groundwater withdrawals are made,

there is also a long time lag between the time when groundwater withdrawals are reduced or

curtailed and the time when resulting increases in streamflow occur, again depending on the

location of the wells from which pumping is reduced or ceases. Consequently, when it is

determined that one or more of the URNRD, MRNRD, or LRNRD has exceeded their

portion of Nebraska's allowable groundwater CBCU in the preceding year, as specified in the

respective IMP, and further reductions are made to consumptive groundwater withdrawals in

the respective NRD, it will be years before the effects of those reductions are expressed as

increased streamflow, again depending on the location of the wells from which groundwater

withdrawals are reduced or curtailed. If a particular NRD's exceedance of its portion of

Nebraska's allowable groundwater CBCU occurs during a prolonged period of dry

conditions, such as occurred from 2002 through 2006, it will likely not be possible for

Nebraska to achieve compliance during the term of the current IMPs without focused

curtailment of consumptive groundwater withdrawals in close proximity to surface water

streams, which is not specifically required in any the IMPs for the URNRD, MRNRD, or

LRNRD. As a result, the limitations on the average annual net streamflow depletions from

consumptive groundwater withdrawals within the URNRD, MRNRD, and LRNRD are likely

inadequate to ensure compliance with the Compact and FSS during prolonged dry-year

conditions, such as occurred from 2002 through 2006.

146. Given Kansas' concerns that the IMPs for the NRDs are inadequate, Nebraska points out that

in 2007 and 2008, Nebraska remained under its allocations by 30,000 acre-feet and 78,000

acre-feet, respectively.245 The years 2007 and 2008, however, were wet years with the

243 Transcript of Arbitration Proceedings, March 16, 2009, Volume VI at 1039:22-23 (Barfield).

244 Transcript of Arbitration Proceedings, March 16, 2009, Volume VI at 1006:13-15 (Larson).

245 State of Nebraska's Post-Hearing Brief at 3.

56

probability of non-exceedance for precipitation being 0.91 and 0.76, respectively,246 and

there were more than adequate surface water supplies. Because of the increased availability

of surface water supplies in 2007 and 2008, Nebraska's Republican River allocations of

243,400 acre-feet and 332,400 acre-feet, respectively,247 were the largest since accounting

pursuant to the FSS was implemented.248 This masks Nebraska's problem in complying with

the Compact and FSS, which is groundwater CBCU, not surface water CBCU. Groundwater

CBCU is by far the largest portion of Nebraska's total CBCU.249 During dry-year conditions,

such as occurred during 2002 through 2006, surface water CBCU varied, but groundwater

CBCU did not vary significantly.250 The provisions in the IMPs that if the 20 percent

reductions in the average annual groundwater withdrawals within the URNRD, MRNRD,

and LRNRD do not achieve compliance with the Compact and FSS, then the net groundwater

depletions within the NRDs will be further reduced to the NRDs respective portions of

Nebraska's allowable groundwater CBCU are not likely sufficient to achieve compliance

with the Compact and FSS during prolonged dry-year conditions for the reasons set forth in

the Finding 145.

147. Aside from seeking changes to the Accounting Procedures and seeking credit for any

damages paid in calculating moving averages of its allocations less CBCU reduced by IWS,

Nebraska and the Republican River NRDs intend to offset exceedances of Nebraska's future

allocations with plans to continue clearing invasive riparian vegetation along the Republican

River and its tributaries, plans to continue participation in incentive programs to retire

irrigated acreage, and plans to implement streamflow augmentation projects.251 However,

the benefits from these plans remain largely unquantified.

148. The primary means that Nebraska and the Republican River NRDs have available to offset

exceedances of Nebraska's future allocations is the leasing of surface water supplies for

conveyance to Kansas, which one of Nebraska's experts referred to as "the lowest hanging

fruit on the tree."252 Although the Nebraska DNR and NRDs successfully leased 25,000

acre-feet, 53,500 acre-feet, and 15,000 acre-feet of surface water in 2006, 2007, and 2008,

246 Kansas Exhibit 6, Expert Report of David W. Barfield, Ensuring Future Compliance by Nebraska, January 20,

2009, Figure 7.

247 State of Nebraska's Post-Hearing Brief at 4.

248 Kansas Exhibit 1, Expert Report of Dale Book, Engineering Analysis of Losses to Kansas Water Users Resulting

from Overuse of Republican River Supply in Nebraska, January 20, 2009, Table 1.

249 Id.

250 Id.

251 Nebraska Exhibit 15, Expert Report of James Schneider and James Williams, Nebraska Compact Compliance,

February 17, 2009, pp. 10-15.

252 Transcript of Arbitration Proceedings, March 12, 2009, Volume IV at 794:8 (Williams).

57

respectively, there is no evidence in the record that similar quantities of surface water could

be leased during a prolonged dry period, such as occurred from 2002 through 2006. The

probability of non-exceedance over the period of record (1918 - 2007) for precipitation in the

Nebraska portion of the Republican River Basin during 2006, 2007, and 2008 was 0.63, 0.91,

and 0.76, respectively,234 which undoubtedly resulted in more surface water being available

for lease than would be available during a prolonged dry period, particularly when the lessor

can use groundwater as a substitute supply such as occurred in the Nebraska Bostwick

Irrigation District during 2006.253

149. If Nebraska and the Republican River NRDs are going to rely on leasing surface water for

conveyance to Kansas to offset exceedances of its future allocations and reduce future

violations of the Compact and the FSS, then Nebraska and the Republican River NRDs

should have permanent, interruptible supply contracts with surface water irrigators that

subject to the call of Nebraska and the Republican River NRDs would provide certain

amounts of surface water, if available. However, there apparently are no efforts underway to

put in place such permanent, interruptible supply contracts.254

150. Because Nebraska has underestimated the amounts by which it is likely to exceed its

allocations during dry-year conditions by perhaps as much as 20,000 acre-feet to 30,000

acre-feet per year,255 the current IMPs adopted by Nebraska and the Republican River NRDs

are inadequate to ensure compliance with the Compact and FSS during prolonged dry-year

conditions, such as occurred from 2002 through 2006. Nebraska and the Republican River

NRDs should make further reductions in consumptive groundwater withdrawals beyond

what's required in the current IMPs, in addition to obtaining permanent, interruptible supply

contracts with surface water irrigators, to ensure compliance with the Compact and FSS

during prolonged dry-year conditions.

151. Neither the Compact nor the FSS require that Nebraska demonstrate in advance how it will

be in compliance in the future. Nonetheless, Nebraska must maintain compliance as

prescribed by the FSS during each 5-year period for normal administration and during each

2-year period for Water-Short Year Administration. While the Nebraska official responsible

for ensuring compliance with the Compact and the FSS clearly understands non-compliance

is not an option,256 it is not clear that this same understanding exists within the NRDs. For

example, in early 2007, the general manager for the MRNRD stated:

As NRDs, we struggle in trying to help others understand that we have been active in the

basin and that given time, our controls will have a positive benefit.

...

253 See Finding 85.

254 Transcript of Arbitration Proceedings, March 13, 2009, Volume V at 963:11-18 (Dunnigan).

255 See Finding 144.

256 See Finding 139.

58

We are concerned on two points: 1) That the formula being used to measure water

allocations for this lawsuit settlement are flawed and are not giving Nebraska irrigators

appropriate credit for groundwater savings; and, 2) That the Nebraska DNR does not

really know what needs to be done in order to bring Nebraska into compliance. We

hesitate to subject the irrigators in the Republican Basin to such drastic reductions - and

the entire region to such economic hardship - based on a guess or an assumption that may

not be accurate or true.257

The fact is Nebraska has not been in compliance with the FSS since it was executed on

December 15, 2002, until the 5-year normal administration period ending in 2008,247

following the wet year of 2007 with wet-year conditions continuing through 2008, as

described in Finding 146.

152. Even if Kansas' experts have not overestimated the amount of reduction in groundwater

irrigated acreage that is necessary in Nebraska for Nebraska to comply with the FSS as

described in Finding 135, it is not necessary to impose Kansas' proposed remedy to ensure

that Nebraska complies with the Compact and FSS in the future.

153. To ensure Nebraska's future compliance with the provisions of the FSS, Kansas is entitled to

injunctive relief enjoining Nebraska from exceeding its future allocations determined in

accordance with the Accounting Procedures using the averaging provisions for normal

administration and Water-Short Year Administration as set forth in the FSS.

154. Should Nebraska fail to comply with the injunction contemplated by Finding 153, sanctions

may be appropriate in addition to the award of additional damages to Kansas. While such

sanctions may be significant, those sanctions should be based on the specific circumstances

of Nebraska's failure to comply, and hence it is not appropriate to recommend the preestablishment

of such sanctions in advance, as requested by Kansas.258

155. Contrary to the viewpoint expressed by one of Nebraska's experts,259 the FSS does not

provide that money can be exchanged for water in determining the 5-year averages of

allocation less CBCU reduced by the IWS credit for normal administration periods or the

2-year averages for Water-Short Year Administration. Consistent with the express

provisions of the FSS and as a sanction for violating the FSS by exceeding its allocations

during Water-Short Year Administration in 2005 and 2006, Nebraska should not receive

credit in subsequent 5-year averages for damages that may be paid to Kansas for those

violations.

257 Kansas Exhibit 61, An Open Letter To All Concerned About Nebraska Water Issues, pp. 2, 3.

258 Kansas Exhibit 6, Expert Report of David W. Barfield, Ensuring Future Compliance by Nebraska, January 20,

2009, § III.b.vi.; Kansas' Post-Trial Brief at 38.

259 Transcript of Arbitration Proceedings, March 12, 2009, Volume IV at 795:12-16 (Williams).

59

156. In addition to its proposed remedy, Kansas also seeks the appointment of a river master to

administer future compliance with the FSS "on an annual basis until such time as Nebraska

can demonstrate an independent ability to achieve compliance."260 Acknowledging that the

"Court rarely appoints a river master,"261 Kansas cites three reasons why it believes the Court

should appoint a river master: (1) Nebraska does not have a central authority or institutions

that are capable of curtailing excessive consumptive groundwater withdrawals in Nebraska's

portion of the Republican River Basin to achieve compliance with the FSS in the short

term;262 (2) there is no incentive for Nebraska to comply with the FSS, since Nebraska's

gain from noncompliance with the FSS is considerably greater than Kansas' losses; and (3)

there is a natural propensity for the states to disagree.

157. While Nebraska does not have a central authority that regulates groundwater withdrawals and

although the Nebraska NRDs may not embrace the reductions in groundwater CBCU that

may be necessary for compliance with the Compact and FSS during prolonged dry-year

conditions, there is a central authority that can impose the necessary actions to ensure

compliance: the State of Nebraska itself. The Nebraska NRDs operate pursuant to statutes

enacted by the Nebraska legislature, and the Nebraska legislature can change those statutes to

ensure that Nebraska complies with the Compact and FSS. As the director of the Nebraska

DNR testified: "The State of the [sic] Nebraska has to live within its allocation."263 With the

injunctive relief suggested in Finding 153 enjoining Nebraska from exceeding its allocations

in the future and sanctions for failure to comply, the cost to Nebraska for noncompliance

should incentivize Nebraska to take whatever steps are necessary to ensure that it does stay

within its allocations under the Compact pursuant to the FSS during all conditions including

prolonged dry-year conditions.

158. Kansas cites to Texas v. New Mexico264 as a precedent for the Court appointing a river

master. In that case, as is the setting here, the Court recognized "the natural propensity of

these two States to disagree."265 But that was not the reason why the Special Master in that

case made the recommendation, which the Court accepted, that a river master be appointed.

In Texas v. New Mexico, the Court specifically noted the Special Master's recommendation

as follows:

... that because applying the approved apportionment formula is not entirely mechanical

and involves a degree of judgment, an additional enforcement mechanism be supplied.

260 Kansas Exhibit 6, Expert Report of David W. Barfield, Ensuring Future Compliance by Nebraska, January 20,

2009, p. 1.

261 Kansas' Post-Trial Brief at 35.

262 Id.

263 Transcript of Arbitration Proceedings, March 13, 2009, Volume V at 954:7-8 (Dunnigan).

264 Texas v. New Mexico, No.65, Original, 482 U.S. 124, 107 S.Ct. 2279.

265 Id. at 134.

60

We accept his recommendation and also his preferred solution: the appointment of a

River Master to make the required periodic calculations.266

In this matter, a river master is not needed "to make the required periodic calculations"

because pursuant to the FSS:

The States will determine Virgin Water Supply, Computed Water Supply, Allocations,

Imported Water Supply Credit, augmentation credit and Computed Beneficial

Consumptive Use based on a methodology set forth in the RRCA Accounting

Procedures, attached hereto as Appendix C.267

159. In Texas v. New Mexico, the river master appointed by the Court had the specific and limited

duty "to make the required periodic calculations" in applying the approved apportionment

formula. In this matter, Kansas has not identified what specific duties and authorities a

Court-appointed river master could or should undertake. Kansas has only proposed the

general duty "to administer Decree compliance on an annual basis"268 Until such time as the

duties and authorities of a river master for the Republican River Basin are specifically

identified, appointment of a river master is not warranted.

CONCLUSIONS

Accounting Procedures

1. For the reasons set forth in the Arbitrator's Final Decision on Legal Issues, which is attached

and incorporated herein, Nebraska's proposed changes to the Accounting Procedures are

proper subjects for this arbitration.

Accounting Procedures - Estimating Computed Beneficial Consumptive Use for Groundwater

and Imported Water Supply

2. The assertion made by Colorado and Kansas that the issue of estimating CBCU of

groundwater and determining the IWS is not a proper subject for this arbitration, because

Nebraska's expert report on this issue had not been submitted to the RRCA for its

consideration, is not convincing. Nebraska's proposal to use 8 differences calculated using

16 runs of the RRCA Groundwater Model for each of 4 aquifer stresses is essentially the

same as what was presented to the RRCA in August of 2008, even though the weighting

coefficients used to combine the differences have changed. Neither Colorado nor Kansas

timely made this assertion when they submitted their respective expert reports in response to

266 Id.

267 Final Settlement Stipulation, Volume 1 of 5, § IV.A., p. 17.

268 Kansas Exhibit 6, Expert Report of David W. Barfield, Ensuring Future Compliance by Nebraska, January 20,

2009, § IV.3.

61

Nebraska's expert report on this issue, and neither timely raised this assertion during the

hearing conducted as part of this arbitration.

3. Nebraska's proposed procedure for determining VWS, whereby what Nebraska terms VWSG,

determined as (θ - CKMN), is more consistent with the definition of VWS established in the

Compact and adopted in the Accounting Procedures than is summing CBCUC, CBCUK, and

CBCUN, less IWS, each calculated in accordance with the existing Accounting Procedures, to

compute VWSG.

4. While Nebraska's proposal for determining what it terms VWSG is consistent with the

definition of VWS established in the Compact and adopted in the Accounting Procedures,

Nebraska's proposed changes to calculate CBCUC, CBCUK, CBCUN, and IWS, are

problematic and adoption of Nebraska's proposed changes by the RRCA is not appropriate.

5. Although Nebraska's proposed changes to calculate CBCUC, CBCUK, CBCUN, and IWS,

should not be adopted by the RRCA, the RRCA should consider reconvening the Technical

Groundwater Modeling Committee to thoroughly re-evaluate the nonlinear response of the

RRCA Groundwater Model when simulated stream drying occurs, re-evaluate the existing

procedures for determining CBCU and IWS, and document its conclusions and any

recommendations in a report to the RRCA.

Accounting Procedures - Haigler Canal

6. During the period of years from 1995 through 2006, the annual amounts of water measured at

the Haigler Canal Spillback gage exceeded the actual annual amounts of water measured at

the Arikaree Gage in 2002, 2003, 2004, and 2005, indicating that a significant portion of the

water measured at the Haigler Canal Spillback gage during these years does not remain in the

Arikaree River as measurable surface water at the Arikaree Gage.

7. While some of the water measured at the Haigler Canal Spillback gage undoubtedly reaches

the Arikaree Gage under certain conditions, there is insufficient information to justify

changing the Accounting Procedures to reduce the diversions from the North Fork

Republican River into the Haigler Canal by the amount of water measured at the Haigler

Canal Spillback gage, as proposed by Nebraska.

8. Consequently, the changes to the Accounting Procedures proposed by Nebraska involving

VWS calculations for the North Fork of the Republican River in Colorado and the Arikaree

River are not justified.

9. During the period of years from 1995 through 2006, the annual amounts of water returning to

the Arikaree River from irrigation using water from the Haigler Canal, as estimated in

accordance with the change to the Accounting Procedures proposed by Nebraska to apportion

49 percent of the return flows to the Arikaree River at the Arikaree Gage, exceeded the actual

annual amounts of water measured at the Arikaree Gage in 2001, 2002, 2003, and 2004.

62

Thus, only a small portion of the return flow from irrigation in Nebraska using water from

the Haigler Canal returns to the Arikaree River, at least during the years since 2001.

10. The conclusion that since 2001 only a small portion of the return flow from irrigation in

Nebraska using water from the Haigler Canal returns to the Arikaree River is supported by

the observations that: (1) the lands irrigated with water from the Haigler Canal in the

Arikaree drainage near Haigler are sandy; (2) many of the systems used to irrigate lands in

Arikaree drainage near Haigler using water from the Hailger Canal have been converted to

center pivot sprinklers reducing return flows comprised by overland flow; and (3) the

direction of groundwater flow under the Arikaree drainage is north towards the Main Stem,

not towards the Arikaree River.

11. While some of the water measured at the Arikaree Gage may be comprised of return flow

from groundwater discharge under certain conditions, there is insufficient information to

justify changing the Accounting Procedures to apportion any of the return flow from

irrigating lands using water from the Haigler Canal to the Arikaree River, as proposed by

Nebraska.

Accounting Procedures - Groundwater Model Accounting Points

12. The "equitable division" or "allocation" of the waters of the Republican River Basin set forth

in Article IV of the Compact for a named "drainage basin" is derived from the "computed

average annual virgin water supply" originating in that drainage basin, which ends at the

confluence of the stream draining that basin and the "Main Stem" of the Republican River as

"Main Stem" is defined in § II. of the Accounting Procedures. This definition of Main Stem

is entirely consistent with Article III of the Compact.

13. The locations of the accounting points in the RRCA Groundwater Model that are used for

calculating CBCU of groundwater for the "Frenchman Creek (River) drainage basin in

Nebraska," "South Fork of the Republican River drainage basin," and "Driftwood Creek

drainage basin," pursuant to § III.D.1. of the Accounting Procedures, are consistent with the

allocations made by named drainage basin in Article IV of the Compact.

14. Changing the locations of the accounting points in the RRCA Groundwater Model that are

used to determine CBCU of groundwater as proposed by Nebraska for the "Frenchman Creek

(River) drainage basin in Nebraska," "South Fork of the Republican River drainage basin,"

and "Driftwood Creek drainage basin," such that the accounting point locations would

correspond to the locations of the stream gages designated in § II. of the Accounting

Procedures, would result in the CBCU of groundwater below the designated stream gages

being included in the CBCU for the Main Stem rather than in the CBCU for the tributary

drainage basins. These changes would be inconsistent with the definitions of these drainage

basins implicit in Article III of the Compact and are not appropriate.

15. However, to the extent groundwater pumping causes depletions to streamflows downstream

of the gages designated in § II. of the Accounting Procedures for the "Frenchman Creek

63

(River) drainage basin in Nebraska," "South Fork of the Republican River drainage basin,"

and "Driftwood Creek drainage basin," and upstream of the confluence of each associated

stream with the Main Stem, the RRCA should modify the Accounting Procedures for these

sub-basins to subtract the CBCU of groundwater below the designated gage for each Subbasin

and above the confluence of that Sub-basin's stream with the Main Stem from the

VWS for that Sub-basin, to avoid a double-accounting of that quantity of water, and add that

increment of groundwater CBCU in the VWS for the Main Stem, such as is currently done in

accounting for the CBCU of surface water below the Sub-basin gages for Medicine Creek,

Sappa Creek, Beaver Creek, and Prairie Dog Creek.

16. The accounting point currently used to determine the CBCU of groundwater in the "North

Fork of the Republican River in Colorado drainage basin" is not located at the confluence

with the Main Stem, as the Main Stem is defined in § II. of the Accounting Procedures. This

is inconsistent with the explicit meaning of the "North Fork of the Republican River drainage

basin in Colorado" in Article III of the Compact and results in CBCU of groundwater that

should be included in the CBCU for the Main Stem being included instead in the CBCU for

the "North Fork of the Republican River in Colorado drainage basin." The RRCA should

move the location of this accounting point to the model cell in which the North Fork of the

Republican River crosses the Colorado-Nebraska state line to provide for the appropriate

determination of CBCU for the "North Fork of the Republican River in Colorado drainage

basin" and CBCU for the Main Stem.

17. The changes to the Accounting Procedures described above should apply to all years for

which the accounting of water use has not been finalized and approved by the RRCA.

Damages - Losses to Kansas Water Users from Overuse in Nebraska

18. Nebraska does not deny that it exceeded its basin-wide allocations in 2005 and 2006 and its

Water-Short Year allocations above Guide Rock in 2005 and 2006.

19. Subsection V.B.2.e. of the FSS explicitly provides that for purposes of determining

Nebraska's compliance during Water-Short Year Administration, Virgin Water Supply,

Computed Water Supply, Allocations, and Nebraska's Computed Beneficial Consumptive

Use, are to be calculated as two-year running averages. The FSS does not explicitly address

the amount of the violation when Nebraska is not in compliance with the FSS during Water-

Short Year Administration.

20. The two-year average of Nebraska's exceedance of its Water-Short Year Administration

allocation above Guide Rock for 2006 should not be used to determine the amount of

Nebraska's violation for 2006 because the two-year average is greater than Nebraska's actual

exceedance in 2006. Rather, the amount of Nebraska's violation for 2005 and 2006 should

be equal to Nebraska's exceedance of its Water-Short Year Administration allocations above

Guide Rock for each of those years.

64

21. Based on a document accepted as Kansas Exhibit 84 on the last day of hearing, irrigators in

the Nebraska Bostwick Irrigation District chose to substitute water supply from Nebraska's

allocation below Guide Rock for water supply from the Superior Canal in 2006. Given the

explicit provision in § IV.A.e)(1) of the Accounting Procedures pertaining to use of

substitute supplies for the Superior Canal from Nebraska's allocation below Guide Rock, a

portion of the 2006 evaporation from Harlan County Lake should be assigned to Nebraska.

22. Adding half of the net evaporation from Harlan County Lake for 2006 to Nebraska's estimate

of its 2006 allocation exceedance results in a revised estimate of the 2006 exceedance that is

sufficiently close to Kansas' estimate of the 2006 exceedance to justify acceptance of

Kansas' estimate, which allocated evaporation from Harlan County Lake "... based on longterm

average uses."

23. Nebraska's exceedance of its Water-Short Year Administration allocation above Guide Rock

is estimated to be 42,860 acre-feet for 2005 and 36,100 acre-feet for 2006, which are the

amounts estimated by Kansas' expert.

24. To provide a basis for estimating the direct economic impacts to Kansas caused by

Nebraska's exceedance of its Water-Short Year allocation above Guide Rock, the additional

amount of water that should have been available for use in Kansas was routed in accounting

simulations by the experts for Kansas and Nebraska to where the direct economic of impacts

of the shortages occurred: the farm headgates in KBID and downstream of KBID. To

perform these simulations the experts for both Kansas and Nebraska assumed that the

additional amount of water that should have been available for use in Kansas was regulated

through Harlan County Lake. After deducting for additional net evaporation from Harlan

County Lake, the additional amounts of water that should have been available from Harlan

County Lake were estimated to be 41,519 acre-feet for 2005 and 33,383 acre-feet, the

amounts estimated by Kansas' expert.

25. The accounting simulations routing the additional water from Harlan County Lake performed

by Kansas' expert results in estimated amounts of water that would have been available for

delivery to KBID from the Courtland Canal at the Nebraska-Kansas state line of 40,551 acrefeet

(rounded to 40,600 acre-feet) for 2005 and 32,605 acre-feet (rounded to 32,600 acrefeet)

for 2006. These estimated amounts are overstated. Kansas' expert only subtracted the

consumptive canal losses (losses that do not recharge computed as 18 percent of the total

canal losses in accordance with RRCA accounting) from the Courtland Canal diversions in

Nebraska, leaving the non-consumptive losses (losses that do recharge computed as 82

percent of the total canal losses in accordance with RRCA accounting) as part of the

simulated additional supplies available to KBID from the Courtland Canal at the Nebraska-

Kansas state line in 2005 and 2006. While some, if not all, of the non-consumptive losses

from the Courtland Canal in Nebraska would reasonably be assumed to be available to

Kansas irrigators as groundwater and as additional flow in the Republican River, the nonconsumptive

canal losses are losses from the canal and can not be part of the water supply

available to KBID from the Courtland Canal at the Nebraska-Kansas state line.

65

26. There is insufficient information in the record to allow a reasonably reliable estimate of how

the additional groundwater and flow in the Republican River from non-consumptive losses

from the Courtland Canal in Nebraska might have been used by irrigators in Kansas.

27. The accounting simulations routing the additional water from Harlan County Lake performed

by Nebraska's experts properly exclude all of the estimated canal losses from the Courtland

Canal in Nebraska. However, Nebraska's experts made no attempt to estimate the amounts

of canal losses that would have been available to Kansas as groundwater or as additional

flow in the Republican River. Nebraska's experts have understated the additional amounts of

water that would have available to Kansas irrigators below the Nebraska-Kansas state line in

2005 and 2006.

Damages - Direct Economic Impacts

28. The approach used by Kansas' experts to project irrigated crop yields that would have been

realized, had overuse of water by Nebraska not occurred, is not materially the same as the

approach used in Kansas v. Colorado, No. 105, Original, in several respects that are

important. First, the crop response functions in Kansas v. Colorado were based on the

response of crop yield to precipitation and irrigation only, whereas the version of IPYsim

employed by Kansas' experts includes not only crop-yield response to precipitation and

irrigation but also includes crop-yield response to total usable nitrogen. Second, the crop

response functions in Kansas v. Colorado do not include economic considerations, whereas

IPYsim incorporates costs for both nitrogen fertilizer and water. Third, Kansas' experts

adjusted the IPYsim response functions first so that the economically optimal yields equaled

trend yields and then secondly so that yields for fully irrigated crops (termed fully irrigated

"expected yield" for an individual crop) equaled observed yields under actual irrigation

multiplied by the ratios of simulated yield under full irrigation and simulated yield under

actual irrigation, both simulated when the economically optimal yields equaled trend yields.

This resulted in the fully irrigated "expected yield" for corn, which Kansas' experts identified

as the most appropriate crop for their proposed yield modeling framework, of 206

bushel/acre. This fully irrigated "expected yield" is 10 percent higher than the historical

maximum yield of 187 bushel/acre in KBID, which was observed in 2005. Kansas did not

provide any information to verify the reasonableness of the resulting response functions that

were then used to assess impacts, whereas the crop response functions in Kansas v. Colorado

were based on empirical relationships; that is, relationships based on observations that can

be verified or disproved by observation or experiment.

29. The experts for Colorado and Nebraska on the issue of economic impacts were both critical

of the adjustment of the IPYsim crop response functions to estimate the crop-specific fully

irrigated "expected yield."

30. Kansas did not sufficiently address variations in soil types and climate between western

Kansas, where the crop-yield functions for precipitation and irrigation were developed and

upon which the IPYsim crop response functions were based, and north-central Kansas

66

several hundred miles to the northeast, where KBID and the other impacted areas in Kansas

are located.

31. There is no evidence in the record of an active water market in or adjacent to south-central

Nebraska, where Nebraska leased surface water in 2006 that could be diverted by KBID at

the Guide Rock Diversion Dam. Therefore, the unit cost that Nebraska paid to lease water in

its attempt to comply with the FSS in 2006 is not the same as the unit value of water to

Kansas from lost profits due to overuse by Nebraska in 2006.

32. In seeking damages, Kansas bears the burden of proof concerning the extent of such damages

based upon a preponderance of the evidence269, 270 and must show such damages to

reasonable certainty.271

33. The preponderance of evidence at this juncture does not support the estimates of additional

water that would have been available at the headgates of Kansas irrigators but for Nebraska's

overuse of water in 2005 and 2006, the lack of significance of soil and climate variations

assumed by Kansas' experts, the methodology used by Kansas's experts to project irrigated

crop yields that would have been realized had overuse of water by Nebraska not occurred, or

the estimates of the total direct economic impacts in 2005 and 2006 made by Kansas' experts

with reasonable certainty. Kansas's estimates of the total direct economic impacts in 2005

and 2006 are not sufficiently reliable to form an appropriate recommendation for awarding

damages to Kansas.

34. The alternative estimates of total direct economic impacts in 2005 and 2006 developed by

experts for Colorado and Nebraska are also not sufficiently reliable to form an appropriate

recommendation for awarding damages to Kansas.

35. Because this arbitration is non-binding, the legal principle res judicata is not applicable and

Kansas may submit additional information to support or revise its estimates of actual

damages caused by Nebraska's overuse of water in 2005 and 2006. Such additional

information can be presented in arbitration supplemental to this present proceeding, before

the same or a different arbitrator, or such information can be presented during a

determination of damages by the Court.

269 "In a typical civil suit for money damages, plaintiffs must prove their case by a preponderance of the evidence."

Herman & MacLean v. Huddleston, 459 U.S. 375, 103 S.Ct. 683 (1983), at 387.

270 "The burden of showing something by a ‘preponderance of the evidence,' the most common standard in the civil

law, ‘simply requires the trier of fact to believe that the existence of a fact is more probable than its nonexistence

before [he] may find in favor of the party who has the burden to persuade the [judge] of the fact's existence.'"

Concrete Pipe & Products of California, In. v. Construction Laborers Pension Trust for Southern California, 508

U.S. 602, 113 S.Ct. 2264, at 2279 (internal citations omitted).

271 "It is well understood that such evidence must show damages to reasonable certainty. Mere ‘plausible

anticipation' does not merit consideration nor are flights into the realm of pure speculation entitled to be treated as

evidence. Connecticut RY. & Lighting Co. v. Palmer et al., 305 U.S. 493, 59 S.Ct. 316 (1939), at 505.

67

36. Clearly Kansas incurred damages resulting from Nebraska's overuse of water in 2005 and

2006 and those damages may well be in the range of one to several million dollars.

However, until such time Kansas can demonstrate with a preponderance of evidence that its

assumptions and methodology for estimating lost profits and establishing damages is

reasonably reliable (either through independent peer review or with empirical data), during

subsequent arbitration or before the Court, only an award of nominal damages should be

made.

37. Nominal damages are "by definition, minimal monetary damages."272 While nominal

damages could be $ 1 or less,273 given that Kansas has clearly been harmed by Nebraska's

overuse of water but has not shown the extent of such harm with sufficient certainty, an

award of nominal damages in the amount of $10,000 is recommended.

Damages - Indirect Economic Impacts

38. The gross indirect economic impacts, or "Value Added Impact" or "Indirect Value Added

Loss" estimated by Kansas' experts for both 2005 and 2006 of 44 percent of the direct

economic impacts (gross income loss), meaning that total economic impacts are estimated to

be 1.44 times the estimated direct economic impacts, are reasonable.

39. Kansas' experts should have attempted to reasonably quantify the indirect benefits resulting

from Nebraska's payments for actual damages. Also, there is no evidence in the record for

this proceeding whether opportunity costs offsetting or reducing gross secondary impacts, as

found to be appropriate by the Court in Kansas v. Colorado, No. 105, Original, were

considered by Kansas' experts, or whether such offsets are even relevant in this instance.

40. Since an award of only nominal damages for direct economic impacts is recommended in this

proceeding, no award of damages for indirect economic impacts should be made.

41. If Kansas seeks to demonstrate with a preponderance of evidence the amounts of additional

water that would have been available at the headgates of Kansas irrigators, but for

Nebraska's overuse of water in 2005 and 2006, and that its assumptions and methodology for

estimating lost profits and establishing actual damages is reasonably reliable during

subsequent arbitration or before the Court, Kansas should also attempt to reasonably quantify

indirect benefits resulting from Nebraska's payment for actual damages and should also

include any offsetting opportunity costs if such are relevant.

Future Compliance

42. To ensure future compliance with the FSS, Kansas has proposed that Nebraska reduce its

groundwater-irrigated acreage in the Basin by approximately 515,000 acres. Kansas' experts

272 22 Am. Jur. 2d Damages § 8 (2008).

273 Colorado Investment Services v. Hager, 685 P.2d 1371 (1984) at 1375.

68

estimate that this would reduce consumptive groundwater withdrawals by an average of

619,000 acre-feet per year.

43. Kansas has adequately demonstrated that its proposed remedy would result in Nebraska's

compliance with the FSS, even during dry-year conditions similar to what occurred during

the period 2002 through 2006. However, given the magnitude of the assumed increase in

surface water CBCU from reductions in groundwater CBCU and the fact that Kansas'

experts used datasets from years when precipitation was above average overall, Kansas'

experts likely have overestimated the amount of reduction in groundwater irrigated acreage

that is necessary in Nebraska for Nebraska to comply with the FSS. Therefore, Kansas has

not adequately demonstrated that its proposed remedy is the "minimum remedy necessary for

compliance" as it has asserted.

44. In its attempts to ensure future compliance with the Compact and FSS, Nebraska and the

URNRD, MRNRD, and LRNRD have jointly developed revised IMPs for the 5-year term

from 2008 through 2012. These revised IMPs first rely on 20 percent reductions in the

average annual groundwater withdrawals within the URNRD, MRNRD, and LRNRD

(intended to be achieved in the LRNRD through reduced allocations for individual

irrigators), compared to the average withdrawals for 1998 through 2006. This would reduce

consumptive groundwater withdrawals within the portion of the Republican River Basin in

Nebraska by an average of 217,120 acre-feet per year from the 1998 - 2006 average of

1,083,530 acre-feet per year. An average reduction in consumptive groundwater withdrawals

of 217,120 acre-feet per year is 35 percent of the average annual reduction of 619,000 acrefeet

per year that Kansas estimates would result from its proposed remedy.

45. Simulations by Nebraska's experts of the performance of the IMPs, assuming 20 percent

reductions in the average annual consumptive groundwater withdrawals within the URNRD,

MRNRD, and LRNRD from the 1998 - 2006 average withdrawals, under a scenario of

repeated dry conditions, during which compliance would be crucial, showed that Nebraska

would be over its allocation under normal year administration by an average amount of

340 acre-feet per year, over the 5-year simulation period, and would be over by an average

amount of 8,288 acre-feet per year under Water-Short Year Administration. However,

Nebraska's basin-wide allocation from these simulations averaged 20,000 acre-feet per year

more than the average basin-wide allocation of about 211,000 acre-feet per year that was

determined by the RRCA for the actual dry-year period of 2002 through 2006, and

Nebraska's allocation above Guide Rock from these simulations for Water-Short Year

Administration averaged 32,000 acre-feet per year more than the actual average allocation

above Guide Rock of 189,820 acre-feet per year that was determined by the RRCA for the

Water-Short Year Administration in 2005 and 2006. Consequently, Nebraska has

underestimated the amounts by which it is likely to exceed its allocations during dry-year

conditions by perhaps as much as 20,000 acre-feet to 30,000 acre-feet per year. As a result,

the 20 percent reductions in the average annual groundwater withdrawals within the

URNRD, MRNRD, and LRNRD, compared to the average withdrawals for 1998 through

2006, are unlikely sufficient to ensure compact compliance during prolonged dry-year

conditions, such as occurred from 2002 through 2006.

69

46. When a 20 percent reduction in the average annual consumptive groundwater withdrawals

within the URNRD, MRNRD, and LRNRD, compared to the 1998 - 2006 average

withdrawals, is not sufficient to achieve compliance with the Compact and FSS, Nebraska

then relies on the provisions in the IMPs that limit the net groundwater depletions for the

URNRD, MRNRD, and LRNRD to 44 percent, 30 percent, and 26 percent, respectively, of

Nebraska's allowable groundwater. The difficulty in ensuring compliance with the Compact

and FSS through these provisions of the IMPs is that just as for groundwater withdrawals

where there is a long time lag between the time when the pumping actually occurs and the

time when it manifests itself on streamflows, depending on the location of the wells from

which consumptive groundwater withdrawals are made, there is also a long time lag between

the time when groundwater withdrawals are reduced or curtailed and the time when resulting

increases in streamflow occur.

47. When it is determined that one or more of the URNRD, MRNRD, or LRNRD has exceeded

their portion of Nebraska's allowable groundwater CBCU in the preceding year, as specified

in the respective IMP, and further reductions are made to consumptive groundwater

withdrawals in the respective NRD, it will be years before the effects of those reductions are

expressed as increased streamflow, depending on the location of the wells from which

groundwater withdrawals are reduced or curtailed. If a particular NRD's exceedance of its

portion of Nebraska's allowable groundwater CBCU occurs during a prolonged period of dry

conditions, such as occurred from 2002 through 2006, it will likely not be possible for

Nebraska to achieve compliance during the term of the current IMPs without focused

curtailment of consumptive groundwater withdrawals in close proximity to surface water

streams, which is not specifically required in any the IMPs for the URNRD, MRNRD, or

LRNRD. As a result, the limitations on the average annual net streamflow depletions from

consumptive groundwater withdrawals within the URNRD, MRNRD, and LRNRD are likely

inadequate to ensure compliance with the Compact and FSS during prolonged dry-year

conditions, such as occurred from 2002 through 2006.

48. Nebraska has not been in compliance with the FSS since it was executed on December 15,

2002, until the 5-year normal administration period ending in 2008, following the wet year of

2007 with wet-year conditions continuing through 2008. Although the IMPs for the

Republican River NRDs are enforceable, the current IMPs adopted by Nebraska and the

Republican River NRDs are inadequate to ensure compliance with the Compact and FSS

during prolonged dry-year conditions, such as occurred from 2002 through 2006. Nebraska

and the Republican River NRDs should make further reductions in consumptive groundwater

withdrawals beyond what's required in the current IMPs, in addition to obtaining permanent,

interruptible supply contracts with surface water irrigators, to ensure compliance with the

Compact and FSS during prolonged dry-year conditions.

49. Neither the , Compact nor the FSS require that Nebraska demonstrate in advance how it will

be in compliance in the future. Nonetheless, Nebraska must maintain compliance as

prescribed by the FSS during each 5-year period for normal administration and during each

2-year period for Water-Short Year Administration. To ensure Nebraska's compliance with

the Compact and FSS into the future, it is not necessary to impose Kansas' proposed remedy.

However, Kansas is entitled to injunctive relief enjoining Nebraska from exceeding its future

70

allocations determined in accordance with the Accounting Procedures using the averaging

provisions for normal administration and Water-Short Year Administration as set forth in the

FSS.

50. Should Nebraska fail to comply with an injunction, sanctions may be appropriate in addition

to the award of additional damages to Kansas. While such sanctions may be significant,

those sanctions should be based on the specific circumstances of Nebraska's failure to

comply, and hence it is not appropriate to recommend the pre-establishment of such

sanctions in advance, as requested by Kansas.

51. Consistent with the express provisions of the FSS, which do not provide that money can be

exchanged for water in determining the 5-year averages of allocation less CBCU reduced by

the IWS credit for normal administration periods or the 2-year averages for Water-Short Year

Administration, and as a sanction for violating the FSS by exceeding its allocations during

Water-Short Year Administration in 2005 and 2006, Nebraska should not receive credit in

subsequent 5-year averages for damages that may be paid to Kansas for those violations.

52. With the injunctive relief enjoining Nebraska from exceeding its allocations in the future and

sanctions for failure to comply, the cost to Nebraska for noncompliance should incentivize

Nebraska to take whatever steps are necessary to ensure that it does stay within its allocations

under the Compact pursuant to the FSS during all conditions including prolonged dry-year

conditions.

53. In Texas v. New Mexico, the Court appointed a river master with the specific and limited duty

"to make the required periodic calculations" in applying the approved apportionment

formula.274 Since the specific duties and authorities that a river master appointed by the

Court could or should undertake in the Republican River Basin have not been specifically

identified, appointment of a river master is not warranted at this time.

274 Texas v. New Mexico, No.65, Original, 482 U.S. 124, 107 S.Ct. 2279, at 134.

71

RECOMMENDATIONS

1. As described in the Arbitrator's Final Decision on Legal Issue, Question 3, the Accounting

Procedures should be modified so that evaporation from Harlan County Lake is allocated

between Kansas and Nebraska in proportion to each state's use of water from Harlan County

Lake for all purposes, including use to offset streamflow depletions from consumptive

groundwater withdrawals.275

2. Nebraska's proposed changes to the Accounting Procedures to calculate CBCUC, CBCUK,

CBCUN, and IWS, should not be adopted. However, the RRCA should consider reconvening

the Technical Groundwater Modeling Committee to thoroughly re-evaluate the nonlinear

response of the RRCA Groundwater Model when simulated stream drying occurs, reevaluate

the existing procedures for determining CBCU and IWS, and document its

conclusions and any recommendations in a report to the RRCA.

3. Nebraska's proposed changes to the Accounting Procedures involving calculation of VWS

for the North Fork of the Republican River in Colorado and the Arikaree River should not be

adopted.

4. Nebraska's proposed changes to the Accounting Procedures to apportion return flows from

irrigation using water diverted through the Haigler Canal between the North Fork of the

Republican River in Nebraska and the Arikaree River should not be adopted.

5. Nebraska's proposed changes to the Accounting Procedures to move the location of the

accounting points in the RRCA Groundwater model to correspond to the location of the Subbasin

gages for "Frenchman Creek (River) drainage basin in Nebraska," "South Fork of the

Republican River drainage basin," and "Driftwood Creek drainage basin," should not be

adopted. However, to the extent groundwater pumping causes depletions to streamflows

downstream of the gages in these sub-basins and upstream of the confluence of each

associated stream with the Main Stem, the Accounting Procedures for these sub-basins

should be modified to subtract the CBCU of groundwater below the designated gage for each

Sub-basin and above the confluence of that Sub-basin's stream with the Main Stem from the

VWS for that Sub-basin, to avoid a double-accounting of that quantity of water, and add that

increment of groundwater CBCU in the VWS for the Main Stem.275

6. Nebraska's proposed change to the Accounting Procedures to move the location of the

accounting point in the RRCA Groundwater model for the "North Fork of the Republican

River in Colorado drainage basin" to the location where the North Fork of the Republican

River crosses the Colorado-Nebraska state line should be adopted.275

7. Kansas should be awarded nominal damages of $10,000 for Nebraska's overuse of water in

2005 and 2006 until Kansas can correct its estimates of the amounts of water that would have

been available to KBID from the Courtland Canal, but for Nebraska's overuse, and can

275 Changes should apply to all years for which the accounting of water use has not been finalized and approved by

the RRCA.

Powered by WorldNow
General and news information or questions:
Across the Midlands Information:
events@nebraska.tv
Comments: comments@nebraska.tv
News:
news@nebraska.tv
Phone: 308-743-2494
News Fax: 308-743-2660
All content © Copyright 2000 - 2014 WorldNow and NTV. All Rights Reserved.
For more information on this site, please read our Privacy Policy and Terms of Service.