Adapting Water Allocation Systems: Challenges and Opportunities...

Adapting Water Allocation Systems: Challenges and Opportunities

The Upper Guadiana and Llobregat Delta cases.

Jesus CarreraGHS UPC-CSIC

IDAEA, CSIC

April, 25th, 2019Adapting Water Rights to Face Climate Change Impacts: A Comparison of California

and SpainThe Botin Foundation and Rosenberg International Forum on Water Policy

Consejo Superior de Investigaciones Científicas

I am going to:

• Briefly summarize the main challenges, and why I am discussing the two cases I will be discussing

• Addressing climate change at the UpperGuadiana Basin

• Addressing seawater intrusion at the Llobregat Delta Aquifer

Rainfall patterns of Spain similar to thoseof California

Guadiana River, now…

… and before

And we face similar challenges:1) water over use (and loss of base flow)

After pumping starts, the river flow rate decreases progressively (by the same amount but with 10 y delay)

(MIMAM, 2000; Custodio, 2017).

And we face similar challenges:2) Agricultural pollution

Nitrate concentrations legally OK

But all our large reservoirs are eutrofized

Map of the trophic state of more 10 hm3 reservoirs (Libro blanco del agua, 2005)

We have a problemwith the law

And we face similar challenges:3) Seawater intrusion in all our Med aquifers

SWI causes not only salinization of costal aquifers, but also the loss of submarine groundwaterdischarge, which affects coastal ecosystems

(IPCC, 2018)

And we face similar challenges:4) Rainfall is dropping due to climate change

The Upper Guadiana case

• Relevant because overpumping for irrigationwas causing– The drying of an important wetland (Tablas de

Daimiel) and only fed by surface water– The loss of the Guadiana River

For proper understanding, we built a «fancy» coupled groundwater surface water model

The only input to the model are wheather variables (rainfall, moisture, temperature, etc.) and pumpingrates and soil use

(Sapriza et al, WRR, 2015)

The model reproduces the fall of heads, and Guadiana River and Daimiel wetland drying…

The some 60 m drop in heads, starting in the 1970’s, therecovery of wet years and

reduction in pumping (Sapriza et al, WRR, 2015)

We addressed the impact of climate changeby looking for changes in circulation paterns

1. Find which GCMs did best during historicalrecords.

2. Calibrate during the historical record therainfall for each circulation pattern

3. Examine the future by simply assuming thatthe GCMs produce reliable circulationpatterns

We had to downscale for accuracy

Response to Climate Impacts: GCM historical (1960-1999 Green) and GCM-RCP85 (2060-2099 Red)

So, we went to the Guadiana River Basin Authority (the President was a classmate)

Response:1) We are forced to use the «legal» climate change

projections 2) We cannot enforce water use3) What we do is to buy (i.e., rent) yearly water rights

(which had been appropriated, contrary to Spanish law)

8-7-20111-3-2005

(1) Past situation of «ojos»

Has it worked? Google Earth images

13-11-201528-5-2015

(2)

(4)

(3) 2011 2012 2013 2014 2015 20166

10

14

18

Prof

undi

dad

(m)

“Ojos”Startflowing

Image 2

Image (1)

04.04.263

GW head recovery

It has dried again (because of drought and economic crisis), ….

But we are close!

Second case: the Llobregat Delta suffered severe seawater intrusion threatening the water supply of Barcelona

Again, we calibrated (1965-2001) and validated (2002-2004) a fancy numerical model

Also for salinity

The state of salinization was bad

1995

And it could get much worse

0

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

22000

After 36 years of pumping

Proposed actions: Reduce pumping and adopt corrective measures to increase resources

Pumping well 1

Pumping well 2

ARTIFICIAL RECHARGE PONDS

2 recharge pondsToral area = 11 haProjected infiltration rate = 0.25 m/d

Total projected recharge =11hm3/y

SEAWATER INTRUSION BARRIER

Divided in 4 sectors+ 2 pumping wells (extract trapped salt)

Total injection rate = 3.65 hm3/y

Recharge pond 1

Recharge pond 2

We posed the problem as an linear programing problem

3

8387.962

3598.126

1912.000

17050.812

14780.677

4334.562

3866.931

10402.932

91976.839

10312.991

153982.228

122613.720

1490316.985

477642.604

477642.604

-25 - -12 hm3/y -12 - -9 hm3/y -9 - -6 hm3/y -6 - -3 hm3/y -3 - -1 hm3/y -1 - 0 hm3/y 0 - 0.01 hm3/y 0.01 - 3 hm3/y

122613. 720

1490316.985

477642.604

477642.6 04

8387.962

3598.126

1912.000

17050.812

14780.677

4334.562

3866.931

10402.932

91976.839

10312.991

153982.228

1

2

3

4

5

122613. 720

1490316.985

477642.604

477642.6 04

8387.962

3598.126

1912.000

17050.812

14780.677

4334.562

3866.931

10402.932

91976.839

10312.991

153982.228

1

2

3

4

5

6

Maximize pumping subject to:1) high heads at the shore, to prevent SWIand 2) fixed artificial recharge rates

Results: unacceptable without corrective actions

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 6

hm3 /y

y

Without corrective measures

8387.962

3598.126

1912.000

17050.812

14780.677

4334.562

3866.931

10402.932

91976.839

10312.991

153982.228

122 613. 720

1490 316. 98 5

477 642 .60 4

4 7764 2. 604

1

2

3

4

5

6

Reference pumpingOptimal pumping

-15.00-10.00-5.000.005.00

10.0015.0020.0025.0030.0035.00

Zone 1 Zone 2 Zone 3 Zone 4 Zone 5 Zone 6 Ponds Barrier

hm3 /y

y

8387.962

3598.126

1912.000

17050.812

14780.677

4334.562

3866.931

10402.932

91976.839

10312.991

153982.228

122 613. 720

1490 316. 98 5

477 642 .60 4

4 7764 2. 604

1

2

3

4

5

6

But the most interesting results where the shadow prices (hydraulicefficiency)

ARTIFICIAL RECHARGE PONDS = 0.6 SEAWATER INTRUSION BARRIER = 1.7

With corrective measures

Results: OK with corrective actions

The end of the story

1) The seawater Intrusion Barrier was built, but with osmotized water (a negative influence of California!!) so it stopped operation during the crisis

2) The Groundwater Users community did not want to hear of “imposed” pumping reductions, but they adopted the model have been using it themselves for self-control

3) SWI has been controlled

In summary

• Reliable models are possible. Do not trust GCMs forrainfal, but for circulation patterns

• The present is not bad for the people (good quality watersupply, good agricultural production).

• But the present is concerning for water dependentecosystems (rivers in poor shape, depleted coastalecosystems)

• Water rights, often appropriated (not acknowledged bythe law), often exceed availability

• The administration lacks tools to enforce «scientific» allocation plans.

• What works are «user’s communities» and the «rental» of water rights.