Agricultural impact studies in areas of complex orography: the need for high resolution climate models in the Iberian Peninsula Universidad Politécnica.

Agricultural impact studies in areas of complex orography:

the need for high resolution climate models in the Iberian Peninsula

Universidad Politécnica de Madrid&

Universidad de santiago de Compostela

(PRUDENCE PROJECT)

CAgM EXPERT TEAM MEETING, Geneva 15-18 Nov. 2004

http://prudence.dmi.dk/main.html

PRUDENCE PROJECTPrediction Uncertainties Describing

European Climate change and Effects

M. Inés Mínguez, Margarita Ruiz-Ramos, Carlos H. Díaz-Ambrona, Miguel Quemada and Federico Sau

IT: Daniel del Valle Sergio de Gruijter UPM & USC

Agricultural Impact studies in areas of complex orography


http://prudence.dmi.dk/main.html

RCMs

A2 & B2 SCENARIOS (SRES)

FORCING

Agricultural Impact (Crop Models)

AGCMs



RCMs in PRUDENCE

• Comparison among 9 RCMs • Forcing with 1 RGM: HadAM3H (less in

Arpège RCM which has it’s own forcing s.)

• 2 climate change scenarios: A2 & B2



RCMs in PRUDENCE

1. HadRM3H (HC: Hadley Centre for Climate Prediction and Research)

2. RegCM (ICTP: The Abdus Salam Intl. Centre for Theoretical Physiscs)

3. CHRM (ETH: Swiss Federal Institute of Technology)

4. RACMO (KNMI: The Royal Netherland Meteorological Institute)

5. ARPÈGE (MF: Meteo France)



RCMs in PRUDENCE

6. CLM (GKSS: GKSS FORSCHUNGSZENTRUM; Institute for Coastal Research)

7. REMO (MPI: Max Planck Institute for Meteorological Research)

8. PROMES (UCM: Universidad Complutense de Madrid)

9. HIRHAM (DMI: Danish Meteorological Institute)



Crop models used in PRUDENCE

• Physiological type models: DSSAT3.5-CERES

(mechanistic model; Jones et. al., 2003)

• Cropping system-type model: CropSyst

(simpler model; Stockle et al., 2003)



Non-linear response in both types.

Compensating effects of climate change reflected in models

• CO2 increase (for a given T)

– Increase in photosynthetic rates– Decrease in Transpiration rates (ET)

• T increase - Increase in evapotranspiration (ET) - Shorter crop cycles (thermal time)



Crop models used in PRUDENCE

INPUTS:

• Daily weather data: SR, Tmax, Tmin, P (Optional: VPD, wind speed)

• Soil profile caracteristicsLL, DUL & SAT (for each soil layer)

• Species charaterisitics (eg wheat) and variety (eg cv. Anza, Marius) Genetic coefficients that allow to define thousands of genotypes

• Soil initial conditions (soil water content)



Crop simulations• Climate change impact assessment

- Use of winter (wheat) vs summer crops (maize) to discriminate between seasons

- Crop management (Irrigated and Rainfed) for changing uncertainty levels



Adding Uncertainties• from the RCMs predictions

• from the crop models per se

• from the extrapolation of crop models, tested within a limited sample of real conditions to a much larger domain

• soil profiles (DUL, LL, SAT and depth)



Changing uncertainty levels


Greater uncertainty level: rain-fed systems

• Rainfall data

• Soil profiles: LL, DUL, SAT, depth

• Initial soil conditions

• Crop responses to water stress


Smaller uncertainty level: irrigated systems (if water

available)

• Potential yield simulation (elimination of water stress)

• Spatial variability in soil textures and depth has much less influence in yield

• Rainfall predictions do not affect yields

Effect of Tmax, Tmin and SR



Crop model calibration and validation has to focus on the main processes involved at crop and cropping systems level

• phenology, biomass and yield• ETc and net irrigation requirements



Spatial scales with impact models

• Crop model calibration and validation:

Site-specific

Some agricultural areas outside the calibration (validation)

• Crop models generate data within the range recorded by MAPYA





34 soil groups specific profiles

Soil data base:

PROMES land-grids



Species/cvs for season discrimination

• Autumn-winter-spring for

winter (cv. Marius) and spring wheat (cv. Anza)

rainfed and irrigated (ETmax)

Sow. 15 october

• Spring-summer

maize (irrigated)

Sow. 15 April; 15 May; 1 June



Filters• Varieties (CVs) with different cold

requirements for flower induction: spring (Anza) vs winter (Marius) cultivars

enhancing the important differences in T (Tmin) in winter

Winter wheat: vernalization needs have to be fulfilled.



Simulations• Current: 1960-1990

• A2 or B2: 2070-2100

Outputs: average of 30 years (current & A2 or B2)

Results are presented in maps:

A2 or B2/Current (%) or Relative Yield (%)



Crop failures

A2/current Relative Yield (%)< 5050 - 7070 - 9090 - 110110 - 130130 - 150150 - 200> 200

REMO winter wheat rain-fedCERES

C ro p fa ilu res

A 2 /co nt r o l R e la tiv e Y ie ld (% )

< 5 050 - 7 070 - 9 090 - 1 10

11 0 - 13 013 0 - 15 015 0 - 20 0> 2 0 0

A2/crtl



Crop failures


REMO winter wheat irrigatedCERES

C ro p fa ilu res

A 2 /cu rre n t R e la tiv e Y ie ld (% )

< 5 050 - 7 070 - 9 090 - 1 10

11 0 - 13 013 0 - 15 015 0 - 20 0> 2 0 0

A2/crtl



HadAM3H A2driving conditions

Relative yield

CERES- Spring Wheat (DSSAT)

Irrigated

Sowing date: 15 october



Crop failures


HadAM3H (HC)CERES-Spring Wheat

C ro p fa ilu res


< 5 050 - 7 070 - 9 090 - 1 10

11 0 - 13 013 0 - 15 015 0 - 20 0> 2 0 0



Irrigated

A2

Crop failures

A2/current Relative Yield (%)

< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200

RegCM (ICTP)CERES-Spring Wheat



Irrigated

A2

Crop failures


CHRM (ETH)CERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

Crop failures


RACMO (KNMI)CERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

ArpègeCERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

Crop failures


CLM (GKSS)CERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

Crop failures


HadRM3H (HadleyC)CERES-Spring Wheat

C ro p fa ilu res


< 5 050 - 7 070 - 9 090 - 1 10

11 0 - 13 013 0 - 15 015 0 - 20 0> 2 0 0



Irrigated

A2

Crop failures


REMO (MPI)CERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

Crop failures


PROMES (UCM)CERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

Crop failures


HIRHAM (DMI)CERES-Spring Wheat

Crop failures


< 5050 - 7070 - 9090 - 110

110 - 130130 - 150150 - 200> 200



Irrigated

A2

Crop model

• Differences between the two crop models outputs exist:

CERES-DSSAT discriminates more among polygons



Crop failures

B2/current Relative Irrigation (%)< 5050 - 7070 - 9090 - 110110 - 130130 - 150150 - 200> 200

Crop failures

A2/current Relative Irrigation (%)< 5050 - 7070 - 9090 - 110110 - 130130 - 150150 - 200> 200

Changes in irrigation requirements for maize

A2 PROMES B2



MAIZE irrigated (kg/ha)

HIRHAM HIRHAM high

Córdoba Avg CV Avg CV

Xerorthent Control 8837 11 8122 12

A2 2893 84 6681 13

Rel(%) 33 82

Chromoxerert Control 8497 14 4645 46

A2 2891 84 6223 16

Rel(%) 35 134

50 km vs 25 kmAgricultural Impact studies in areas of complex orography


Some conclusions

• Regions/areas show similar yield trends in many RCM models

• Influence in rainfed crops of Tmin vs rainfall• All models detect winter vs spring wheat • Higher resolution: more differences among soil

polygons



Still working on:

• A2, B2 scenarios

• 50 km vs 25 km (grid)

• Other forcings: try different GCMs with the same RCM.

THANK YOU



SpWheatKg/ha

SR HadAM3H PROMES RegCM

M CV M CV M CV

Navarra

42ºN

500 m

2074 (53)

Ctl 3530 43 663 32 6337 44

A2 4821 42 1466 33 8678 26

B2 -- 1096 47 7330 37Valladolid

41º N

400 m

1533 (60)

Ctl 3480 53 1050 44 5432 40

A2 5918 37 2200 36 4000 72

B2 --- 2226 37 4127 64Córdoba

38ºN

120 m

3047 (39)

Ctl 627 39 219 77 3264 55

A2 1720 58 793 66 3997 72

B2 ---- 472 74 4127 64CAgM EXPERT TEAM MEETING, Geneva 15-18 Nov. 2004


Agricultural impact studies in areas of complex orography: the need for high resolution climate models in the Iberian Peninsula Universidad Politécnica.

Documents

climate change scenarios

climate prediction

prudence inputs

prudence comparison

european climate change

high resolution climate

meteorological research

seasons crop management