1. Dr. Adriana-Cornelia Marica & Alexandru Daniel
Using seasonal climate forecasts and the CROPWAT model for irrigation planning and management
Dr. Adriana-Cornelia Marica & Alexandru Daniel
ROMANIA
WMO & COST 718 Expert Meeting on Weather, Climate and Farmers, Geneve, Switzerland, November 2004

2. Aim:
 describe the experience of Romania as an example of application of seasonal climate forecast in the agriculture sector;
 demonstrate how seasonal climate forecast combined with the CROPWAT model can estimate the soil water deficits during maize growing season of 2003.

3. Objectives:
 evaluate and predict a few months in advance the daily and total soil moisture deficit during maize vegetation period in the rainfed and irrigated conditions;
 compare various options for water supply and irrigation management;
 assess yield reduction due to crop stress under rainfed conditions or deficit irrigation;
 compare the model results and analyze the skill level of seasonal forecast;

4. Sites in the southern region of Romania used for soil water deficits assessment

5. Benefits for Agriculture:
Seasonal Forecasts:
predict major climate trends over a period of several months to a few seasons;
 indicate areas where there is an increased likelihood of some deviation from the climate mean (dry or wet, warm or cold conditions);
Benefits for Agriculture:
offer the potential for farmers to plan and take decisions on crop management (if to sow or not, to spray or not, to irrigate or not);
 modify decisions to decrease unwanted impacts or take advantage of expected favorable conditions.

6. The monthly means values of temperature and rainfall estimated by seasonal forecast for the interval June-September 2003, as compared with the normal

7. Functions of CROPWAT model:
Is a method to support decision making for irrigation planning and management;
Calculates reference evapotranspiration, crop water requirements and irrigation requirements;
Develop irrigation schedules based on a daily soil-moisture balance;
Allows the development of recommendations for improved irrigation practices, the planning of irrigation schedules and the assessment of production under rainfed conditions or deficit irrigation

8. Input data used: Monthly means climatic data:
measured during April-May 2003 (min.& max. temp., humidity, sunshine duration, wind speed and monthly rainfall)
estimated by seasonal forecasat for June-September 2003 (air temperature and rainfall)
Crop data:
 real sowing date: 20 April (Calarasi) / 5 May 2003 (Tg.Jiu)
 standard crop coefficient (Kc), crop yield data (Ky)
and depletion fraction (P)
Soil data:
 total available moisture: 227/191 mm
 initial available soil moisture: 170 mm/163 mm
 maximum root infiltration rate: 40 mm/day
 maximum rooting depth: 1m

9. Model application:
Running the CROPWAT model with rainfed maize in the forecasted and real weather conditions;
Running the model with application of irrigation using different scheduling criteria:
 irrigate at fixed intervals and depths;
 irrigate when 70% of total available soil moisture
depletion occurs;
 irrigate when 70 and 100% of readily soil moisture
depletion occurs.

10. Results
Daily soil moisture deficit simulated with CROPWAT model during rainfed maize growing season, in the weather forecast conditions for summer 2003
TAM: total available moisture,
RAM: readily available moisture
SMD: soil moisture deficit

11. Cumulated values on the whole vegetation period of maize variables simulated with CropWat in the estimated weather condition of 2003
Sites
Total Rain (mm)
Eff.Rain (mm)
SMD (mm)
Yield Red. (%)
Calarasi
Tg. Jiu
144
398
135
339
443
165
53% 4%

12. Simulated results for irrigated maize at Calarasi site in the forecast weather conditions for summer 2003, using two scheduling options:
4 irrigation of 60mm Irrigate 70% of TAM

13. Simulated results for irrigated maize at Calarasi site in the forecast weather conditions for summer 2003, using “optimal” scheduling criteria:
Irrigate 70% of RAM
Irrigate 100% of RAM

14. The effects of the rainfed and different irrigation scheduling simulated with CROPWAT at Calarasi site
Options
Net irrigation
(mm)
Yield reduction
(%)
Rainfed
Irr.fixed int&depth
Irr. 70% of TAM
Irr. 70% of RAM
Irr. 100% of RAM -
240
366
405
449
53%
24%
10%

15. Skill level
Changes in growing season rainfall and soil moisture deficit in the seasonal weather forecast as compared with the real weather conditions
-60%
-35%
+11%
+91%

16. Results
Effects of estimated and real weather conditions on rainfed maize yield reduction due to crop stress

17. CONCLUSIONS
 The application of seasonal weather forecast combined with CROPWAT model allows the estimation of soil water supply conditions with 3-4 months in advance and in case a skillful forecast can provides useful information for farmers to make decisions on irrigation planning and management in a dry season, avoiding yield reduction.
 According to the seasonal forecast for summer 2003 and the first month of autumn, in Targu Jiu the maize crop was not affected by the soil water deficit, while in Calarasi the high soil moisture deficit led to yield reduction up to 53% from the productive potential;

18. CONCLUSIONS
 In the real weather conditions, the maize crop was affected by drought at both sites, the yield reduction percentage was 64% in Calarasi and 41% in Targu Jiu;
 In the most sensitive areas to drought by application of irrigation the yield losses would be significantly reduced;
 The comparative analysis of the simulated results has emphasized a higher skill level of seasonal forecast for Calarasi site than for Targu Jiu site;
 The precision level of forecast information for temperature was higher than for rainfall at both sites.

19. THANK YOU FOR YOUR ATTENTION