Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Titre : Evaluation of agricultural management from field to catchment scale : Development of a parsimonious agro-hydrological model

Evaluatie van landbouwbeheer van veld- tot bekkenschaal : Ontwikkeling van een eenvoudig agro-hydrologisch model

Auteur : Van Gaelen, Hanne

Université de soutenance : KU Leuven

Grade : Doctor of Bioscience Engineering (PhD) (Leuven) 2016

Résumé
Due to the increasing world population and prosperity, global food production needs a 70% increase by 2050. To achieve this taking into account the limited land and water resources, an increase in productivity needs to be accompanied by an increase in crop water productivity. Improved agricultural management is one of the key solutions for upgrading (crop) water productivity, especially in rainfed cropping systems in drought-prone regions. However, agricultural management practices are only sustainable if they are selected considering their suitability under changing environmental conditions. Also, their potential impact on regional water availability should be taken into account. Agro-hydrological models are suitable tools to investigate the impact of several agricultural management strategies under various environmental conditions. While simple crop and hydrological models are limited with respect to the number and accuracy of the processes they incorporate, complex models have high demand for data. Due to these limitations, there is a need for new agro-hydrological models that accurately simulate both crop productivity and water availability in agricultural catchments but have low data and calibration requirements. This study aimed at developing a widely applicable, parsimonious agro-hydrological model, AquaCrop-Hydro, by linking the process-based AquaCrop crop water productivity model with a conceptual hydrological model. First, the AquaCrop submodel to simulate the effect of agricultural management at field scale was evaluated and further improved. AquaCrop can simulate the effect of various agricultural management practices on the soil water balance, crop canopy development, crop transpiration and crop (water) productivity of an agricultural field. Next to irrigation management, the model considers crop management, soil management, field surface management, mulches, soil fertility management and weed management. Two of these practices were further studied in detail. First, AquaCrop’s semi-quantitative simulation procedure to simulate crop response to soil fertility stress was elaborately discussed and evaluated against field experimental data of maize and wheat in Nepal, quinoa in Bolivia and tef in Ethiopia. Second, a new procedure to simulate crop production in weed-infested fields was developed and tested against field data of barley in Ethiopia and winter wheat in Australia. Evaluation of simulation results showed that AquaCrop performed well to simulate the soil water content, crop development and production under various environmental conditions and different water, soil fertility and weed infestation levels. Furthermore, a scenario analysis demonstrated that AquaCrop enables efficient analysis of a broad range of agricultural management practices in order to develop management strategies that are tailored to the local agronomic and environmental conditions. Subsequently, the AquaCrop model was linked to a conceptual hydrological model. The resulting AquaCrop-Hydro model was evaluated and applied to the Plankbeek catchment, an agricultural catchment in Flanders, Belgium. Comparison against historical observations showed that AquaCrop-Hydro performed well to simulate crop production and river discharge at the outlet of the catchment. Moreover, an impact analysis demonstrated AquaCrop-Hydro’s ability to evaluate various agricultural management strategies for climate change adaptation with respect to their effect on crop production as well as water availability. Finally, the strengths and limitations of AquaCrop-Hydro as compared to other agro-hydrological models was assessed. The model is widely applicable to agricultural catchments with varying characteristics. Due to its parsimonious nature it is especially useful for application in data-scarce regions, where it provides good estimates while alleviating the burden of high data and calibration requirements. Although there is room to improve model accuracy and functionality, AquaCrop-Hydro can be applied to evaluate agricultural management strategies and support sustainable water management from field to catchment scale.

Présentation (ORBI)

Version intégrale (10,86 Mb)