INTEGRATION OF REMOTE SENSING DERIVED PARAMETERS IN A CROP MODEL: CASE OF HAY

The aim of this study is to assess the interests of integrating remote-sensing-derived parameters (LAI, harvest and irrigation dates) in a crop model (PILOTE) that simulates vegetation growth for hay crop. Results show that optical images are suitable to feed PILOTE with LAI values without inducing significant errors on the predicted Total Dry Matter (TDM) values (Root Mean Square Error "RMSE" = 0.41 t/ha and Mean Absolute Percentage Error "MAPE" = 22%). Moreover, optical images with revisit times lower than 16 days are adequate to feed PILOTE with remotely sensed harvest dates (RMSE < 0.44 t/ha, MAPE < 10.8%). Finally, results show negligible errors for the TDM predictions when radar images are used to determine all irrigation dates (RMSE = 0.14 t/ha, MAPE = 3.1%). Disregarding one or two irrigations within a period with enough rainfalls does not induce significant errors for the predicted TDM values. However, it causes noticeable underestimations in drier periods (a maximum of 1.55 t/ha for TDM out of 3.43 t/ha predicted using all irrigation dates). This gives an insight about necessary revisit times of radar satellites in order to correctly predict the growth of an irrigated crop.