Application of the cross-simulation method to the analysis of the rain-flow relationship from the GR2M model: Case of the Taghjijt sub-basin (GUELMIM- MOROCCO)

Abstract

The study of the relationship between precipitation and run-off is very interesting especially, in arid and semi-arid regions, and in watersheds where flow data are scarce. The case of the Taghjijt basin, which has an area of 1400m2, is a case study for our research using the monthly model of rural engineering with two optimizable parameters X1 and X2 (GR2M), consisting of the application of the method of cross-simulations and the evaluation by the criterion of NASH-SUTCLIFFE during the period of calibration and validation. The model performance evaluation criteria, numerical and graphical, yielded satisfactory results in calibration and validation, including the correlation coefficients between observed and calculated flows. The monthly model showed better results for the calibration in comparison to the validation. This is due to the drought in the region, which has significantly affected the surface flows. In addition, tectonics, lithology, and Geographic Information Systems (GIS) are used to represent the general characteristics of the studied watershed. The data of the period from 1986 to 2016 of the station of Taghjijt are used and results show the performance of the estimation model GR2M for the period of calibration as well as in the period of validation, in the prediction of floods, even if it applies only two parameters. The phase shift between observed and simulated flows and precipitation is explained by lithology and tectonics. The creation of a significant real flow is dependent on the lithology and tectonics of the superficial rocks, similarly, these two factors can explain the observed mismatches between observed and simulated flows. The results of this study are useful for the adequate management of water resources, for the prediction of floods, and finally for a better adaptation of semi-arid territory to climate change.