Seasonal comparisons of meteorological and agricultural drought indices in Morocco using open short time-series data

Abstract

Although the preliminary investigations of NDWI demonstrated its sensitivity to vegetation water content, drought indices based on NDWI short time-series are still understudied compared to those derived from NDVI and LST, such as VCI, SVI and TCI. On the basis of the open data, this paper introduces a new index derived from NDWI short time-series, and explores its performance for drought monitoring in Mediterranean semi-arid area. The new index, Standardized Water Index (SWI), was calculated and spatiotemporally compared to both meteorological drought index (TRMM-based SPI) and to agricultural drought index (NDVI-based SVI) for the hydrological years and autumn, winter and spring seasons during a period of 15 years (1998–2012). Furthermore, the response and spatial agreement of the meteorological and agricultural drought indices (SWI, SVI and SPI) were compared over two land use classes, rainfed agriculture and vegetation cover, for the studied years and seasons. The validation of SWI was based on in situ SPI and cereal productions. The analysis of the 336 cross-tables, proportions of concordance and Cohen's kappa coefficients indicate that SWI and SVI are concordant comparing to other combinations for hydrological years and for the three seasons. The study points that the spatial agreements of drought indices over rainfed agriculture and over vegetation cover are different. It is relatively more important in the rainfed agriculture than in the vegetation cover areas. Our results show that the agreement between vegetation drought indices and meteorological drought indices is moderated to low and the SPI is slightly more concordant with SWI when it is compared to SVI in autumn and winter seasons. The validation approach indicates that drought affected area, according to SWI, is highly correlated with cereal production. Likewise, a satisfactory correlation was revealed between SWI and in situ SPI.