Drought Monitoring for Northern Part of Iraq Using Temporal NDVI and Rainfall Indices

Abstract

Climate change is the major global challenge facing water resources managers. Drought is a natural hazard temporarily affecting almost every region in the world. In this study, the climate change in term of rainfall fluctuation in the northern part of Iraq (Mosul, Kirkuk and Salah Al-Din) has been investigated using a set of data containing monthly precipitation for the period from 1980 to 2010, and the MODIS time series images for the period from 2000 to 2010. All data series have been used to calculate standardized precipitation index (SPI) and Normalized Difference Vegetation Index (NDVI). Monthly rainfall data from 12 stations were used to derive the SPI at several time scales (3, 6 and 12-months), the analysis was carried out for the period from 1980 to 2010. Results of the SPI analyses showed that the year 2007–2008 was an extremely drought year for the whole study governorates (Mosul, Kirkuk and Salah Al-Din) with the lowest SPI-12 values −2.67, −2.07 and −2.0 for the three above mentioned governorates, respectively. The results also pointed to the importance of using short time scales in detecting and monitoring the agricultural drought during the crop growing season. The multiple time scales analyzed in this study reflected a clearer picture of the severity and frequencies of drought events, which happened in the study area. The NDVI results were analyzed to get the agricultural drought risk map. The highest NDVI values were 0.33 in 2001, 0.39 in 2003 and 0.20 in 2001 for Mosul, Kirkuk and Salah Al-Din, respectively. While the lowest NDVI values were 0.10 in, 0.19 and 0.13 in 2008 for the three above mentioned governorates respectively. This study emphasized the use of Remote Sensing and GIS in the field of drought risk evaluation. The results showed that the NDVI is an efficient way to monitor changes in vegetation conditions (weekly or daily) during the growing season, and can be used as simple and cost-efficient drought index to monitor agricultural drought at a small or large scale. The NDVI and rainfall were found to be highly correlated 0.83, 0.70 and 0.72 for Mosul, Kirkuk and Salah Al-Din, respectively. Therefore, the temporal variations of NDVI are closely linked with precipitation. Results of statistical correlation analysis between NDVI and SPI (3, 6 and 12-months) time scales showed that the highest correlation coefficients were between NDVI and SPI-6, which verified that the short time scales could be related closely to soil moisture. It was observed that the studied indices (NDVI & SPI) could be effectively used for monitoring and assessing agricultural productions and in that way, proper agricultural policies can be adopted to mitigate drought impacts.