Assessment of Agricultural Drought Using the Normalized Difference Drought Index (NDDI) to Prediction Drought at Corong River Basin

Abstract

As a complex and widespread natural phenomenon, drought poses a significant threat to the agricultural sector, especially in developing countries, resulting in significant economic losses. Its close relationship with water resilience and crop production necessitates sophisticated monitoring approaches for agricultural drought. Leveraging satellite remote sensing technology and various data types such as multispectral, thermal infrared, and microwave, can monitor drought on a large scale. This technology provides a comprehensive perspective for timely and spatial data collection, facilitating the monitoring of vegetation in vast agricultural areas. The study focuses on developing an agricultural drought model from 2017 to 2021, using Landsat 8 imagery. The model integrates the Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI), resulting in the establishment of the Normalized Difference Drought Index (NDDI) method. To predict agricultural drought in the Corong River Basin, the study employs the Seasonal Autoregressive Integrated Moving Average (SARIMA) model. Findings reveal varying degrees of dryness in the Corong River River Basin, with 77% categorized as Strong dry conditions, 1% as Dry, 0.3% as Moderate wetness, and 21.6% as wetness. Drought predominantly occurs between July and October, impacting approximately 78% of the total dry area and extending across almost the entire region. The SARIMA (0,0,1)(3,0,0)¹² model, with a MAPE value of 0.2399, emerges as the most effective for predicting agricultural drought. These forecasted results provide critical insights into the level of agricultural drought in the Corong River Basin and valuable information for drought mitigation strategies, especially in regulating the distribution of irrigation water.