Monitoring pre-monsoon drought in Bangladesh using remote sensing technique

Abstract

Drought has been a prevalent concern in Bangladesh over the past few decades, and the findings of several studies have indicated that Bangladesh has a high risk of drought, in association with a significant increasing trend of temperature. However, little attention has so far been given in Bangladesh to the mitigation and monitoring of drought, although few studies have been conducted for drought assessment based on either rainfall and temperature or a drought index based on rainfall, such as the Standardized Precipitation Index (SPI). The objective of this study is to assess drought conditions in Bangladesh using long-term satellite data from January to May (2001-2014). Temperature-Vegetation Dryness Index (TVDI) is a drought index based on remote sensing data that exploits the relationship between Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) for estimating soil moisture condition has been used. A systematic approach was adopted in the methodology for considerations to; i) identify the spatial and temporal variation of drought using TVDI, ii) examine the relationship between TVDI and other climatology as well as environmental variables (such as soil moisture, LST, NDVI, rainfall, and Land Use Land Cover (LULC)), and iii) compare TVDI result with field investigation. Results indicate that drought is a concerning problem in Bangladesh and drought conditions varies spatially and temporally. It is clearly observed from the TVDI results that the problem of drought was not prominent in January and February (2001-2014) due to low temperatures. But the effect of drought was considerably high for the rest of the three months, of March, April and May (2001-2014) due to high temperature. However, there were still severe drought conditions have been observed in several small parts of the study areas where no additional water supply were available during that time, except rainfall. Nevertheless, a large part of the study area was still unaffected by the drought even during very hot weather condition due to massive irrigation which has been ascertained during the field investigations. Among the investigated parameters, very closed agreements were found between TVDI and LST as well as NDVI and LULC, although relationships between TVDI and rest of the parameters were not well-defined. This study also found that the TVDI result is in a good agreement with the field investigation. Most importantly, the correlation between TVDI and field investigation clearly indicates the important of this TVDI for the investigation of drought in a complex environment in Bangladesh, where drought estimation using only meteorological data (rainfall and temperature) is ineffective due to anthropogenic and environmental factors which modified the soil moisture condition across the ground very severely. Finally, this study highlights the potential of drought monitoring using remote sensing technique especially on the use of TVDI in Bangladesh due to lack of meteorological data.