Use of a MODIS Satellite-Based Aridity Index to Monitor Drought Conditions in the Pearl River Basin from 2001 to 2021

Kunlong Niu, Junliang Qiu, Shirong Cai, Wenxin Zhang, Xiaolin Mu, Edward Park, Xiankun YangISPRS Int. J. Geo Inf 2022 · DOI

TL;DR

This study introduces and applies a novel satellite-based aridity index (SbAI) to monitor drought conditions in the Pearl River Basin (PRB) from 2001 to 2021. The findings reveal a general trend towards increased wetness in the PRB, particularly in the dry season, though human activities like urbanization are exacerbating localized drought conditions in the monsoon season.

What Problem Does This Paper Address?

Droughts are a significant natural hazard with increasing prevalence due to global climate change. While much research has focused on arid and semi-arid regions, humid and sub-humid areas, like the Pearl River Basin (PRB), also experience seasonal droughts and water shortages that are less understood. Traditional drought monitoring often relies on meteorological stations, which have sparse and uneven distribution, limiting large-scale analysis. This paper aims to address this gap by using satellite data to provide a comprehensive spatiotemporal analysis of drought conditions in the PRB, offering insights into regional trends and the influence of both natural and anthropogenic factors.

Methods

The core of this study is the application of the Satellite-Based Aridity Index (SbAI). The SbAI is calculated as the ratio of the difference between day and night land surface temperatures (LST) to the absorbed solar radiation. This index reflects land surface wetness, with higher SbAI values indicating drier conditions and lower values indicating wetter conditions. The researchers utilized MODIS satellite data for LST and surface reflectance to derive the SbAI.

To analyze the temporal trends of the SbAI, the Mann-Kendall (M-K) test was employed to assess the significance of trends, and the Theil-Sen Median slope estimation was used to quantify the trend magnitude. The Pettitt test was used to detect abrupt change years. For spatial analysis, the SbAI was calculated at a pixel scale across the PRB. The study also involved comparing the SbAI with precipitation data from TRMM and GPM missions for validation.

Datasets

  • MODIS Satellite Data:
    • MOD09GA (Terra satellite): Surface reflectance data (1 km spatial resolution, daily temporal resolution) used for albedo calculations.
  • MOD11A1 (Terra satellite): Land Surface Temperature (LST) data (1 km spatial resolution, daily temporal resolution) used for day/night temperature differences.
  • Precipitation Data:
    • TRMM (Tropical Rainfall Measuring Mission): Historical precipitation data.
  • GPM (Global Precipitation Measurement mission): Higher-precision precipitation data (0.1° spatial resolution, 0.5-hour temporal resolution).
  • Other Datasets:
    • 2001–2020 Water Resources Bulletin for the PRB (from Pearl River Water Resources Commission).
  • 30 m annual land cover datasets in China (1990–2021).
  • DEM data.

Key Insights and Ideas

  • Humid Basin, Wetter Trend: Despite being a humid region, the PRB is generally becoming wetter over the past two decades, a trend observed both annually and particularly in the dry season.
  • Seasonal Dichotomy: While the overall trend is wetter, the monsoon season shows a slight increase in drought conditions in some areas, contrasting with the significant shrinking of drought areas in the dry season.
  • Human Impact Amplifies Drought: In areas like the Pearl River Delta, rapid urbanization and increased impervious surfaces are reducing water storage capacity, leading to drier conditions despite potentially stable rainfall. This highlights how human activities can exacerbate drought even in generally wet regions.
  • SbAI as a Robust Indicator: The SbAI, derived from satellite data, proves effective in monitoring drought in a region with sparse meteorological stations, offering a valuable alternative to traditional indices.

Key Results

  • Overall Trend: The inter-annual SbAI in the PRB showed a statistically significant decreasing trend (-0.624/yr) from 2001 to 2021, indicating a trend towards increased wetness.
  • Seasonal Trends: The decreasing trend in SbAI was statistically significant in the dry season but not in the monsoon season. Drought conditions showed an insignificant expansion in the monsoon season but a statistically significant shrinking in the dry season.
  • Regional Variations: Areas 1-3 (headwater and western regions) were identified as relatively drier, especially in the dry season, while Areas 4-7 (eastern and delta regions) were generally wetter.
  • Drought Events: Most detected drought events occurred in November, December, January, and February, with durations typically lasting 2-5 months.
  • Correlation with Precipitation: The SbAI showed a significant negative correlation with GPM precipitation data, as expected, with higher rainfall weakening drought conditions.

Limitations and Future Work

  • Spatial Resolution: The study acknowledges that MODIS products, with their 1 km spatial resolution, are primarily suitable for large-scale analysis and may not capture fine-scale drought variations.
  • Data Consistency: While GPM data and SbAI showed a consistent negative correlation, they were not completely consistent. This discrepancy is attributed to the fact that GPM reflects natural climate change, while SbAI is influenced by both natural factors and human activities.
  • Future Work: The study suggests that the findings provide a valuable reference for drought assessment, prevention, and environmental management in the PRB. Further research could explore the specific impacts of different human activities on drought conditions in more detail and potentially integrate higher-resolution satellite data for finer-scale analysis.

Summary updated

BibTeX
@article{niu2022use,
  author = {Kunlong Niu and Junliang Qiu and Shirong Cai and Wenxin Zhang and Xiaolin Mu and Edward Park and Xiankun Yang},
  title = {{Use of a MODIS Satellite-Based Aridity Index to Monitor Drought Conditions in the Pearl River Basin from 2001 to 2021}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2022},
  doi = {10.3390/ijgi11110541},
}
Figure