Abstract

Terrestrial groundwater dependent vegetation (TGDV) are crucial ecosystems which provide important goods and services such as carbon sequestration, habitat, water purification and aesthetic benefits in semi-arid environments. Global climate change and anthropogenic effects on surface water resources have led to increased competing claims on groundwater resources to meet an exponential water demand for environmental needs, agricultural and developmental needs. This has led to the unsustainable exploitation of groundwater resources, resulting in groundwater table declines threatening the sustainability of TGDV. It is on this premise that the review aims to provide a detailed overview on the progress in remote sensing of TGDV. More specifically, the paper provides a background on TGDV and threats, and then further explores recent knowledge on vegetation response to groundwater variability and climate change impacts on TGDV. This review also focuses on recent progress in remote sensing (RS) and geographic information systems (GIS) based techniques for mapping and monitoring of TGDV and explores the available satellite products and classification techniques. Finally, the challenges of remote sensing and future research direction are explored. To date, research on TGDV has gained considerable interest with the year 2020 releasing the most publications. Of significant importance is an increase in studies integrating field measurements, model-based techniques with remotely sensed estimates. Despite this progress, only 0.06% of groundwater dependent ecosystems (GDE) research has utilized remote sensing techniques in the past 20 years, with the top three publishing countries namely, Australia, USA, and China. The literature reveals that TDGV are highly heterogenous, complex ecosystems with unique responses to variable groundwater tables. The vegetation responses differ with the landscape, vegetation type, and seasonality at specific groundwater table thresholds. Despite progress in TGDV scientific research, further remote sensing studies are required to understand the annual and inter-annual vegetation response to groundwater variability at local scales. Further, climate impacts are difficult to discriminate from other influences such as disturbances, management, and anthropogenic activities. Moreover, new generation remote sensing products integrated with machine learning techniques have the potential to improve TGDV delineation. Despite these challenges, the development of cloud computing technologies such as google earth engine (GEE) and artificial intelligence (AI) provide advanced computer-processing capabilities for long-term monitoring and integration of multi-source datasets required to capture the effects of climate and groundwater variability on TGDV.