Soil moisture plays a critical role in numerous environmental and agricultural processes, influencing water availability, crop growth, and climate regulation. While ground-based and satellite-derived measurements are valuable tools for monitoring soil moisture, they come with notable limitations. Ground-based observations, though accurate, are often restricted by sparse spatial coverage and logistical challenges, while satellite data, although providing broader coverage, can be limited by temporal resolution and atmospheric interference.
Additionally, most publicly available datasets, whether from ground or satellite sources, often lack the spatial resolution needed for detailed regional analysis and decision-making. To bridge this gap, we present the historical soil moisture data derived from the Noah-MP land surface model (https://alice-lab.com/noah-mp-land-surface-model) that delivers high-resolution coverage across the entire Asian continent (Fig. 1). This approach enables more comprehensive monitoring and management, enhancing the understanding of regional water cycles and supporting agriculture and climate resilience strategies across diverse landscapes.
Figure 1: The 1 km soil moisture at 0-10 depth across Asia in September 2024 (a), which aligns well but provides more spatial detail than public data such as GLDAS (b vs. c) It is also consistent with SMAP (9 km) data in various areas (b vs. d), with the disagreement likely related to sensible soil depth, sensor sensitivity, and model uncertainty.
The average soil moisture in Central, North, North East, and East Thailand is evaluated with the ESA-CCI satellite-derived (merged) product. Figure 2 shows significant agreement, especially in terms of seasonal variations. The model shows superior performance during the wet season but underestimates soil moisture during the dry season. Overall, the model performs well, with a correlation up to 0.97 and a KGE up to 0.9 (Table 1). It also performs well in bias, with an RMSE of about 0.03 m3/m3 versus the soil moisture error of 0.04 m3/m3.
Figure 2: The average soil moisture in Central, North, North East, and East Thailand from 1958 to the present. For comparison, soil moisture from satellite data acquired from the ESA-CCI product from 1979 to 2022 is shown.
The 1 km resolution meteorological data across Asia is developed at the Asian Land Information for Climate and Environmental Laboratory (ALICE-LAB; https://alice-lab.com) at the Asian Institute of Technology in Thailand. To address Asia’s data scarcity concerns, we prioritize integrating advanced approaches such as hydrology and land surface modeling, remote sensing, artificial intelligence (AI), data assimilation, and computational hydrology (via high-performance computing; HPC). The produced data comprise high-resolution historical and seasonal forecast meteorological data, streamflow simulation, and hydrologic projections under climate change, not only in Asia but also globally.
Our findings demonstrate that the high-resolution, model-derived soil moisture data align well with satellite products, confirming their reliability in various environmental and agricultural applications. Importantly, our dataset extends beyond the temporal span of satellite observations, offering longer time series and finer spatial detail that enhance its utility for long-term analysis and regional planning. However, the analysis reveals a consistent underestimation of soil moisture during dry periods, underscoring the need for further refinement of model parameters and underlying physics to achieve better agreement with satellite or ground measurements. Addressing these limitations will enhance the model’s accuracy and broaden its applicability for drought monitoring, agricultural management, and climate studies.
Download the pdf version here.
ALICE-LAB: Asian Land Information for Climate and Environmental Research Laboratory
Artificial Neural Network (ANN)
Modeled after the brain’s neural structure, ANNs are powerful tools in AI, excelling in tasks like hydrology and climate prediction by processing complex data patterns, from rainfall to extreme weather events, to enhance forecasting accuracy and resource management.
Introduction on Big Data
Big Data revolutionizes water and climate prediction by leveraging vast datasets from satellites, sensors, and historical records. Advanced analytics and machine learning improve prediction accuracy, aid in flood and drought management, and help assess climate change impacts, enabling informed decisions for a sustainable future.