Geospatial Information Tools That Use Machine-Learning to Enable Sustainable Groundwater Management in West Africa
In West Africa, particularly in Niger, rain-fed agriculture is unsustainable. Groundwater development will provide a more stable water source and enhance food security. Groundwater development requires collecting and analyzing data produced at global and national levels and disseminating that data and knowledge to end users such as States, NGOs, municipalities, businesses, and agropastoralists in a format that is useful for planning and decision-making. We will develop a set of geospatial tools for stakeholders at all levels to assess and interpret groundwater data. Users will access these tools via a simple browser interface that requires minimal infrastructure or computer expertise. We will also assist hub partners in developing and using groundwater models.
A) Groundwater Resource Assessment
The Groundwater Level Mapping Tool (GW-Level) is the foundation of the system. GWLevel ingests in situ water level measurements and performs temporal and spatial interpolation to build interactive animated maps or times series plots. GW-Level uses machine learning to find correlations among Earth observation data, such as precipitation or soil moisture, with water level data and perform more accurate interpolation. This approach is particularly applicable to West Africa because it ensures that scarce in situ data are used as effectively and accurately as possible. These products help water managers gain a better understanding of groundwater resources and determine how aquifers are responding to groundwater development, droughts, and climate change.
The Groundwater Volume Analysis Tool (GW-Volume) provides lower resolution data that gives a current picture of groundwater conditions in terms of overall water storage volumes. GW-Volume can identify and characterize conditions in data-poor areas or identify trends in other regions often obscured by noise from well data. GW-Volume uses GRACE mission data to compute and display changes in water storage in a web-based mapping system and will integrate GRACE-FO data. GW-Volume uses NASA GLDAS surface water data to derive changes in groundwater. It displays the results as time series plots at selected points or map animations. This tool will complement the GW-Level tool, providing a secondary measure of how groundwater changes in space and time.
B ) Groundwater Model Development
We will work with SERVIR-West Africa to update and improve existing groundwater models and develop new regional models. We will implement workflows using data derived from Earth Observations as forcings to these models. Groundwater modeling typically requires advanced computer and software capabilities with trained and experienced modelers. We will develop the Groundwater Model Scripting Tool (GWModel) to support groundwater model simulations for selected scenarios that managers face on a routine basis. GW-Model uses a web-based interface to add or eliminate wells, change pumping rates, or evaluate changes such as drought (e.g., changing recharge rates) or temperature (e.g., changing irrigation needs). This tool will allow stakeholders to access and use models without having to rely on modeling experts. We have helped develop similar modeling tools for the Virginia Department of Environment Quality and the Utah Division of Water Resources.
C) Stakeholder Rraining and Technology Transfer
We will work with stakeholders, including AGRHYMET, The Ministry of Hydraulics and Sanitation (DGRE), the Association for the Redynamisation of Pastoralism in Niger (AREN), and the National Network of Chambers of Agriculture of Niger (RECA), to develop and provide training for these tools. We will leverage training to refine the capabilities, tool interfaces, visualizations, and data reports.
We are uniquely qualified for this project due to our extensive experience in web-based tool development, groundwater data management, groundwater modeling, and education and training.