In the arid western United States, accurate water supply predictions are crucial for managing resources and planning for the future. A recent study published in Water Resources Research, the English translation of the journal name, has shown promising results in improving these predictions using next-generation satellite data. The research, led by Sean W. Fleming of the US Department of Agriculture’s Natural Resources Conservation Service, could have significant implications for various sectors, including energy, which relies heavily on consistent water supplies for cooling and generation processes.
Fleming and his team tested a new space-based remote sensing product, called spatially and temporally complete (STC) MODSCAG fractional snow-covered area (fSCA), as input for the Natural Resources Conservation Service’s operational US West-wide water supply forecast (WSF) system. The study focused on four test watersheds: the Walker and Wind Rivers in California and Wyoming, the Piedra River in Colorado, and the Gila River in New Mexico.
The results were encouraging. “We found that fSCA-enabled accuracy gains are most consistent and explainable for short-lead, late-season forecasts,” Fleming explained. “This is particularly useful in operational practice, as snowlines often rise above in situ measurement sites, making it challenging to gather accurate data.”
The improvements in accuracy ranged from 10% to 25%, depending on the watershed and the time of year. The gains were most significant in areas where spring-summer runoff is dominated by snowmelt and where in situ networks struggle to monitor late-season snowpack.
The study also compared the performance of statistical and AI-based hydrologic models. The AI-based model generally outperformed the statistical model and, in some cases, better capitalized on satellite remote sensing data.
So, what does this mean for the energy sector? Accurate water supply predictions are vital for energy companies, which often rely on water for cooling and generation processes. Improved forecasts can help these companies plan for the future, manage resources more effectively, and potentially reduce costs.
Moreover, the study suggests that combining satellite and in situ products in data-driven hydrologic models could help guide new SNOTEL site selection. This could lead to more efficient and effective monitoring of snowpack, benefiting not just the energy sector, but also agriculture, recreation, and other industries that depend on water resources.
Looking ahead, this research could shape future developments in the field by demonstrating the value of integrating satellite data with traditional in situ measurements. As Fleming noted, “This approach could be particularly useful in sparsely monitored regions, where satellite data might be the only reliable source of information.”
The study also highlights the potential of AI and machine learning in hydrologic modeling. As these technologies continue to advance, they could revolutionize the way we predict and manage water resources.
In the meantime, the energy sector and other stakeholders can start exploring how to incorporate these findings into their own operations. By doing so, they can help ensure a more sustainable and resilient future for all.