Downscaling climate projections to 1 km with single-image super resolution
This work addresses the need for high-resolution climate data for local decision-making, but it is incremental as it applies existing super-resolution methods to a new domain without a major methodological breakthrough.
The paper tackled the problem of low spatial resolution in climate projections by using single-image super-resolution models to downscale them to 1 km resolution, achieving results where the downscaled projections did not increase error in climate indicators compared to low-resolution ones.
High-resolution climate projections are essential for local decision-making. However, available climate projections have low spatial resolution (e.g. 12.5 km), which limits their usability. We address this limitation by leveraging single-image super-resolution models to statistically downscale climate projections to 1-km resolution. Since high-resolution climate projections are unavailable for training, we train models on a high-resolution observational gridded data set and apply them to low-resolution climate projections. We propose a climate indicator-based assessment using observed climate indices computed at weather station locations to evaluate the downscaled climate projections without ground-truth high-resolution climate projections. Experiments on daily mean temperature demonstrate that single-image super-resolution models can downscale climate projections without increasing the error of climate indicators compared to low-resolution climate projections.