PL1GD-T: a high-resolution gridded daily air temperature dataset for Poland

Abstract

This paper presents a high-resolution gridded dataset of daily minimum (TN), mean (TG), and maximum (TX) near-surface air temperatures over Poland, covering the period from 1951 to 2020, with a spatial resolution of 1 km2. The PL1GD-T dataset was developed using radial basis functions (RBFs), which were applied to quality-controlled observations from 347 ground weather stations at the Institute of Meteorology and Water Management - National Research Institute. TG is calculated consistently using the same formula throughout the entire period, and daily TN, TG, and TX fields are generated based on available daily records. Cross-validation methods evaluated the gridding procedure on a monthly basis. The linear RBF was selected by hold-out cross-validation (HO-CV) as the most suitable for the gridding procedure among other RBFs. The leave-one-out cross-validation (LOO-CV) was performed to ensure the ability to reproduce the original characteristics variability. The values of the scores averaged over all stations for individual months are in the range of 0.3-0.2, 0.3-0.2, and 0.1-0.2 (K) for the bias; in the range of 1.23-1.46, 0.69-0.92, and 0.84-0.99 (K) for the root-mean-squared difference (RMSD); and in range of 0.91-0.97, 0.98-0.99, and 0.98-0.99 for the correlation for TN, TG, and TX, respectively. The RMSD is clearly altitude-dependent, increasing from lowland to mountainous regions. The dataset's scope and resolution allowed for the robust estimation of local climate variability characteristics and observed trends. The availability of high-resolution datasets in both spatial and temporal contexts is essential for climate change impact analysis on a smaller scale. This new dataset provides a quality-validated, high-resolution, and open-access dataset that could be utilised by society, administrative bodies, or research institutions for climate-related applications. The dataset is publicly available from the repository of the Institute of Meteorology and Water Management - National Research Institute at 10.26491/imgw_repo/PL1GD-T (Jaczewski et al., 2024).