Summary
This paper describes the development and validation of HRLT, a high-resolution (1 km) daily gridded climate dataset for China spanning 1961–2019, interpolated from coarser meteorological administration data using machine learning, generalised additive models, and thin plate splines with topographic covariates. Validation against station observations demonstrated high accuracy for temperature variables (MAE 1.07–1.08 °C) and moderate accuracy for precipitation (MAE 1.30 mm), with performance comparable to or exceeding three existing datasets whilst offering superior spatial resolution or temporal span. The publicly available dataset is intended to support diverse climatological and agricultural studies requiring detailed long-term climate information.
UK applicability
This dataset is specific to China's geography and meteorological networks and has limited direct applicability to UK conditions. However, the interpolation methodology combining machine learning with topographic covariates may inform development of similar high-resolution climate datasets for the UK or other regions with sparse station networks.
Key measures
Mean absolute error (MAE), root mean square error (RMSE), Pearson's correlation coefficient, coefficient of determination (R²), and Nash–Sutcliffe modelling efficiency (NSE) for maximum temperature, minimum temperature, and precipitation estimates
Outcomes reported
The study produced a publicly available daily gridded dataset (HRLT) of maximum temperature, minimum temperature, and precipitation at 1 km spatial resolution covering 1961–2019 across China. Accuracy was assessed against meteorological station observations using multiple error metrics and comparative validation against three existing datasets.
Topic tags
Dig deeper with Pulse AI.
Pulse AI has read the whole catalogue. Ask about this record, its theme, or how the findings apply to UK farming and policy — every answer cites the underlying studies.