Estimation of Hourly near Surface Air Temperature Across Israel Using an Ensemble Model

Victor Novack,Alexandra Shtein,Allan C Just,Bin Zhou,Jonathan Rosenblatt,Evyatar Erell,Itai Kloog,Ian Hough

doi:10.3390/rs12111741

Abstract

Mapping of near-surface air temperature (Ta) at high spatio-temporal resolution is essential for unbiased assessment of human health exposure to temperature extremes, not least given the observed trend of urbanization and global climate change. Data constraints have led previous studies to focus merely on daily Ta metrics, rather than hourly ones, making them insufficient for intra-day assessment of health exposure. In this study, we present a three-stage machine learning-based ensemble model to estimate hourly Ta at a high spatial resolution of 1 × 1 km2, incorporating remotely sensed surface skin temperature (Ts) from geostationary satellites, reanalysis synoptic variables, and observations from weather stations, as well as auxiliary geospatial variables, which account for spatio-temporal variability of Ta. The Stage 1 model gap-fills hourly Ts at 4 × 4 km2 from the Spinning Enhanced Visible and InfraRed Imager (SEVIRI), which are subsequently fed into the Stage 2 model to estimate hourly Ta at the same spatio-temporal resolution. The Stage 3 model downscales the residuals between estimated and measured Ta to a grid of 1 × 1 km2, taking into account additionally the monthly diurnal pattern of Ts derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) data. In each stage, the ensemble model synergizes estimates from the constituent base learners—random forest (RF) and extreme gradient boosting (XGBoost)—by applying a geographically weighted generalized additive model (GAM), which allows the weights of results from individual models to vary over space and time. Demonstrated for Israel for the period 2004–2017, the proposed ensemble model outperformed each of the two base learners. It also attained excellent five-fold cross-validated performance, with overall root mean square error (RMSE) of 0.8 and 0.9 °C, mean absolute error (MAE) of 0.6 and 0.7 °C, and R2 of 0.95 and 0.98 in Stage 1 and Stage 2, respectively. The Stage 3 model for downscaling Ta residuals to 1 km MODIS grids achieved overall RMSE of 0.3 °C, MAE of 0.5 °C, and R2 of 0.63. The generated hourly 1 × 1 km2 Ta thus serves as a foundation for monitoring and assessing human health exposure to temperature extremes at a larger geographical scale, helping to further minimize exposure misclassification in epidemiological studies.

Highlights

The last two decades have witnessed an increase in global mean temperature and episodes of extreme temperatures, which is attributed mainly to altered patterns in atmospheric circulation and in sea surface temperature caused by human-induced climate change [1,2]
As the frequency and magnitude of these extreme events is expected to increase in the future [3,4], their environmental consequences raise serious public health concerns globally, given that exposure to temperature extremes is well associated with mortality and other adverse health effects [5,6,7,8]
Conventional measurements at 2 m height from scattered weather stations are inadequate for this purpose, as they are incapable of capturing the spatio-temporal variability of temperature within a large area, especially in cities, where the urban heat island effect could considerably modify the local micro-climate [9,10,11]

Summary

Introduction

The last two decades have witnessed an increase in global mean temperature and episodes of extreme temperatures, which is attributed mainly to altered patterns in atmospheric circulation and in sea surface temperature caused by human-induced climate change [1,2]. Conventional measurements at 2 m height from scattered weather stations are inadequate for this purpose, as they are incapable of capturing the spatio-temporal variability of temperature within a large area, especially in cities, where the urban heat island effect could considerably modify the local micro-climate [9,10,11]. Studies based on such sparse and spotty data introduce exposure error/misclassification and likely underestimate the true effect [12]. Physically based numerical models, such as the Weather Research and Forecasting (WRF) model, can simulate and project Ta at different spatio-temporal resolutions under current and future scenarios [13], they normally entail high-level expertise, regional knowledge, and resource-intensive computing infrastructure, restricting their applications as a global practice in epidemiological studies

Methods

Discussion

Conclusion