Geographically Weighted Machine Learning and Downscaling for High-Resolution Spatiotemporal Estimations of Wind Speed

Abstract

High-resolution spatiotemporal wind speed mapping is useful for atmospheric environmental monitoring, air quality evaluation and wind power siting. Although modern reanalysis techniques can obtain reliable interpolated surfaces of meteorology at a high temporal resolution, their spatial resolutions are coarse. Local variability of wind speed is difficult to capture due to its volatility. Here, a two-stage approach was developed for robust spatiotemporal estimations of wind speed at a high resolution. The proposed approach consists of geographically weighted ensemble machine learning (Stage 1) and downscaling based on meteorological reanalysis data (Stage 2). The geographically weighted machine learning method is based on three base learners, which are an autoencoder-based deep residual network, XGBoost and random forest, and it incorporates spatial autocorrelation and heterogeneity to boost the ensemble predictions. With reanalysis data, downscaling was introduced in Stage 2 to reduce bias and spatial abrupt (non-natural) variation in the predictions inferred from Stage 1. The autoencoder-based residual network was used in Stage 2 to adjust the difference between the averages of the fine-resolution predicted values and the coarse-resolution reanalysis data to ensure consistency. Using mainland China as a case study, the geographically weighted regression (GWR) ensemble predictions were shown to perform better than individual learners’ predictions (with an approximately 12–16% improvement in R2 and a decrease of 0.14–0.19 m/s in root mean square error). Downscaling further improved the predictions by reducing inconsistency and obtaining better spatial variation (smoothing). The proposed approach can also be applied for the high-resolution spatiotemporal estimation of other meteorological parameters or surface variables involving remote sensing images (i.e. reliable coarsely resolved data), ground monitoring data and other relevant factors.

A Robust Deep Learning Approach for Spatiotemporal Estimation of Satellite AOD and PM2.5

“…Residual connections have been effectively leveraged to improve learning efficiency in CNN [63,84], and multilayer perceptron [85,86]. Here, residual connections were introduced into the base models of autoencoder.…”

“…Residual connections provide the shortcuts from the encoding layer to the decoding layer, and these shortcuts boost the efficient backpropagation of errors from the deep decoding layers to the shallow encoding layers in a parallel way in the learning of the parameters [84,87]. Thus, the introduction of residual connections can effectively solve the potential issues of saturation of gradients and degradation of accuracy in the traditional neural network, as demonstrated in the fusion of the meteorological parameters [85,86]. For technical details, please see Supplementary Section S3.2.…”

“…The proposed approach was used in the spatiotemporal estimation of meteorological parameters in a considerably extensive region, i.e., mainland China with reliable performance [85,86]. In the future, the proposed approach will be used to predict PM 2.5 concentration at a high spatial (1 km) and temporal (daily) resolution for an extensive spatial (mainland China) and temporal (2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016) domain.…”

“…The method of the residual deep network was used for the interpolation of meteorological data. For details, please refer to Supplementary Section S2 and[85,86]. (4) Imputation of missing satellite AOD.…”

Deep Residual Autoencoder with Multiscaling for Semantic Segmentation of Land-Use Images

“…In this architecture, in addition to classic residual units within each encoding and decoding layer [36,37], residual connections are established through identity mapping shortcuts from the encoding layers to the corresponding decoding layers in a nested way. In the author's previous work, similar nested residual connections have been used in multilayer perceptrons (MLPs), resulting in considerably improved performance compared with nonresidual MLPs for regression and spatiotemporal predictions of meteorological parameters [38], aerosol optical depth and particulate matter of diameter <2.5 µm (PM 2.5 ) [39]. In this architecture, residual interconnections are established to lengthen the paths for the backpropagation of errors to improve the efficiency of residual learning, and multiscaling via ASPP or resizing is also incorporated to make the trained model robust to scale variations.…”