New land use regression model to estimate atmospheric temperature and heat island intensity in Taiwan

Chin Yu Hsu, Uen Ching Ng, Chiao Ying Chen, Yu Cheng Chen, Mu Jean Chen, Nai Tzu Chen, Shih Chun Candice Lung, Huey Jen Su, Chih Da Wu

Research output: Contribution to journalArticlepeer-review

6 Citations (Scopus)


This paper is about spatial–temporal variability of atmospheric temperature across Taiwan, an island with diverse local emission sources partly because of its Asian cultural characteristics. To develop a new land use regression (LUR) model for this study, we used the temperature data collected by the Taiwan Central Weather Bureau from 2000 and 2016, while using the data from 2017 as the external data verification to assess model reliability. Because incense and joss money burning is a cultural-specific emission source in Asia, we further included location of temples, cemeteries, and crematoriums as potential predictors. The overall model performance and tenfold cross-validated are R2 of 0.88 and R2 of 0.87, respectively, which presents a high level of prediction performance. Moreover, we used our LUR model to estimate urban heat islands intensity (UHII) for six metropolises in Taiwan and found Taichung City has the highest UHII value (4.60 °C) among them. These results provide important insights in expanding the remote sensing application on spatial–temporal variation of atmospheric temperature and its further application on UHI effects.

Original languageEnglish
Pages (from-to)1451-1459
Number of pages9
JournalTheoretical and Applied Climatology
Issue number3-4
Publication statusPublished - 2020 Aug 1

All Science Journal Classification (ASJC) codes

  • Atmospheric Science


Dive into the research topics of 'New land use regression model to estimate atmospheric temperature and heat island intensity in Taiwan'. Together they form a unique fingerprint.

Cite this