Impact of climate change on SARS-CoV-2 epidemic in China

demonstrated the difficulty in controlling the epidemic owing to the lack of economic develop- ment and convenient transportation. Hubei Province recorded the highest number of confirmed cases, with 68,128 confirmed cases as of April 30, constituting 80.73% of the total, whereas confirmed cases in the other 32 provinces (excluding Tibet) constituted only 19.27%. This was followed by Guangdong, Henan, Zhejiang, and Hunan with 1,588, 1,276, 1,268, 1,037, and 1,019 confirmed cases, respectively, as of April 30, with a total of 6,188 confirmed cases, constituting 7.33%. The over- all number of confirmed cases was 58.92 per million of the population, and the average for each province was 51.47 per million people. The highest ratio was observed in Hubei (1,149.45 per million). Although the number of confirmed cases in Guangdong, Henan, and Hunan provinces was relatively high, the proportion of confirmed cases in the population was rela- tively low. By contrast, provinces which indicated value above the average value included Bei- jing (27.53 per million), Shanghai (26.85 per million), Heilongjiang (25.17 per million), Zhejiang (21.68 per million), and Jiangxi (21.68 per million), Jiangxi (20.08 per million), Chongqing (18.53 per million), and Hainan (17.78 per million). Notably, the total provincial population, GR, and confirmed cases were not sufficient to allow a unilateral assessment regarding the severity of the regional epidemic. Spatial and temporal variation characteristics of climatic factors. The long-term trends

of the six weather factors were calculated based on the Theil-Sen trend. Based on the results, only the SP indicated a downward trend, whereas the remaining five factors showed an upward trend, as shown in Table 2. The statistical indicators for each factor are shown in S1 Table. Among them, T and UV showed a significant upward trend (P-value < 0.05), whereas the WS and TP fluctuated significantly, although the trend was not significant (P-value > 0.05). The values of T and UV were 0.15˚C/day and 0.56 kJ�m-2�h-1/day, respectively. The long-term change trend indicates that the GR and SP decreased simultaneously, whereas the GR and the other five factors showed the opposite trend; the statistics of each province are shown in S4 Fig. The spatial distribution and cluster characteristics of each factor are shown in Fig 6. H and T showed consistent cluster characteristics, with HH and LH clusters in Guangdong and Hei- longjiang, respectively. UV and WS showed opposite cluster characteristics, with an LL cluster in Yunnan and an HL cluster in Xinjiang for UV, and a LL cluster in Xinjiang and an HL clus- ter in Sichuan for WS. In terms of the SP, Xinjiang was an LL cluster, Shandong was an HH cluster, Liaoning and Tianjin were LH clusters, and Yunnan and Qinghai were HL clusters. Relationship between climatic factors and SARS-CoV-2 Correlation of SARS-CoV-2 with weather factors. Among the correlations between UV and the other five factors, all were significant (P-value < 0.05), except for WS, whereas UV

indicated the lowest correlation with WS (RWS = 0.13). UV was significantly negatively corre- lated with GR, TP, and SP, whereas the other parameters were positively correlated with them. Table 2. Theil-Sen change trend of various weather factors. H T TP UV WS SP CR (unit/day) 2.59E-03 0.15 0.01 0.56 4.30E-03 -0.01 Significant - * - * - - Note: * represents P-value < 0.05. The Theil-Sen change rate describes a long time series that differs from the growth rate (GR). The GR represents the change rate of two time points, which can be regarded as the short-term change rate. https://doi.org/10.1371/journal.pone.0285179.t002 PLOS ONE Impact of climate change on SARS-CoV-2 epidemic in China PLOS ONE | https://doi.org/10.1371/journal.pone.0285179 July 27, 2023 10 / 21