A Comprehensive Evaluation Model for Local Summer Climate Suitability under Global Warming: A Case Study in Zhejiang Province

: In the context of global warming, how to measure summer climate suitability at a local scale is important for meteorological services. Considering meteorological and ecological condi ‐ tions, body comfort, and the atmospheric environment, an assessment method for summer climate suitability for Zhejiang Province is proposed. In this paper, a summer suitable index (SSI) for Zhejiang is calculated, including four secondary indices: a summer cool index (SCI), a comfort days index (CDI), a good air days index (GADI) and a vegetation cover index (VCI). Using a local evalu ‐ ation criterion, summer climate suitable areas are distinguished objectively according to the SSI. The results show that especially suitable regions account for 4.97% of Zhejiang province, very suit ‐ able regions account for 22.2%, suitable regions account for 39.58%, and general regions account for 33.25%. The summer climate suitable areas are located mainly in high mountains and hills and coastal island areas while plain areas cannot be considered a suitable destination for summer tour ‐ ism. By comparison and discussion, the SSI is demonstrated to capture summer climate suitability well. In contrast to a fixed evaluation index, benchmark values obtained for the SSI depend on the local climate and the index is straightforward to apply.


Introduction
With global warming, high impact extreme weather and climate events are now happening frequently in China [1][2][3]. Extreme heat waves have become one of the main meteorological events affecting people's work and life [4,5]. High temperatures can cause sweating, increase electrolyte metabolism and affect brain activity. They can also cause a variety of heat-related conditions, such as heat stroke, heat sickness, heat spasm, cardiocerebrovascular disease, etc. [6,7]. Thus, in hot summer, it is necessary to find cool and suitable places as outdoor travel destinations. A suitable summer resort can ensure that tourists are not only healthy but are also happy. The tourism economy and summeravoiding economy have gradually become of major concern for urban development, attracting significant attention from society and government.
Global warming is one of the most challenging global issues [8][9][10]. However, because of the multiple scales of climate change effects, climates and natural systems at different spatiotemporal scales show different change characteristics [11][12][13]. A key scientific issue in ecometeorological research is how to objectively evaluate the suitability of summer climate as a basis for recommending suitable summer tourism city and scenic spot destinations [14,15]. Wu and Ge [16] analyzed Hainan's climate comfort and its correlation with tourist flow applying the comfort index of W. H. Terjung. It was found that temperature-dominated climate comfort was the main factor influencing Hainan's tourist flow and tourist decisions. Lin [17] used the tourism climate index (TCI), physiological equivalent temperature (PET) and the climate tourism information scheme (CTIS) to analyze the climate comfort of 25 summer cities in northern China between 1980 and 2009 and simulated the maximum metabolic rate of the human body in summer cities. Hou et al. [18] established a summer weather index based on climatic analysis results for the daily mean temperature, maximum temperature, precipitation, sunshine duration and relative humidity of 14 summer tourism scenic spots of Liaoning Province from 2006 to 2015, dividing the conditions into four levels to provide a comprehensive evaluation of the suitability of summer tourism in Liaoning Province.
The climate of Zhejiang, located in east China, is controlled by subtropical monsoon. The summer of Zhejiang begins from late May or early June. The early summer is mainly characterized as Meiyu season, while the midsummer is hot and dry [19][20][21]. With global warming and urbanization, annual summer high temperature days and maximum temperature have shown an increasing trend according to data from most meteorological stations in Zhejiang [22]. Heat waves have occurred frequently in recent years. For example, a very serious heat wave affected Zhejiang in the summer of 2013; the recorded average temperature, high temperature days and extreme maximum temperature all broke records kept since 1951. The highest temperature recorded during the 2013 heat wave was 44.1 °C in Shaoxing city. Previous studies have shown that extreme weather and climate events will become more and more serious in the context of global warming, particularly heat waves and rainstorms [23][24][25]. Hence, both society and government are paying attention to the need for summer meteorological services in Zhejiang Province.
Although the summer climate has been analyzed and evaluated in some regions of China [26][27][28], the evaluation indices used cannot be applied directly to Zhejiang because of the great climate differences at different spatiotemporal scales. There has been limited study of the Zhejiang summer climate, and its characteristics and variation, such as high temperature duration, high temperature intensity and summer body comfort require to be clarified. It is therefore important that more scientific research, including evaluation of suitable summer climate conditions, is undertaken to provide a reference basis and guidance for further development of the summer tourism industry. Based on previous research, this paper reports the development of a comprehensive evaluation system for summer high temperature persistence, high temperature intensity, body comfort and other climatic elements, and estimates the distribution of summer climate suitability in Zhejiang Province.

Data and Methodology
A multi-source grid product was obtained from the China Meteorological Administration CMA Land Data Assimilation System (CLIDAS), published by the National Meteorological Information Center (NMIC) of CMA. The time span was 2008-2011, the time resolution was hourly and the spatial resolution was 0.062° × 0.0625° [29]. In addition, air quality data were obtained from the Department of Ecological Environment of Zhejiang Province, and vegetation coverage data from moderate resolution imaging spectroradiometer (MODIS) remote sensing with a spatial resolution of 250 m × 250 m; the data formats were unified using Kriging interpolation [30]. In this paper, the summer was taken to include only July and August (62 days), because June is a rainy month and is not hot in Zhejiang [31]. In meteorological and touristic studies, calculation of a meteorological index is a common way to estimate the climate suitability [32][33][34]. In this paper, four elements which summer tourists care most about were selected. According to Equation (1), the summer suitable index was calculated as follows: where SSI is the summer suitable index, SCI is the summer coolness index, CDI is the comfort days index, GADI is the good air days index, and VCI is the vegetation coverage index. A total of 20 meteorological experts from different cities were invited to estimate the local summer climate suitability. Weights were set according to the importance of each element to summer resorts and were comprehensively scored by these experts. W1, W2, W3, and W4 represent the weight scores for each index, respectively (Table 1).
where DBCMI4, DBCMI5, and DBCMI6 are the days where the body comfort meteorology index (BCMI) reaches 4, 5 and 6 grads, respectively. Dall is the summer total days and is taken to be 62 in this paper. β1, β2 and β3 are the weight coefficients for DBCMI4, DBCMI5, and DBCMI6, respectively. β1 and β3 are 0.8, and β2 is 1.0. In Equation (4), t is the air temperature, rh is the relative humidity, and v is the wind speed. The details of the levels are shown in Table  2. 3. Good air days index (GADI) where DAQIY and DAQIL are the days the air quality index (AQI) reached excellent and good grades, respectively [35]. Dall is the total days and is defined as 62 in this paper. W1 and W2 are the weight coefficients for DAQIY and DAQIL, being 1.0 and 0.8, respectively.

Vegetation coverage index (VCI)
The VCI refers to the percentage of vertical projection of the vegetation canopy in the cross-sectional area of the statistical area. It is an important index for measuring the regional ecological environment. This study calculated VCI according to the pixel dichotomy method [36].
where NDVI refers to the normalized difference vegetation index and data from the MODIS monthly synthetic products; NDVIsoil represents the value at which the pixel is pure bare soil, and NDVIvegetation represents the value at which the pixel is pure vegetation.
According to the characteristics of vegetation in China, the NDVIsoil and the NDVIvegetation values in this study were 0.05 and 0.95, respectively.

SCI Pattern in Zhejiang Province
More than 25 high temperature days were recorded in northern and central Zhejiang, corresponding to the Hangjiahu Plain and the Jinqu Basin. Fewer than 10 high temperature days were mainly recorded in northwest Zhejiang and southern Zhejiang, where the Tianmu Mountain, Yandang Mountain and other mountain areas are located. High temperatures (≥35 °C) of more than 140 h in Zhejiang Province occurred in northern Zhejiang and southern central Zhejiang, with the observed range being wider than for the high temperature days. Daily maximum temperatures over 39 °C were mainly distributed in the north of Zhejiang Province and the south of central Zhejiang Province.
Based on the high temperature days, high temperature hours and daily maximum temperature, the SCI was calculated according to Equation (2). Overall, south Zhejiang was cooler than north Zhejiang, and the eastern coast was relatively cooler (Figure 1), which is consistent with the distribution of high temperature days, high temperature hours and maximum temperature. The SCI was over 0.75 in the high mountains and coastal areas, and below 0.40 in the Hangjiahu Plain and the Jinqu Plain.

CDI Pattern in Zhejiang Province
The comfort days index (CDI) is an important index of summer climate suitability; the comfort of the human body is closely related to the local temperature, humidity and wind conditions. Based on meteorological observation data, the daily average temperature, relative humidity and wind speed in Zhejiang for July and August from 2008 to 2017 was analyzed. Mean temperatures of more than 39 °C were mainly distributed in northern Zhejiang and southern central Zhejiang, similar to the distribution of high temperature days in Zhejiang. The areas with average relative humidity over 80% were mainly distributed in the mountainous areas of northwest Zhejiang and the east coast. Areas of average relative humidity below 80% were mainly distributed in the central region of Zhejiang province. Due to differences in the underlying surface, the sea surface wind speed was significantly higher. The regions with an average wind speed of more than 1.0 m/s were mainly distributed in the coastal area of eastern Zhejiang, while the regions where the average wind speed was lower than 0.5 m/s were mainly distributed in the southwestern mountainous area.
According to Equations (3) and (4), based on the above indices, CDI was calculated and analyzed as shown in Figure 2. Recorded CDI values were relatively higher in the northwest and southwest of Zhejiang. In other regions, such as the central and northern areas and the southeast coastal region, due to higher average temperature, lower relative humidity and faster wind speed, comfortable days, corresponding to BCMI index grades 4-6, were less. In general, the high-altitude area of Zhejiang was more comfortable, with a CDI value of more than 0.5. Other regions were relatively low.

GADI Pattern in Zhejiang Provinc
To determine the climate distribution of the GADI in summer, we counted the days the air quality index (AQI) reached an excellent or good grade for July and August in Zhejiang, and then calculated the GADI according to the weight of the different air grades, applying Equation (5). Overall, the southern Zhejiang area had better air quality than the northern area; in northern Zhejiang, the recorded GADI values near coastal and riverine areas were higher than for the inland areas. Comparing areas with different elevations, the GADI values at higher elevations were higher than for the plain areas. The regions with fresher air were located in the mountainous areas of southern Zhejiang and the coastal areas of Taizhou, with GADI values above 9. Recorded GADI values in northern Zhejiang were relatively lower, especially in Shaoxing.

VCI Pattern in Zhejiang Province
The spatial distribution of vegetation coverage in Zhejiang province was obtained using remote sensing satellite data. Most areas had more than 50% vegetation coverage, though some plain and coastal areas had a vegetation coverage of less than 50%. On this basis, we used GIS to calculate the average vegetation coverage of villages in Zhejiang province according to Equation (6), and normalized the VCI of Zhejiang province in summer, as shown in Figure 3.

SSI Pattern in Zhejiang Province
Based on the meteorological observation data, satellite remote sensing data and geographic information data, a comprehensive evaluation model for the climate conditions of summer was established as Equation (1), and the distribution of SSI in Zhejiang Province was drawn according to a comprehensive scoring method, as shown in Figure 4.  Table 3. The especially suitable regions (90 ≤ SSI) accounted for 4.97% of Zhejiang province, the very suitable regions (75 ≤ SSI < 90) accounted for 22.2%, the suitable regions (60 ≤ SSI < 75) accounted for 39.58%, and the general regions (SSI < 60) accounted for 33.25%. The summer suitable climate areas of Zhejiang province were mainly located in the high mountains and hills and the coastal island areas, while for plain areas, suitable summer climate areas were not common.
Based on climate regionalization, the summer climate resorts were evaluated and ranked using the SSI values from 2000 meteorological observation stations in Zhejiang province. According to the analysis, the summer climate resorts in Zhejiang province were mainly located in the mountains and coastal island areas (Figure 4), which is consistent with the evaluation results of existing studies and reports by the mass media. The number of summer climate resorts varied greatly among different cities, being closely related to the richness of the local climate resources.

Discussion and Conclusions
In the context of climate change, the responses to global warming are inconsistent across different regions. Huang et al. [37] found that surface air temperature trends were particularly enhanced in the boreal cold season over semi-arid regions between 1901 and 2009, with a drier and warmer trend in semi-arid regions. Local climatic conditions were also affected by the atmospheric circulation. For example, changes in the intensity of the Hadley circulation affected the vertical transport of water vapor in tropical regions, influencing water cycle processes and the distribution of precipitation [38]. The evolution trends of the Hadley circulation have obvious regional and seasonal characteristics, and are one of the reasons for regional climate differences [39][40][41]. In addition, the surrounding geographical environment and human social activities also play an important role in regional climate conditions; therefore, assessment of climate suitability requires consideration of local conditions.
In the past, limited by the regional area, land and sea differences and the topographic factors of Zhejiang Province, previous indicators have not been suitable for the evaluation of the climate suitability of summer resorts. Therefore, it was necessary to comprehensively consider the issues and to develop a set of indicators suitable for Zhejiang Province. In this regard, by analysis of the SCI, CDI, GADI and VCI, and considering the situation of Zhejiang Province, we proposed the SSI to evaluate the climate suitability of summer vacation destinations according to local indicators. In contrast to fixed evaluation indices, the benchmark values for the SSI depend on local climate, and it can easily be applied in different locations.
However, it should be noted that global warming is becoming more and more intense-the warming trend in China has shown no signs of abating in recent years. Hot summers will become even more frequent in the future [42]. Several regions in China show different temperature trends. From 1901 to 2015, the homogenized annual mean surface air temperature exhibited a range of trends: between 1.1 °C and 4.0 °C/century in northeastern China, between 0.4 °C and 1.9 °C/century in southeastern China, and between 1.4 °C and 3.7 °C/century in western China to the west of 105 ° E [43]. Cheng et al. [44] found that, in eastern China, the four temperature extremes (warm days, cold days, warm nights and cold nights) for winter have decreased almost everywhere, while those for summer have decreased in the north but increased in the south. Therefore, under global warming, whether the summer assessment system for particular regions needs to be updated synchronously to adapt to the background of global warming needs to be considered. It is necessary to check the summer climate resorts in Zhejiang Province at regular intervals of several years.
Finally, the original intention of the evaluation and recommendation of summer resorts was to provide information to support people when choosing summer resorts, and to make it more straightforward for them to be able to identify travel destinations. At its present stage of development, the evaluation model uses coolness and comfort as the main scoring criteria, and can capture summer climate suitability well. With continuous improvement in living standards, the requirements for a suitable ecological environment and air quality will also increase, so the weighting of relevant scoring items may become more important in the evaluation and recommendation of summer resorts in the future. In addition, due to the limitations of observation and data processing, whether it will be necessary to add new evaluation indicators to make the evaluation more systematic and comprehensive remains to be tested.