The impact of human activities on the heat island effect - A case study of annual activities Ren-De Chiu 1, Yi-Shiang Shiu 1, Re-Yang Lee 1, Tsu-Chiang Lei 1 1 Feng Chia University,No. 100, Wenhwa Rd., Seatwen, Taichung, Taiwan 40724, Email:autumn_left@hotmail.com, ysshiu@fcu.edu.tw, tclei@fcu.edu.tw, rylee@fcu.edu.tw KEY WORDS:Remote sensing, Landsat, heat island effect ABSTRACT: Nowadays the rising of temperature makes global warming and climate change become very popular topics. Previous studies pointed out that human activity is one of the causes of the heat island effect. In Taiwan, more and more seasonal tourism activities are held in recent years along with the improvement of the life quality. However, such activities attract a lot of crowds and traffic flow making the heat island intensity change. In order to monitor this effect, this study uses thermal infrared channels of Landsat series imagery and retrieves the ground temperature based on the methodology proposed by Artis and Carnahan (1982) and Weng et al. (2004). Two seasonal activity cases, Da Hu strawberry festival in Miao Li County and Xinshe flowery field in Taichung City, are chosen to monitor the change of heat island effect at different periods. With supervised maximum likelihood classification, land features are divided into five categories: water, soil, concrete, forest and grassland; and then the emissivity values are assigned to each category. The results show that the relatively high temperature regions are the areas with: (1) intensive construction; (2) frequent human activities; (3) intensive traffic flow. In addition to the aforementioned regional comparison, we also join the statistical data with the number of visitors in different years to explore the relationship between the intensity of heat island and the visitors. This study investigates and shows the impact of seasonal events on the heat island effect. We conclude that transient human activities in non-urban areas rather than the urban should also be monitored and the ecological impact has to be considered by the activity organizer. 1. INTRODUCTION People demand for leisure and quality of life improved. The marketing of agricultureincreases a lot of related activities and attractslots of people.however, land cover changes a lot after the activities. For example in Taiawn, Da Hu strawberry festival in Miao Li County is popular with strawberry cultivating, and Xinshe flowery is also famous in Taichung City. Events held every year causes changes of land cover. In recent years, excessive development and climate change result in significant impact, such as global warming, land subsidence and landslides.so people began to pay attention to climate change and urgent environmental issues. However, development of the industry is bound to bring environmental impact. Human development has changed the original landscape and vegetation. The land without green cover is one of the causes of disasters. The changes of landscape make a lot of influence, and the most unnoticed is the heat island effect. Urban heat island is mainly due to the artificial impermeable pavement used to replace the original ground surface. The replacement has caused unbalanced temperature with higher temperature in urban areas than the rural ones(ho et al, 2011). As the development of thermal remote sensing and spatial information technology, the temperature retrievalmethods have significantlydeveloped, such as mono-window algorithm(qin et al, 2001),
split-windowalgorithm (Jimenez-Munoz et al. 2009; Jimenez-Munoz et al. 2014) and S-SEBI algorithm (Roerink et al., 2000). Finally, combined with many features of activities, they are consistent with the causes of the heat island effect. Therefore, this study combined the characteristics of remote sensing and thermal infrared band of Landsat-8 satellite images and used temperature retrieval to investigate the temperature of study areas. The relationship between temperature and different land use was discussed based on the surface temperature during the activity season and non-season periods. 2. RELATED WORK 2.1Study Area There are two areas in the study. First area is set to Xinshe district, Taichung City. The annual flower sea event venue and the surrounding area for the study area are shown in Figure 1. Figure 2 shows that the second area is set to Dahu Township, Miaoli County. Annual strawberry season is opened in the Dahu winery. Therefore, this study chose strawberry cultivation areas around the winery and compared with other areas in Dahu Township. Figure 1Xinshe District, Taichung City Figure 2 Dahu Township, Miaoli County
2.2Landsat Imagery Landsat series still working are Landsat-7 and Landsat-8. Landsat-7 began its mission in April 1999, while the Landsat-8 began in February 2013.After 2003, Landsat-7 has encountered malfunction. Because all of the seasonal activities we chose are after 2003, this study used Landsat-8 images. Details of each band of Landsat-8 are shown in Table 1. Table 1Details about each bands of Landsat-8 Spectral bands Wavelength Resolution (micrometers) (meters) Band 1-coastal/aerosol 0.43-0.45 30 Band 2-blue 0.45-0.51 30 Band 3-green 0.53-0.59 30 Band 4-red 0.64-0.67 30 Band 5-near IR 0.85-0.88 30 Band 6-SWIR 1 1.57-1.65 30 Band 7-SWIR-1 2.11-2.29 30 Band 8-panchromatic 0.50-0.68 15 Band 9-cirrus 1.36-1.38 30 Band 10-TIRS 1 10.60-11.19 100 Band 11-TIRS 2 11.50-12.51 100 Source: Center for Space and Remote Sensing Research, National Central University http://www.csrsr.ncu.edu.tw/08csrweb/chinver/c6techsupp/optical/landsat.php 2.3 Limitations of the Study In this study, research topic is annual activities. Therefore, images are limited to time. Landsat images are taken every 16 days in the same area, and we must choose study area without the obstruction of cloud to prevent the emissivity being disturbed. Accordingly, Table 2 shows the date of image. The range covers two paths, which is shown in Table 3. Table 2Date of image Annual activities Date of season Date of non-season Xinshe s flower sea 2014/12/06 2014/10/19 Dahu s strawberry season 2015/02/01 2014/10/19 Table 3Path/Row Date Time Path/Row 2014/10/19 02:27:14 118/043 2014/12/06 02:27:09 118/043 2015/02/01 02:20:46 117/043
3 RESEARCH DESIGN 3.1 Framework This framework can be divided into five processes (Figure 3), the first part is data collection, the second part will be pre-processed map data, to prepare for the supervised classification, and the third part is select training samples. They are water, cloud, impermeable surface, soil and vegetation. And in fourth part, give radiance to each classification (Table 4).Then, running the temperature inversion formula (as shown in Formula 1) to get the final results. At last, we calculated heat island intensity (as shown in Formula 2) to present heat island intensity change. Figure3 Flow chart Table 4Emissivity of each classification Classification Emissivity Water 0.95 Impermeable surface 0.8 Soil 0.92 Vegetation 0.98 Cloud 0.95 Source: Jensen, John R.,2007
3.2 Methodologies 3.2.1Temperature Inversion Formula b b/*1 + [10.8 ] log(a)} (1) 14380 a= Emissivity, b=band 10 3.2.2 Heat Island Intensity (2) The mean temperature in the study area, = The lowest temperature in the study area 4.RESULTS AND DISCUSSION 4.1 Results of Xinshe District As can be seen from Figure 4 at the time of the season, soil and vegetation distribution is average, the site survey, soil position as a car park use and food area is arranged, and in Figure 5 you can see the range of non-seasonal flowers, surface covered with soil segment most abundant. After an investigation, soil distribution area is parking and food fair, and in Figure 5 you can see the range of Flowers Sea, the largest distribution of land covered with soil. Comparing Figure 6 and Figure 7 can be found in surface temperature varies with the land cover will be different, temperature of soil obviously higher than the surrounding vegetation in Figure 6,and temperature of impermeable surface higher than the soil. Presented by Table 5, you can easily see heat island intensity of season is less than non-season. With the chart presentation, it can be reasonably presumed that, although annual activity attracts a lot of crowds and traffic flow, making the heat island intensity change is land cover change.
Figure 4 Classification of Xinshe s Flower Sea, Season Figure 5 Classification of Xinshe s Flower Sea, Non-season
Figure 6Surface Temperature of Xinshe s Flower Sea, Season Figure 7Surface Temperature of Xinshe s Flower Sea, Season Table 5Heat Island Intensity of Xinshe s Flower Sea Area Surface Temperature ( ) Heat Island Intensity Season (2014/12/06) Non-Season (2015/02/01) Flower Sea 23 Xinshe District 14.5 Flower Sea 26.1 Xinshe District 14.7 8.5 11.4
4.2 Results of Dahu Township We can observe the same situation with Flower Sea from Figure 8, strawberry cultivation areas mostly distributed in the soil during the non-season. In the season is based on vegetation mostly, as shown in Figure 8 and Figure 9. In temperature calculation results (Figure10), the temperature of impermeable surface is higher than the soil, same as Xinshe District, and the lowest temperature is vegetation. Accordingly, we can get a conclusion, the impact of land cover significantly more than human activity. Figure 8Classification of Dahu s Strawberry Cultivation Area, Season Figure 9Classification of Dahu s Strawberry Cultivation Area, Non-season
Figure 10Surface Temperature of Dahu s Strawberry Cultivation Area, Season Figure 11Surface Temperature of Dahu s Strawberry Cultivation Area, Non-season Table 6Heat Island Intensity of Strawberry Cultivation Area Area Surface Temperature ( ) Heat Island Intensity Season (2015/02/01) Non-Season (2014/10/19) Strawberry Cultivation Area 20.7 Dahu Township 11.3 Strawberry Cultivation Area 31.9 Dahu Township 19.3 9.4 12.6
5. CONCLUSIONS This study confirmed that the temperature of vegetation area is lower than soil area. And the impermeable surface has the highest temperature in study area. For example, the temperature of densely built area is higher than the vegetation area. Although there will be many visitors during the activity, but the season of heat island intensity lower than non-season. While human activities may cause heat island effect, the change of land use is still the mainly impact of the heat island effect. 6. REFERENCE 1. Ho, C.W.,2011. The study of Taichung area land use change on heat island effect. Journal of Photogrammetry and.remote Sensing, 16(2), pp. 139-149. 2. Jensen, J. R., 2007, Remote Sensing of the Environment: An Earth Resource Perspective, 2nd Ed., Upper Saddle River, NJ: Prentice Hall, 592 pages. 3. Jimenez-Munoz,J. C., Cristobal, J., Sobrino, J. A., Soria, G., Ninyerola, and M., Pons, X., 2009. Revision of the Single-Channel Algorithm for Land Surface Temperature Retrieval from Landsat Thermal-Infrared Data. Geoscience and Remote Sensing, IEEE Transactions on, 47(1), pp. 339-349. 4. Jimenez-Munoz, J. C., Sobrino, J. A., Skokovic, D., Mattar, C., and Cristobal, J., 2014. Land Surface Temperature Retrieval Methods from Landsat-8 Thermal Infrared Sensor Data. Geoscience and Remote Sensing Letters, IEEE, 11(10), pp. 1840-43. 5. Qin, Z. H., Zhang, M. H., Karnieli, A. and Berliner, P., 2001. Mono-window Algorithm for Retrieving Land SurfaceTemperature from Landsat TM 6 data. ActaGeographicaSinica, 56(4), pp. 456-466. 6. Roerink, G., Su, Z., and Menenti, M., 2000. S-SEBI: A simple remote sensing algorithm to estimate the surface energy balance. Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere, 25(2), pp. 147-157. 7. Roger. K, Thayne. M, 1998. Urban tree transpiration over turf and asphalt surfaces. Atmospheric Environment, 32(1), pp. 35-41. 8. Weng, Q., Lu, D., and Schubring, J., 2004. Estimation of land surface temperature vegetation abundance relationship for urban heat island studies. Remote Sensing of Environment, 89(4), pp. 467-483.