Based on the Landsat remote sensing images of 1986 and 2011, higher resolution terrain data and 2-D urban fraction values, the impact of land use and land cover change (LUCC) on the near-surface air temperature is investigated for Shenzhen by carrying out two sets of numerical experiments using the weather research and forecasting model (WRF) coupled with the Noah urban canopy model. Comparison with measured near-surface air temperatures of 40 ground-based atmospheric observatories in the region shows a good agreement between observed and simulated data for the simulation periods. The results indicate significant contributions of urban sprawl and accompanying LUCC to the near-surface air temperature. Simulations suggest that LUCC has caused average temperatures with an increase of up to 0.42°C in January and 0.91°C in July respectively. The conversion from natural vegetation covers into urban and built-up area results in a decrease in surface albedo and thereby enhances the surface net solar radiation and the sensible heat flux, which leads to the most obvious warming effect.

Hourly precipitation data of 1457 stations during 1980?2012 were used to investigate the spatial-temporal patterns of PCI (precipitation concentration index) from May to September over eastern and central China. The relationships between PCI and other precipitation variables based on different precipitation intensities or different precipitation durations were also analyzed to reveal the variables highly correlated with PCI. The result indicated that the values of PCI over eastern and central China ranged from 0.56 to 0.78 with evident regional variations. The values of PCI in southeastern China were higher than the values in northwestern China. There were three concentrations of high PCI value centers, i.e., Northeast China, North China, and Sichuan Basin. The variations of PCI were found in areas with great altitude difference. The linear trend analysis showed an increasing trend of PCI values over eastern and central China over the past 33 years. In addition, significantly positive correlation relationship between PCI and extreme precipitation was detected, and the correlation coefficient of the two variables increased with the decrease of the total precipitation amount.

Using the Landsat TM imagery and land use data of Shenzhen in Oct. 2009, this paper studied the relationship between land use structure and spatial differentiation of thermal environment, by land surface temperature (LST) retrieval, temperature zoning and overlay mapping. Besides, investigation of plots in three temperature partitions and analysis of landscape metrics were also used to study the relationship between landscape characteristics and LST. The results showed that, 1) there were significant variation in both land use structure and spatial pattern of different temperature partition. The landscape of mid-temperature partition was highly fragmented and patch shape was most complex. In cooling-island partition, heterogeneity of landscape was low and patch shape were relatively simple with an aggregated structure. 2) relationships between spatial pattern and LST also differed among the three partitions. The landscape-level configuration metrics were all significantly negatively correlated to LST in heat-island partition , however, they displayed a positive relationship in cool-island partition. 3) In greenspace and water area, several patch-level landscape metrics displayed an inconsistent relationship among three partitions, suggested there existed a threshold character in cooling effect of ecological land uses. Research on relationship between urban landscape pattern and LST can help us understand the influence of the land use pattern on surface thermal environment, and it also provide support for studies about optimization of landscape pattern to ease heat island effect.