城市气象学课件：05_3城市热岛_影响.ppt

Discussion 1、The first area for generalisation is operationalisation of concepts. The literature is reasonably successful in this regard, as most studies demonstrate good conceptual understanding of the heat island effect and establish appropriate definitions to test these concepts. 2、The second area for generalisation is controlled measurement. The literature is generally poor in this regard. Approximately half of all heat island studies fail to sufficiently control their measurements for the confounding effects of weather, relief or time. 3、The third area for generalisation is openness of method. The literature is highly inadequate in this area, with three-quarters of the sample failing to communicate, in most basic terms, the precision of instruments used to measure UHI magnitude and the physical nature of the surfaces surrounding those instruments at the time of measurement. Recommendations and closing remarks 1、Reduce the spatial and temporal resolution of your data. For the purpose of quantifying UHI magnitude, fewer field sites in representative locations is preferable to more sites in unrepresentative locations. Likewise, a smaller dataset of controlled measurements is preferable to a larger dataset of uncontrolled measurements. A simple comparison of two representative sites will provide a reasonably good measure of UHI magnitude, provided that the measurements sufficiently regulate the effects of weather, relief, time and random variation. 2、Follow standardised guidelines for site reporting. Guidelines in Aguilar et al. (2003) and Oke (2004) include descriptive templates for reporting the micro-, local- and meso-scale settings of temperature measurements in urban and rural environments. The information contained in these templates is essential to any heat island paper and to proper interpretation and comparison of its reported UHI magnitudes. 3、Disclose the limits of your data. Observational data in environmental science are limited in their certainty an