The Impact Of Urbanization In The State's Basic Reference Station Surface Temperature Sequence Detection And Revised

Climatic change detection and attribution depend on reliable surface air temperature dataset. The urbanization-induced bias, however, is probably the largest source of systematic error so far identified for land surface air temperature observation data. Further more, surface air temperature data used for climate change research in mainland China is generally from the National Stations (i.e. national basic meteorological stations and national reference climate stations), and the dataset has been significantly affected by urbanization effect as shown by many case studies. A proper assessment and correction of the urban bias in the dataset from the National Stations is therefore of great importance for the land surface air temperature change research in China.One of the most crucial issues in identifying and correcting urban bias is the screening of the rural stations. The selection of 138 rural stations has been made in combination with another study. Based on a dataset of 614 National Stations and the 138 rural stations, and the widely adopted grid-weighted method for calculating the regional average climate series, annual and seasonal mean temperature series of six regions and China as a whole are constructed for time period 1961-2004, and the analysis of urbanization effect on surface air temperature changes observed from the National Stations is conducted.Results show that, excluding North Xinjiang, all the regions and China as a whole witness significant urbanization effect on the surface air temperature trends as recorded by the National Stations during period 1961-2004. The region-averaged annual urban warming, as defined by the difference between the annual mean surface air temperature trends of the National Stations and the rural stations, reaches 0.086℃/10a in the Middle and Lower Yangtze basin and the Huaihe basin, accounting for 55.48% of the overall warming. The annual urban warming are estimated to be 0.060℃/10a, 0.059℃/ 10a, 0.042℃/10a and 0.042℃/10a respectively for the regions of Northeast and North Plains, Tibetan Plateau, Southeast, and Northwest, and 0.076℃/10a for the whole country. Taking China as a whole, the contribution of the annual urban warming to the overall annual mean temperature increase recorded by the National Stations is 27.33%. Therefore, the urbanization effect on the trend of the region-averaged annual mean air temperature estimated from the National Stations is not negligible, and it should be eliminated from the current regional surface air temperature series in order to make a more reliable detection of climate change.It is also important to adjust the urban bias in the current land surface air temperature dataset. With reference to the adjustment methods by the previous researches, this work makes an attempt to adjust the urban bias in monthly and annual land surface temperature records of the 614 National Stations. Thus a new dataset with the in-homogeneities and urban bias being adjusted has been obtained. The comparison analysis of temperature trends of pre- and aft- adjustment dataset indicates an obvious difference, with the adjusted temperature trends being significantly decreased. However, Most regions of mainland China are still undergoing a significant warming in terms of annual mean surface air temperature change.