| SWAT 2000气象模拟器的随机模拟原理、验证及改进 |
| |
| 引用本文： | 张东,张万昌,张东.SWAT 2000气象模拟器的随机模拟原理、验证及改进[J].资源科学,2004,26(4):28-36. |
| |
| 作者姓名： | 张东 张万昌 张东 |
| |
| 作者单位： | 1. 南京大学,国际地球系统科学研究所(ESSI),南京,210093;南京师范大学地理科学学院,南京,210097
2. 南京大学,国际地球系统科学研究所(ESSI),南京,210093 |
| |
| 基金项目： | 国家重点基础研究发展计划(973计划)，海外青年学者合作研究基金，教育部科学技术基金 |
| |
| 摘 要： | 该文详细介绍了SWAT 2000气象模拟器(最高、最低气温、太阳辐射)的随机模拟原理,利用甘肃河西地区酒泉、张掖等7个站的完整气象资料,提出了气候条件相似的区域气象台站月平均气象数据的计算方法,并计算出模拟器需要的输入参数,对气象模拟结果进行了验证.在此基础上利用逐日平均气温修正模拟生成的最高、最低气温并求5日滑动平均,对模拟器进行了改进.通过模拟结果与气象台站实测资料的对比可以看出,改进后最高气温相关系数从0.85提高到0.90,最低气温相关系数从0.89提高到0.94,气温数据模拟精度有一定提高.最后利用改进后的气象模拟器对河西黑河地区祁连、札马什克、莺落峡3个数据短缺站点进行了气温和太阳辐射的模拟,通过SWAT模型的应用证明模拟结果能够满足分布式水文模型的气象输入数据精度要求.
|
| 关 键 词： | 气象模拟器 随机模拟 气温修正 5日滑动平均 分布式水文模型 |
| 文章编号： | 1007-7588(2004)04-0028-09 |
| 收稿时间： | 1/1/1900 12:00:00 AM |
| 修稿时间： | 2003年8月6日 |
Stochastic Simulation Theory,Validation and Improvement for SWAT 2000 Climate Simulator |
| |
| ZHANG Dong.Stochastic Simulation Theory,Validation and Improvement for SWAT 2000 Climate Simulator[J].Resources Science,2004,26(4):28-36. |
| |
| Authors: | ZHANG Dong |
| |
| Institution: | ZHANG Dong~ |
| |
| Abstract: | The Objective of this article is to improve the simulation efficiency of the climate simulator which is used to simulate maximum, minimum temperature and solar radiation as the input climate driver forces for SWAT2000 distributed hydrological model. Firstly the stochastic theory of climate simulator was described in detail. And then, an analysis was made to the full serial weather data in 7 calibration stations located on Hexi corridor, Gansu, China, such as Jiuquan and Zhangye. It indicated that the correlation coefficients for observed monthly mean maximum, minimum temperatures versus those related mean temperatures were above (0.99) and their standard deviations versus those related ones were all greater than 0.89. Based on these conclusions, a methodology to calculate monthly mean maximum, minimum temperature and their corresponding standard deviations where similar weather conditions do exist was developed. The monthly statistical data could be obtained by adding its monthly mean air temperature to a certain increments and they had a fairy well precision. Using these data as the original climate simulator input, a calibration was made to the 7 calibrated stations. It showed there had some abnormal surges in the simulation results and even some logical errors in them such as the daily maximum temperature lower than daily minimum one. They were caused by the stochastic-noise-matrix to the climate simulator. By using daily mean air temperature to correct the logical errors in the simulated daily maximum and minimum temperatures and by calculating 5-day overlapping mean on them, the correlation coefficients increased from 0.85 to 0.90 and 0.89 to 0.94 respectively by the comparison between simulated results and observed ones. The precision of simulated weather data was fairy improved. And lastly, the temperature and solar radiation data for Qilian, Zhamashike and Yingluoxia weather stations were simulated by the modified climate simulator. Putting them into SWAT hydrological model as the input parameters, the coefficient for the simulated hydrological results with those observed ones reached 0.82, which meet the requirements of the data precision for the distributed hydrological model. |
| |
| Keywords: | Climate simulator Stochastic simulation Temperature revision 5-day overlapping mean Distributed hydrological model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《资源科学》浏览原始摘要信息 |
| 点击此处可从《资源科学》下载免费的PDF全文 |
|
