投稿中心

审稿中心

编辑中心

期刊出版

网站地图

友情链接

引用本文:陈喜,吴敬禄,王玲.人工神经网络模型预测气候变化对博斯腾湖流域径流影响.湖泊科学,2005,17(3):207-212. DOI:10.18307/2005.0303
CHEN Xi,WU Jinglu,WANG Ling.Prediction of Climate Change Impacts on Streamflow of Lake Bosten Using Artificial Neural Network Model. J. Lake Sci.2005,17(3):207-212. DOI:10.18307/2005.0303
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
←前一篇|后一篇→ 过刊浏览    高级检索
本文已被:浏览 4649次   下载 2540 本文二维码信息
码上扫一扫!
分享到: 微信 更多
人工神经网络模型预测气候变化对博斯腾湖流域径流影响
陈喜1, 吴敬禄2, 王玲1
1.河海大学水文水资源与水利工程科学国家重点实验室, 南京 210098;2.中国科学院南京地理与湖泊研究所, 南京 210008
摘要:
温室气体排放量增加造成气候变化,对全球资源环境产生重要影响.本文利用人工神经网络模型建立月降水、气温与径流关系,利用开都河流域降水、气温、径流资料对模型进行训练和验证,通过试算法确定网络模型结构.气温升高和降水量增加对径流影响的敏感程度分析表明,气温升高和降水增加对该区域径流影响较大,且气温升高的影响更为显著,径流增加主要集中在夏季.根据区域气候模型(RCMs)推算的CO2加倍情况下西北地区气候的可能变化,预测位于博斯腾湖流域的开都河大山口站年径流量增加38.6%,其中夏季增加71.8%,冬季增加11.4%.
关键词:  人工神经网络  气候变化  径流  预测
DOI:10.18307/2005.0303
分类号:
基金项目:中国科学院南京地理与湖泊所知识创新工程重要方向性项目(XXNIGLAS-A01-2)国家自然科学基金(40273004)联合资助.
Prediction of Climate Change Impacts on Streamflow of Lake Bosten Using Artificial Neural Network Model
CHEN Xi1, WU Jinglu2, WANG Ling1
1.Key Laboratory of Water Resources Development, Hohai University, Nanjing 210098, P. R. China;2.Nanjing Institute of Geography and Limnology, CAS, Nanjing 210008, P. R. China
Abstract:
Climate change takes a great effect on global hydrology and Welter resources, ecology and environment, and social economic development due to an increasing concentration of greenhouse gases in the atmosphere. Relationship among streamflow and its influences of precipitation and temperature in monthly scale is developed using an artificial neural network model. The model is trained and validated based on inputs of precipitation and temperature data in Bayinbuluke hydrological station within the study catchment and output of streamflow data in the Dashankou hydrological station which controls streamflow of the Kaidu River into the Bosten Lake. The model structure is determined with a trial and error method. Sensitivity analysis of modeling slremflow to temperature rise and precipitation increase demonstrates that influences of temperature rise is more significant than that of precipitatioo in-crefise, and streamflow increase is primarily concentrated in summer season. Based on input of possible future climate scenarios predicted by regional climate models (RCMs), the model prediction presents thal annual streamflow would increase 38.6 percent, 71.8 percent in summer and 11.4 percent in winter.
Key words:  artificial neural networic  climate changes  streamflow  prediction
分享按钮