| 基于数据融合算法的灌区蒸散发空间降尺度研究 |
| |
| 引用本文： | 白亮亮,蔡甲冰,刘钰,陈鹤,张宝忠,黄凌旭.基于数据融合算法的灌区蒸散发空间降尺度研究[J].农业机械学报,2017,48(4):215-223. |
| |
| 作者姓名： | 白亮亮 蔡甲冰 刘钰 陈鹤 张宝忠 黄凌旭 |
| |
| 作者单位： | 中国水利水电科学研究院,中国水利水电科学研究院;国家节水灌溉北京工程技术研究中心,中国水利水电科学研究院;国家节水灌溉北京工程技术研究中心,中国水利水电科学研究院;国家节水灌溉北京工程技术研究中心,中国水利水电科学研究院;国家节水灌溉北京工程技术研究中心,中国水利水电科学研究院;国家节水灌溉北京工程技术研究中心 |
| |
| 基金项目： | “十二五”国家科技支撑计划项目(2012BAD08B01)、国家自然科学基金项目(51679254)和国家重点研发计划项目(2016YFC0400101) |
| |
| 摘 要： | 采用Landsat和MODIS数据,通过增强自适应融合算法(Enhanced spatial and temporal adaptive reflectance fusion model,ESTARFM)对蒸散发进行空间降尺度,构建田块尺度蒸散发数据集;利用2015年田间水量平衡方法计算的蒸散发数据对融合结果进行评价。在融合蒸散发基础上,结合解放闸灌域2000—2015年间种植结构信息,提取不同作物各自生育期和非生育期内年际蒸散发量,并分析了大型灌区节水改造以来,作物蒸散发占比的年际变化。研究结果表明:融合蒸散发与水量平衡蒸散发变化过程较吻合,小麦耗水峰值出现在6月中下旬—7月初,玉米和向日葵峰值出现在7月份。在相关性分析中,玉米、小麦和向日葵的决定系数R2分别达到了0.85、0.79和0.82;生育期内玉米(5—10月份)、小麦(4—7月份)和向日葵(6—10月份)的均方根误差均不高于0.70 mm/d;平均绝对误差均不高于0.75 mm/d;相对误差均不高于16%。在农田蒸散发总量验证中,融合蒸散发与水量平衡蒸散发相关性较好,两者决定系数达到了0.64。基于ESTARFM融合算法生成的高分辨率蒸散发(ET)结果可靠,具有较好的融合精度。融合结果与Landsat蒸散发的空间分布和差异性一致,7月23日、8月24日和9月1日相关系数分别达到0.85、0.81和0.77;差值均值分别为0.24 mm、0.19 mm和0.22 mm;标准偏差分别为0.81 mm、0.72 mm和0.61 mm。ESTARFM融合算法在农田蒸散发空间降尺度得到较好的应用,可有效区分不同作物蒸散发之间的差异。不同作物在生育期和非生育期内耗水量差别较大;生育期内套种(4—10月份)耗水量最大,达到637 mm,玉米(5—10月份)和向日葵(6—10月份)次之,分别为598 mm和502 mm,小麦(4—7月份)最低为412 mm;非生育期内,小麦(8—10月份)耗水量最大,年均达到214 mm,玉米(4月份)和向日葵(4—5月份)分别为42 mm和128 mm。不同作物多年平均耗水量(4—10月份)差异较小,其年际耗水总量主要随作物种植面积的变化而变化。
|
| 关 键 词： | 遥感 数据融合 蒸散发 地表能量平衡模型 增强时空自适应融合算法 河套灌区 |
| 收稿时间： | 2017/1/7 0:00:00 |
Spatial Downscaling of Evapotranspiration in Large Irrigation Area Based on Data Fusion Algorithm |
| |
| BAI Liangliang,CAI Jiabing,LIU Yu,CHEN He,ZHANG Baozhong and HUANG Lingxu.Spatial Downscaling of Evapotranspiration in Large Irrigation Area Based on Data Fusion Algorithm[J].Transactions of the Chinese Society of Agricultural Machinery,2017,48(4):215-223. |
| |
| Authors: | BAI Liangliang CAI Jiabing LIU Yu CHEN He ZHANG Baozhong and HUANG Lingxu |
| |
| Institution: | China Institute of Water Resources and Hydropower Research,China Institute of Water Resources and Hydropower Research;National Center for Efficient Irrigation Engineering and Technology Research-Beijing,China Institute of Water Resources and Hydropower Research;National Center for Efficient Irrigation Engineering and Technology Research-Beijing,China Institute of Water Resources and Hydropower Research;National Center for Efficient Irrigation Engineering and Technology Research-Beijing,China Institute of Water Resources and Hydropower Research;National Center for Efficient Irrigation Engineering and Technology Research-Beijing and China Institute of Water Resources and Hydropower Research;National Center for Efficient Irrigation Engineering and Technology Research-Beijing |
| |
| Abstract: | In order to construct the high spatial-temporal dataset of evapotranspiration (ET), the Landsat and MODIS data were used to achieve spatial downscaling of ET by using the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM). The result of data fusion was evaluated by field ET output from root zone water balance model. According to crop planting structure information from 2000 to 2015 in the study area, the water consumption of different crops was exacted during their growth and non-growth periods. Based on the fusion ET, the interannual variation of total agricultural water consumption was analyzed since the implement of water-saving project in large irrigation district. The result showed that the process of fusion ET was more consistent with ET output from water balance. In the correlation analysis of water balance and fusion ET, the determination coefficients (R2) of maize, wheat and sunflower reached 0.85, 0.79 and 0.82, respectively. During the growth period, the root mean square errors (RMSE) of maize (May to October), wheat (April to October) and sunflower (June to October) were lower than 0.70mm/d, the mean absolute error (MAD) was all lower than 0.75mm/d, and the relative error (RE) was all less than 16%. On the spatial scale, the spatial characteristics of fusion results were consistent with the Landsat ET. The correlation coefficients of July 23, August 24 and September 1 reached 0.85, 0.81 and 0.77, the mean values of the differences were 0.24mm, 0.19mm and 0.22mm, and the standard deviations were 0.81mm, 0.72mm and 0.61mm, respectively. The high resolution ET based on ESTARFM fusion algorithm was reliable and had good fusion precision. The water consumption of different crops varied greatly both in the growth period and non-growth period. During the growth period, the maximum water consumption was 637mm for interplanting (April to October), followed by maize and sunflower, which were 598mm (May to October) and 502mm (June to October), respectively, the minimum water consumption of wheat was 412mm (April to July). During the non-growth period, wheat (August to October) had the highest water consumption with an annual average of 214mm, and those of maize (April) and sunflower (April to May) were 42mm and 128mm, respectively. Due to the difference of average annual water consumption of different crops was not significant during April to October, the variation of total water consumption for different crops was varied with the changes of crop acreage. |
| |
| Keywords: | remote sensing data fusion evapotranspiration surface energy balance model enhanced spatial and temporal adaptive reflectance fusion model Hetao irrigation district |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《农业机械学报》浏览原始摘要信息 |
| 点击此处可从《农业机械学报》下载免费的PDF全文 |
|
