面向冬小麦遥感同化估产的叶面积指数空间尺度差异校正

发布时间：2018-03-21 17:14

本文选题：尺度效应　切入点：尺度转换　出处：《中国地质大学(北京)》2017年博士论文　论文类型：学位论文

【摘要】：遥感数据与作物生长模型在农作物估产上能够优势互补。然而,作物模型尺度与遥感观测尺度不匹配是影响同化模型精度的重要因素,这将极大增加遥感反演和数据同化的不确定性。本研究以基于作物生长模型与遥感数据同化的冬小麦估产为目标,针对遥感尺度这一重要科学问题,借助统计模型、物理方法、数据融合等技术,剖析了同化观测量(叶面积指数,LAI)尺度效应产生的主要原因,定量描述并校正了空间异质性与模型非线性所引起的尺度效应误差,探讨了多源遥感数据差异及其形成的内在机理,研究了Landsat与MODIS之间的尺度转换方法,建立了空间尺度差异校正的基本框架,并面向研究区对尺度校正框架进行了适用性验证,利用数据融合技术对同化观测量进行了时间尺度扩展,进而耦合WOFOST模型与多时空尺度遥感数据对河北省衡水地区冬小麦产量进行估测。论文的研究工作及主要结论如下:(1)在分析尺度效应产生根源、明确多源遥感数据差异的基础上,定量分析多空间尺度遥感反演农作物同化观测量LAI的总体差异。结果表明,多源遥感数据引起的差异高于尺度效应带来的误差。(2)考虑同化观测量LAI的空间异质性,细化尺度效应产生过程,结合小波变换和分形理论,定量分析冬小麦LAI不同反演过程对尺度效应的贡献,并有效校正尺度效应引起的误差。(3)从系统内在机理出发,归纳分析Landsat和MODIS数据差异,提取可通过数理方法模拟的相关信息,基于点扩散函数及粒子群优化算法,定量校正遥感观测数据差异,建立尺度转换模型,降低尺度差异导致的不确定性。(4)基于多尺度遥感数据反演同化观测量差异的定量分析,借助各类数理模型,构建并完善空间尺度差异校正框架,将多尺度遥感定量反演同化观测量LAI的总体不确定性降低了50%以上。(5)在空间尺度差异校正和时间尺度扩展的基础上,通过四维变分同化算法及SCE-UA优化算法,耦合多尺度遥感信息与WOFOST作物生长模型,对冬小麦进行区域化时空同化估产,在保证同化精度的前提下,大幅提高同化效率。
[Abstract]:The remote sensing data and crop growth model can be complementary in the estimation of crop yield. However, crop model scale and remote sensing observation scale, is an important factor affecting the assimilation model precision, which will greatly increase the retrieval and data assimilation uncertainty. Based on the yield of winter wheat crop growth model and remote sensing data assimilation based on the target in view of this, the scale of remote sensing of important scientific problems, physical methods by means of statistical model, data fusion technology, analyzes the assimilation measurements (leaf area index, LAI) mainly due to the scale effects, quantitative description of the error correction and scale effect of spatial heterogeneity and nonlinear model caused by the inherent mechanism of the difference of multi-source remote sensing data and the formation of the research on the transformation method between Landsat and MODIS scale, establish the basic framework of correction in different spatial scales, and surface To study on scale correction framework for the validation of measurements, the assimilation time scale expansion of the use of data fusion technology, and the coupling of WOFOST model and multi-scale remote sensing data on the yield of Winter Wheat in Hengshui area of Hebei province were estimated. The research work of this thesis and the main conclusions are as follows: (1) in the analysis of causes the scale effect, based on the difference of multi-source remote sensing data, the overall differences in quantitative analysis of multi spatial scale remote sensing crop assimilation measurements of LAI. The results show that the difference of multi-source remote sensing data caused by the above error scale effects. (2) considering the spatial heterogeneity of assimilation measurements of LAI, produced in the process of refining the scale effect, combined with wavelet transform and fractal theory, quantitative analysis of winter wheat LAI inversion process different contribution to the scale effect, scale effect and effective error correction caused by (3) from the Department. The system of internal mechanism, analyzed the difference between Landsat and MODIS data, the relevant information can be extracted by mathematical simulation method, point spread function and particle swarm optimization algorithm based on the difference of quantitative calibration of remote sensing observation data, a scale transformation model, reduce the scale difference leads to uncertainty. (4) quantitative analysis of differences of multi-scale remote sensing data based on the concept of retrieval and assimilation, with all kinds of mathematical model, construct and perfect the spatial scale difference correction framework of multi-scale quantitative remote sensing measurements of LAI assimilation overall uncertainty reduced by 50%. (5) based in correction in different spatial scales and time scales, the four-dimensional Variational Assimilation Algorithm and SCE-UA optimization algorithm, coupled multi-scale remote sensing information and crop growth model WOFOST, regional spatial assimilation yield of winter wheat, under the premise of ensuring the accuracy of assimilation, sharp Improve the efficiency of assimilation.

【学位授予单位】：中国地质大学(北京)
【学位级别】：博士
【学位授予年份】：2017
【分类号】：S512.11;S127

【相似文献】