山地典型生态参量遥感反演建模及其时空表征能力研究

The topographic variation in mountainous area significantly influences the remote sensing imaging and radiative transfer processes, which makes the retrieval of ecological parameters by the remote sensing method much more complex. Therefore, it is urgent to solve the critical scientific problems to monitor the dynamics of the eco-environment in mountainous areas: (1) the scientific description of land surface ecological processes with remote sensing methods in mountainous areas, (2) the particularity and applicability of the key ecological remote sensing estimation and assimilation methods, and (3) the uncertainty of the key ecological parameters estimation and its error propagation mechanism. With the consideration of these key issues, this project will conduct following researches in mountainous areas with the help of field validation sites, physically-based models and Satellite-Airplane-Ground synchronous observation experiment. The study contains:.(1) synergetic correction of the terrain, atmosphere and BRDF effects of the optical remote sensing data in the mountainous area and the construction of remote sensing dataset with the high spatio-temporal resolution, (2) models and methods development for LAI and NPP ecological parameters estimation, (3) analysis of the spatial scale effects of the estimation model and the retrieved key ecological parameters and find their numerical expressions, (4) development of theories and methods for the spatio-temporal evaluation of remote sensing products, and (5) theories exploration and methods integration for quantitative remote sensing in mountainous areas. The successful implementation of this project will realize the automatic process of remote sensing data and rapid derivation of the spatial-temporal information of ecological parameters in the mountain environment. In addition, it will also illuminate the mechanism of the error accumulation and propagation in remotely sensed products. There should be some significant breakthroughs in the theories and methods for quantitative remote sensing of the.mountain environment. This study will be helpful to provide the theoretic, technical, and methodological support for dynamically monitoring the mountainous eco-environment in China.

山地地形起伏影响遥感成像和辐射传输过程，生态参量遥感反演建模更加复杂，亟待解决诸多科学问题：(1)遥感信息对山地陆表复杂格局和快速过程的表征；(2)生态参量遥感反演与同化理论在山区的特殊性和适用条件；(3)山地生态参量遥感反演不确定性及其传播机制等。本项目在星机地一体化同步观测试验协助下，通过真实场景模拟构建和机理模型推导，重点研究：(1)山地光学影像大气-地形-BRDF协同改正建模与高时空数据集构建；(2)山地陆表LAI、NPP等参量遥感反演模型与策略；(3)反演模型及产品的空间尺度效应及其数学表达；(4)反演产品时空表征能力评估；(5)山地定量遥感理论提升、方法集成与应用示范。项目成果将增强山地遥感数据自动化处理和生态参量时空信息获取能力，阐明遥感产品误差积累及其传播机制，可期在山地定量遥感理论和方法上取得突破性进展，为满足我国山区生态环境遥感监测科技需求提供基础理论、技术和方法支撑。

山地环境特有的地形复杂性以及由此引起的生态系统的复杂性和气候环境的多变性，致使山地陆表生态参量遥感反演面临更多理论和技术难题，山地陆表生态参量定量遥感反演不确定性更为突出。本项目选取山地典型生态系统为研究对象，在现有生态参量遥感反演研究进展的基础上，剖析山地陆表生态参量遥感反演建模及其时空表征能力研究所需解决的科学问题，主要研究（1）山地光学遥感地形改正建模与高时空分辨率影像集融合方法、（2）山地陆表典型生态参量遥感反演模型与策略、（3）山地陆表生态参量遥感反演空间尺度效应及其转换机制、（4）山地典型生态参量遥感产品时空表征能力评估、以及（5）山地生态参量定量遥感理论提升、方法集成与应用示范等内容。.自立项以来，围绕以上研究内容系统梳理了山地遥感研究进展、方法、现状及问题；构建了新的坡地冠层反射率模型及大气-地形-BRDF协同校正模型，发展了时空谱归一化的山地高时空分辨率卫星影像时空融合方法；考虑复杂地形特征，兼顾不同分辨率遥感影像间的信息传递过程，构建了多源、多时空尺度山地生态参量遥感协同反演模型，开展了反演建模过程中的地形效应分析；分别从空间升尺度和降尺度两个方面发展了地表参量尺度转换方法；构建了国内首个面向复杂地形特征的王朗山地生态遥感综合观测试验站，设计了样点-样方-样地-样区多尺度嵌套的遥感产品验证技术框架，基于直接验证和交叉验证方法系统评价了地形对遥感产品可靠性的影响程度及其应用效果。截止目前，总计发表项目标注论文94篇，其中第一标注论文39篇；出版专著1部；培养博士8人，硕士22人；3人入选国家级人才计划；6人晋升高级职称；成果获得Elsevier Atlas国际奖；项目中期考核被评为优秀。.通过本项目的研究，在方法上补充和完善了山地地表生态参量遥感反演模型，在实践上通过集成山地典型生态参量遥感反演方法、模型和策略，推动了山地生态参量定量遥感理论提升，为构建典型山区生态环境遥感监测技术体系提供基础理论、技术和方法支撑。研究成果在大熊猫国家公园监测、生物多样性监测、一带一路监测与评估等国家需求中成功应用。