1981-2018年全球0.05度叶面积指数8天数据

本文介绍了集成时间序列的遥感观测数据，生产时空连续的长时间序列的全球叶面积指数（LAI）产品的算法。该算法主要包括有效LAI与真实LAI之间的转换、LAI产品的融合地表反射率数据预处理、集成时间序列反射率数据的LAI反演以及时间连续处理策略等。基于该算法，利用MODIS地表反射率数据生产了2000年至2018年8天500m分辨率的全球LAI产品，利用AVHRR地表反射率数据生产了1981至2018年8天5km分辨率的全球LAI产品.
通过与现有的全球LAI产品进行比较，对生产的全球LAI产品的空间和时间一致性进行了评价，并利用LAI的地面测量数据，对生产的全球LAI产品的精度进行了评估。

This paper presents an algorithm for generating spatially-temporally continuous long-time-series global Leaf Area Index (LAI) products using integrated time-series remote sensing observations. The algorithm mainly includes the conversion between effective LAI and true LAI, preprocessing of surface reflectance data for LAI product fusion, LAI inversion based on integrated time-series reflectance data, and time-continuity processing strategies, among others. Based on this algorithm, global LAI products with 8-day 500 m resolution from 2000 to 2018 were generated using MODIS surface reflectance data, while global LAI products with 8-day 5 km resolution from 1981 to 2018 were generated using AVHRR surface reflectance data.
The spatial and temporal consistency of the generated global LAI products was evaluated by comparing them with existing global LAI products, and the accuracy of these products was assessed using in-situ LAI measurement data.

搜集汇总

背景与挑战

背景概述

该数据集提供了1981年至2018年全球范围的叶面积指数（LAI）8天合成产品，空间分辨率为0.05度（约5公里）。它基于AVHRR遥感数据，通过集成时间序列算法生成，旨在实现时间连续的长期监测，并经过地面验证，在非森林和森林区域均表现出较高的数据一致性（RMSE分别为0.4501和0.6301），适用于全球植被动态和气候变化研究。