青藏高原30米分辨率火烧迹地产品（2020）

1）数据内容：本数据集包含2020年青藏高原地区30米分辨率火烧迹地产品。2）数据来源及加工方法：基于时间序列Landsat地表反射率和火烧迹地敏感光谱参量，通过选取大量典型火烧迹地样本点，利用随机森林等机器学习算法研发并生产的30米分辨率火烧迹地产品；3）数据质量描述：产品总体精度在90%以上。4) 数据应用成果及前景：该数据集可为火灾监测、碳排放研究、生态环境监测、全球变化研究等相关领域的研究和应用提供数据产品支撑。

1) Data Content: This dataset encompasses the 30-meter resolution burned area product covering the Qinghai-Tibet Plateau in 2020.
2) Data Source and Processing Workflow: Based on time-series Landsat surface reflectance data and spectral parameters sensitive to burned areas, the 30-meter resolution burned area product was developed and generated via machine learning algorithms including Random Forest, with a large number of typical burned area sample points selected for model training.
3) Data Quality Assessment: The overall accuracy of the product exceeds 90%.
4) Data Application Outcomes and Prospects: This dataset can provide data product support for research and applications in relevant fields such as fire monitoring, carbon emission research, ecological environment monitoring, and global change research.

搜集汇总

背景与挑战

背景概述

该数据集为2020年青藏高原地区30米分辨率的火烧迹地产品，基于Landsat地表反射率和机器学习算法生产，总体精度超过90%，适用于火灾监测、碳排放研究等领域。数据以GeoTiff格式存储，像素值1代表火烧迹地，空值代表非火烧迹地。