What is it about?
Forest inventories serve a wide range of purposes, e.g. through providing information about carbon stocks and biodiversity. Increasingly, forest inventories are based on remote sensing and models that predict target quantities from the remotely sensed data. This article highlights potential pitfalls in this context, and discusses means to avoid them.
Featured Image
Photo by Geranimo on Unsplash
Why is it important?
Forest inventories are often being based on predictions from remotely sensed data, but properties of such information are not fully understood by practitioners. This article highlights important features of this type of information and discusses how mistakes can be both committed and avoided when this type of information is used.
Perspectives
The study suggests that in many cases it is important to modify predictions using remotely sensed data, obtained from traditional methods such as regression analysis and other machine learning algorithms. Otherwise, mistakes may be committed when planning forest management or developing forest policy.
Göran Ståhl
Swedish University of Agricultural Sciences
Read the Original
This page is a summary of: Why ecosystem characteristics predicted from remotely sensed data are unbiased and biased at the same time – and how this affects applications, Forest Ecosystems, January 2024, Tsinghua University Press,
DOI: 10.1016/j.fecs.2023.100164.
You can read the full text:
Contributors
The following have contributed to this page
