Skip to content
naulon
Not checkedNot checked for notices yet.

Modeling tree species diversity by combining ALS data and digital aerial photogrammetry

Jahangir Mohammadi, Shaban Shataee, Erik Næsset

Science of Remote Sensing · 2020

doi:10.1016/j.srs.2020.100011 (opens in a new tab)

We email you when a new notice or a question about this paper appears.

Abstract

Monitoring and assessment of tree species diversity in forests are essential tasks for forest managers. In the present study, we investigated the effectiveness of airborne laser scanner (ALS) and digital aerial photogrammetry (DAP) data for modeling tree diversity indices using machine learning algorithms in the uneven-aged Hyrcanian forests of Iran. Systematic sampling was adopted for the collection of field data on 300 circular plots (0.1 ​ha) in the study area. Menhenick, Margalef, Simpson’s heterogeneity, reciprocal of Simpson’s heterogeneity, Shanon-Winer heterogeneity and Simpson’s evenness indices were computed as measures of tree species diversity for each plot. Several commonly used variables were extracted from the ALS and DAP data. The results showed that the random forests (RF) algorithm produced the greatest accuracy among all machine learning algorithms. Additionally, combining ALS and DAP (ALS ​+ ​DAP) data increased prediction accuracy compared to separate modeling (0.2–6.6% reduction in relative root mean square error). The smallest RMSE% values based on independent validation data for Menhenick, Margalef, Simpson’s, reciprocal of Simpson’s, Shanon-Winer heterogeneity and Simpson’s evenness indices were 34.6%, 32.2%, 27.6%, 27.4%, 30.7% and 24.4%, respectively. These results demonstrate that the combination of ALS and DAP could be useful for monitoring tree species diversity in multi-story stands in the north of Iran and likely also in forests with complex canopy structures and multiple tree species.

Abstract licence: CC BY-NC

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.