LiDAR Insights into Biodiversity Patterns - Evaluating Vegetation Structure’s Role in Ecosystem Assessments
For the containment of the ongoing biodiversity loss, a targeted documentation of the status and dynamics of ecosystems is necessary. With LiDAR remote sensing, it is possible to quantify the vegetation structure, which plays a vital role for biodiversity. Technical developments in remote sensing...
Saved in:
| Main Author: | |
|---|---|
| Contributors: | |
| Format: | Doctoral Thesis |
| Language: | English |
| Published: |
Philipps-Universität Marburg
2024
|
| Subjects: | |
| Online Access: | PDF Full Text |
| Tags: |
No Tags, Be the first to tag this record!
|
| Summary: | For the containment of the ongoing biodiversity loss, a targeted documentation of the status and dynamics
of ecosystems is necessary. With LiDAR remote sensing, it is possible to quantify the vegetation structure,
which plays a vital role for biodiversity. Technical developments in remote sensing and especially LiDAR
hold great potential to implement comprehensive monitoring strategies but still pose some challenges. In
this thesis, the ability of LiDAR data for predicting animal species across taxa, mapping tree species group
specific successional stages as well as seasonal mapping of the leaf area index was systematically analyzed.
With three case studies that vary by scale, data and ecological process, I aimed at analyzing the utility of
LiDAR data for the mapping of biodiversity in the context of different data sources.
In case study one, high-resolution airborne LiDAR data, were especially obtained at Mount Kilimanjaro
for the purpose of modeling animal species richness. Taxonomic data from an extensive multi-taxa field
campaign were used to study the contribution of LiDAR data compared to environmental co-variables for
species richness models. For most animal groups, the superior influence of elevation compared to vegetation
structure was demonstrated. As the natural processes in the study area at the slopes of Mount
Kilimanjaro are highly influenced by the elevation gradient the especially collected structural LiDAR data
could not significantly add additional value. The study was designed as a multi-taxa approach, therefore
results were compared systematically across taxa which is a valuable component for eventually bridging the
gap between small scale local studies and global mechanisms.
Those specifically commissioned high-resolution LiDAR flight campaigns are resource intensive and not
always within the possibilities of ecological projects. In many countries, however, governmental LiDAR
data with rather low spatial resolution and a repetition rate of a couple of years exist. At the scale of
federal states, wall-to-wall data consist of several observation dates and have varying quality. With that in
mind, the benefit of these data initially appears questionable, which is probably why they are rarely used
for the classification of successional stages.
In the second case study, the potential of such data for modeling forest successional stages for Rhineland-
Palatinate was analyzed. By properly adapting data processing and modeling strategies, it was shown
that for mapping of forest successional stages, even highly heterogeneous LiDAR data, in addition to the
commonly used spectral data, could make a valuable contribution to biodiversity monitoring.
In the third case study included in this thesis, the seasonal changes of leaf area index with spaceborne
LiDAR data were investigated. This study was based on the rather new LiDAR data from the spaceborne
GEDI (Global Ecosystem Dynamics Investigation) mission. However, with its scattered collection of 25
m footprints and with overflights of the same area every couple of weeks the GEDI sampling design is
neither comprehensive nor repetitive for the exact same points. This initially presented novel challenges
since wall-to-wall monitoring is desirable but cannot be directly derived from GEDI data. The regular overflights of the same areas, however, open up new possibilities for the integration with area-wide sensors.
In this study the GEDI data were integrated with radar and optical data (Sentinel-1 and -2) to provide a
spatio-temporally continuous mapping of the leaf area index. This approach was the first to analyze the
monthly dynamics of the leaf area index derived from structural GEDI data comprehensively for a regional
study. The satisfactory results of this integration method for different types of land use seemed promising
in regards to further approaches in the direction of using multi-temporal data sets for regional ecological
monitoring. Hence, even though planned to monitor global ecosystem dynamics, it was shown that the
recorded GEDI data hold the potential of delivering valuable insights even for regional studies.
In summary, this thesis provides novel insights into the applicability of LiDAR data for biodiversity research.
The general relevance of vegetation structure for biodiversity and LiDAR’s ability to quantify
three-dimensional vegetation structure offer great chances to support various ecological questions on different
scales. |
|---|---|
| Physical Description: | 103 Pages |
| DOI: | 10.17192/z2024.0107 |