New Applications of Forest Models in combination with Remote Sensing

Please use this identifier to cite or link to this item:
Open Access logo originally created by the Public Library of Science (PLoS)
Title: New Applications of Forest Models in combination with Remote Sensing
Other Titles: Neue Methoden der Waldmodellierung in Komination mit Fernerkundung
Authors: Henniger, Hans
ORCID of the author:
Thesis advisor: Prof. Dr. Andreas Huth
Thesis referee: Prof. Dr. Uta Berger
Abstract: Forests cover 31% of the global land surface. They play a major role for the global carbon cycle because of their role as carbon storage and their contributions to global carbon fluxes. Forest ecosystems are exposed to environmental changes due to climate warming and related change processes as droughts, heat waves, fires, storms or pest outbreaks, but also deforestation and fragmentation which accelerating and occurring more and more simultaneously. Forest models contribute to the understanding of forests and the dynamics of ecosys- tems under changing environmental conditions. With the increasing availability of remote sensing data and increasing computing power, new opportunities are emerging for the application of forest models. This also means that forest models need to be adapted, extended or applied in new ways to take full advantage of these new condi- tions. This thesis presents innovative and novel applications of forest models, which could help to profit from such new opportunities. The first study of this thesis (Chapter 2) establish a new way of using forest models by extending the forest factory approach by and make it applicable for forests in different biomes. This approach allows to generate forests using the architecture and processes of forest models (here we use the individual-based gap model FORMIND). In this study, 700,000 forest stands in seven different ecoregions are generated by using the Forest Factory 2.0. In contrast to the tradition of investigating the development of individual forest stands over time, we used the Forest Factory 2.0 as a tool to gain knowledge about forests by analyzing the state space of forests. We conducted a structural sensitivity analysis to compare the relationships between structural properties and biomass, productivity, as well as (tree) species evenness of forests. We analyze the state space of forests in different biomes and demonstrate the potential of this approach for theoretical ecology. With the Forest Factory 2.0, researchers can generate virtual forests for their needs or use the open- source forest data to analyze a digital forest universe of forest states. The second study in this thesis (Chapter 3) provides insights into how remote sensing measurements can be incorporated in forest models. It is about a new approach which enable the calculation of hyperspectral reflectance of forests. It uses the multi-layer radiative transfer model mScope and the individual-based forest model FORMIND. This work provides a forward modeling approach for relating forest reflectance to for- est characteristics. With this tool, it is possible to analyze a large set of forest stands and their corresponding reflected radiance (in the visible and near infrared range). This opens up the possibility to understand how forest reflectance is related to succession and different forest conditions. In order to take advantage of the increasing number of remote sensing measurements and to achieve synergy effects with forest models, it would be useful to align the de- sign of satellite missions with the capabilities of forest models. The last study of the thesis (Chapter 4) is about using expected biomass distributions provided by the upcoming RADAR BIOMASS P-band satellite mission (launching in 2024 by the Eu- ropean Space Agency) to predict the productivity of tropical forests. The results show a high correlation for estimating productivity with a biomass distribution at a spatial resolution of 4 ha and 1 ha. Increased vertical resolution leads generally to better pre- dictions for productivity (GPP, NPP). Further, the results demonstrate the influence of spatial resolution with differences between disturbed and mature forests. The pre- sented approach offers a number of innovations: (i) the use of expected remote sensing measurements, (ii) the use of individual-based forest models for preliminary studies of satellite missions, (iii) the use of RADAR satellite measurements for the prediction of productivity and carbon turnover times and (iv) the exploration of the prediction quality for different forest types. The obtained results emphasize the value of the forthcoming BIOMASS satellite mission and highlight the potential of deriving estimates for forest productivity from information on forest structure. The studies presented in the thesis are providing a basis for future applications. In ad- dition, they show first applications of how these newly developed and novel methods can be used to investigate the relationship between forest structure and productivity (Chapter 2 and 4), to explore virtual remote sensing measurements for a case study in Finland (Chapter 3), and finally to estimate productivity for typical tropical forests using RADAR remote sensing data (Chapter 4). The applications presented may ulti- mately contribute to a better understanding of forest ecosystems.
Subject Keywords: forest model; radiative transfer; vegetation indices; individual-based; forest reflectance; machine learning; remote sensing; RADAR; forest generator; ecosystem functions; productivity; forest biomass; forest factory
Issue Date: 18-Apr-2024
License name: Attribution-NonCommercial-NoDerivs 3.0 Germany
License url:
Type of publication: Dissertation oder Habilitation [doctoralThesis]
Appears in Collections:FB06 - E-Dissertationen

Files in This Item:
File Description SizeFormat 
thesis_henniger.pdfPräsentationsformat14,02 MBAdobe PDF

This item is licensed under a Creative Commons License Creative Commons