This thesis focuses on land degradation in the hinterland of Rio de Janeiro, part of the highly endangered Brazilian Atlantic Forest biome. Forest and pasture degradation are of outmost concern in the region. Thus, the main objective of this work is to provide a methodology to identify areas that fulfill the aim of increasing forest area, improving forest quality and reducing overall pasture degradation. Therefore, this work presents an approach for high-resolution land cover and pasture degradation mapping as well as an approach for prioritizing reforestation sites. Based on the prioritized land, forest scenarios are modelled and evaluated. Outcomes of this work include the recommendation to restrict land use on slopes above 10° and theoretical considerations to adapt compensatory payments for reforestation based on the priorities of the identified sites. Methods used in this work encompass image fusion using RandomForest regression, Land Cover Classification with the RandomForest classifier and Multiple Endmember Spectral Mixture Analysis (MESMA) and field methods for pasture degradation mapping.:LIST OF FIGURES XVII
LIST OF TABLES XIX
LIST OF ABBREVIATIONS XXI
1 INTRODUCTION 1
1.1 PROBLEM IDENTIFICATION 1
1.2 RESEARCH OBJECTIVES AND QUESTIONS 3
1.3 VALUE OF THE RESEARCH 5
1.4 STRUCTURE 6
2 CONCEPTUAL BACKGROUND 7
2.1 LAND DEGRADATION 7
2.1.1 Definition of Land Degradation 7
2.1.2 Forest Fragmentation 9
2.1.3 Pasture Degradation 14
2.2 REMOTE SENSING FUNDAMENTALS 16
2.2.1 Optical Remote Sensing 16
2.2.2 Image Classification and Land Cover Mapping 20
2.2.3 Vegetation Discrimination 22
2.2.4 Digital Elevation Model (DEM) Applications 24
2.3 LANDSCAPE METRICS 25
3 STUDY AREA 27
3.1 LOCATION 27
3.2 PHYSICAL ENVIRONMENT 28
3.3 HUMAN ENVIRONMENT 32
4 PART (A): STATE OF THE ART 39
4.1 LAND DEGRADATION IN THE GUAPI-MACACU WATERSHED 40
4.1.1 Forest Fragmentation 40
4.1.2 Pasture Degradation 42
4.2 REMOTE SENSING-BASED LAND DEGRADATION ASSESSMENTS 46
4.2.1 Forest Monitoring in the Brazilian Atlantic Forest Biome 49
4.2.2 Pasture Degradation Mapping 51
4.3 POLICIES AND PROGRAM WITH RELEVANCE TO CONSERVATION AND REFORESTATION PROJECTS IN BRAZIL 54
4.3.1 International Programs, Schemes and Intitiatives 54
4.3.2 National Environmental Policies and Programs 58
4.3.3 Policies and Programs with Focus on the Brazilian Atlantic Forest 64
4.3.4 State-based Regulations in the Rio de Janeiro Federal State 65
4.4 LAND REHABILITATION AND REFORESTATION EFFORTS IN THE RJ FEDERAL STATE 66
4.4.1 Pasture Rehabilitation 66
4.4.2 Forest Restoration 67
5 MATERIAL AND METHODS 71
5.1 GEODATA AND SOFTWARE 71
5.2 PART (B): LAND COVER AND PASTURE DEGRADATION MAPPING 73
5.2.1 Field Survey of Degraded Pastures 73
5.2.2 Satellite Data Processing 75
5.2.3 Hot Spot Analysis 86
5.2.4 Relation of Slope Angle and Degradation Class 87
5.3 PART (C): PRIORITIZATION OF REFORESTATION SITES 87
5.3.1 Preliminary Study 88
5.3.2 Spatial Multi-Criteria Evaluation (SMCE) 90
5.3.3 Forest Scenario Development and Evaluation Using Landscape Metrics 94
6 RESULTS 97
6.1 PART (B): MAPPING OF PASTURE DEGRADATION 97
6.1.1 Categorization of Pasture Degradation 97
6.1.2 High-resolution SWIR band modelling 99
6.1.3 Land Cover Classification with Random Forests 101
6.1.4 Pasture Degradation Mapping Using Spectral Mixture Analysis and Field Data 103
6.1.5 Hot Spot Analysis of Pasture Degradation 106
6.1.6 Slope Influence on Pasture Degradation 107
6.2 PART (C): PRIORITIZATION OF REFORESTATION SITES USING SMCE 109
6.2.1 Characteristics of the Prioritized Areas 109
6.2.2 Forest Scenarios 113
7 DISCUSSION 117
7.1 PART (A/B): ASSESSMENT OF THE STATUS OF LAND DEGRADATION IN THE GUAPI-MACACU WATERSHED 117
7.2 PART (B): METHODOLOGICAL APPROACH FOR PASTURE DEGRADATION MAPPING 119
7.2.1 Satellite images for high-resolution LCC 119
7.2.2 High-resolution SWIR Band Modelling 120
7.2.3 Land Cover Classification 121
7.2.4 Pasture Degradation Mapping Approach 122
7.3 PART (C): SPATIAL PRIORITIZATION FOR REFORESTATION MEASURES 125
7.3.1 Identification of Priority Sites 125
7.3.2 Forest Scenarios 127
8 CONCLUSION AND RECOMMENDATIONS 131
9 OUTLOOK 135
REFERENCES 137
ANNEX I / Die vorliegende Arbeit befasst sich mit der Landdegradation im Hinterland von Rio de Janeiro, Teil des stark gefährdeten brasilianischen Atlantikwaldbioms. Wald- und Weidedegradation zählt zu den Hauptproblemen in der Region. Das übergeordnete Ziel dieser Arbeit ist daher die Bereitstellung einer Methodik, um Flächen zu identifizieren, die zur Vergrößerung der Waldfläche und Verbesserung der Waldqualität sowie gleichzeitiger Verminderung degradierter Weiden beitragen. Aus diesem Grund stellt diese Arbeit einen Ansatz für eine hochauflösende Kartierung der Landbedeckung und der Weidedegradation sowie einen Ansatz für die Priorisierung von Wiederaufforstungs-gebieten vor. Auf der Grundlage der priorisierten Flächen werden Waldszenarien modelliert und bewertet. Ergebnisse dieser Arbeit beinhalten u.a. die Empfehlung zur Einschränkung der Landnutzung auf Hängen über 10° und theoretische Überlegungen zur Anpassung der Ausgleichszahlungen für die Wiederaufforstung auf der Grundlage der Prioritäten der Standorte. Die in dieser Arbeit verwendeten Methoden umfassen Bildfusion mittels RandomForests Regression, die hochauflösende Ableitung der Landbedeckung unter Verwendung des RandomForests Klassifiizierers, sowie spektrale Entmischung mittels Multiple Endmember Spectral Mixture Analysis (MESMA) und Feldmethoden für die Kartierung des Weidezustands.:LIST OF FIGURES XVII
LIST OF TABLES XIX
LIST OF ABBREVIATIONS XXI
1 INTRODUCTION 1
1.1 PROBLEM IDENTIFICATION 1
1.2 RESEARCH OBJECTIVES AND QUESTIONS 3
1.3 VALUE OF THE RESEARCH 5
1.4 STRUCTURE 6
2 CONCEPTUAL BACKGROUND 7
2.1 LAND DEGRADATION 7
2.1.1 Definition of Land Degradation 7
2.1.2 Forest Fragmentation 9
2.1.3 Pasture Degradation 14
2.2 REMOTE SENSING FUNDAMENTALS 16
2.2.1 Optical Remote Sensing 16
2.2.2 Image Classification and Land Cover Mapping 20
2.2.3 Vegetation Discrimination 22
2.2.4 Digital Elevation Model (DEM) Applications 24
2.3 LANDSCAPE METRICS 25
3 STUDY AREA 27
3.1 LOCATION 27
3.2 PHYSICAL ENVIRONMENT 28
3.3 HUMAN ENVIRONMENT 32
4 PART (A): STATE OF THE ART 39
4.1 LAND DEGRADATION IN THE GUAPI-MACACU WATERSHED 40
4.1.1 Forest Fragmentation 40
4.1.2 Pasture Degradation 42
4.2 REMOTE SENSING-BASED LAND DEGRADATION ASSESSMENTS 46
4.2.1 Forest Monitoring in the Brazilian Atlantic Forest Biome 49
4.2.2 Pasture Degradation Mapping 51
4.3 POLICIES AND PROGRAM WITH RELEVANCE TO CONSERVATION AND REFORESTATION PROJECTS IN BRAZIL 54
4.3.1 International Programs, Schemes and Intitiatives 54
4.3.2 National Environmental Policies and Programs 58
4.3.3 Policies and Programs with Focus on the Brazilian Atlantic Forest 64
4.3.4 State-based Regulations in the Rio de Janeiro Federal State 65
4.4 LAND REHABILITATION AND REFORESTATION EFFORTS IN THE RJ FEDERAL STATE 66
4.4.1 Pasture Rehabilitation 66
4.4.2 Forest Restoration 67
5 MATERIAL AND METHODS 71
5.1 GEODATA AND SOFTWARE 71
5.2 PART (B): LAND COVER AND PASTURE DEGRADATION MAPPING 73
5.2.1 Field Survey of Degraded Pastures 73
5.2.2 Satellite Data Processing 75
5.2.3 Hot Spot Analysis 86
5.2.4 Relation of Slope Angle and Degradation Class 87
5.3 PART (C): PRIORITIZATION OF REFORESTATION SITES 87
5.3.1 Preliminary Study 88
5.3.2 Spatial Multi-Criteria Evaluation (SMCE) 90
5.3.3 Forest Scenario Development and Evaluation Using Landscape Metrics 94
6 RESULTS 97
6.1 PART (B): MAPPING OF PASTURE DEGRADATION 97
6.1.1 Categorization of Pasture Degradation 97
6.1.2 High-resolution SWIR band modelling 99
6.1.3 Land Cover Classification with Random Forests 101
6.1.4 Pasture Degradation Mapping Using Spectral Mixture Analysis and Field Data 103
6.1.5 Hot Spot Analysis of Pasture Degradation 106
6.1.6 Slope Influence on Pasture Degradation 107
6.2 PART (C): PRIORITIZATION OF REFORESTATION SITES USING SMCE 109
6.2.1 Characteristics of the Prioritized Areas 109
6.2.2 Forest Scenarios 113
7 DISCUSSION 117
7.1 PART (A/B): ASSESSMENT OF THE STATUS OF LAND DEGRADATION IN THE GUAPI-MACACU WATERSHED 117
7.2 PART (B): METHODOLOGICAL APPROACH FOR PASTURE DEGRADATION MAPPING 119
7.2.1 Satellite images for high-resolution LCC 119
7.2.2 High-resolution SWIR Band Modelling 120
7.2.3 Land Cover Classification 121
7.2.4 Pasture Degradation Mapping Approach 122
7.3 PART (C): SPATIAL PRIORITIZATION FOR REFORESTATION MEASURES 125
7.3.1 Identification of Priority Sites 125
7.3.2 Forest Scenarios 127
8 CONCLUSION AND RECOMMENDATIONS 131
9 OUTLOOK 135
REFERENCES 137
ANNEX I
Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:74326 |
Date | 22 April 2021 |
Creators | Naegeli de Torres, Friederike |
Contributors | Universität Leipzig |
Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
Language | English |
Detected Language | English |
Type | info:eu-repo/semantics/acceptedVersion, doc-type:doctoralThesis, info:eu-repo/semantics/doctoralThesis, doc-type:Text |
Rights | info:eu-repo/semantics/openAccess |
Page generated in 4.3565 seconds