Towards better mapping of forest management patterns: A global allocation approach

Research output: Contribution to JournalArticleAcademicpeer-review

Abstract

Forests provide numerous ecosystem services, such as timber yields, biodiversity protection and climate change mitigation. The type of management has an effect on the provision of these services. Often the demands for these services can lead to conflict – wood harvest can negatively impact biodiversity and climate change mitigation capacity. Although forest management differences are important, spatially explicit data is lacking, in particular on a global scale. We present here a first systematic approach which integrates existing data to map forest management globally through downscaling national and subnational forest data. In our forest management classification, we distinguished between two levels of forest management, with three categories each. Level 1 comprised primary, naturally regrown and planted forests. Level 2 distinguished between different forest uses. We gathered documented locations, where these forest categories were observed, from the literature and a database on ecological diversity. We then performed multinomial logit regression and estimated the effect of 21 socio-economic and bio-physical predictor variables on the occurrence of a forest category. Model results on significance and effect direction of predictor variables were in line with findings of previous studies. Soil and environmental properties, forest conditions and accessibility are important determinants of the occurrence of forest management types. Based on the model results, likelihood maps were calculated and used to spatially allocate national extents of level 1 and level 2 forest categories. When compared to previous studies, our maps showed higher agreement than random samples. Deviations between observed and predicted plantation locations were mostly below 10 km. Our map provides an estimation of global forest management patterns, enhancing previous methodologies and making the best use of data available. Next to having multiple applications, for example within global conservation planning or climate change mitigation analyses, it visualizes the currently available data on forest management on a global level.
LanguageEnglish
Pages776-785
JournalForest Ecology and Management
Volume432
DOIs
StatePublished - 15 Jan 2019

Fingerprint

forest management
climate change
biodiversity
conservation planning
downscaling
allocation
ecosystem service
accessibility
timber
ecosystem services
plantation
socioeconomics
planning
plantations
taxonomy
methodology
climate change mitigation
effect
soil

Keywords

  • Global
  • Forest management
  • Land use mapping
  • Global forest resource assessment
  • FRA 2015
  • Location factors
  • Multinomial logit model

Cite this

@article{595b97e5a0f941a6b8224656730e6c8c,
title = "Towards better mapping of forest management patterns: A global allocation approach",
abstract = "Forests provide numerous ecosystem services, such as timber yields, biodiversity protection and climate change mitigation. The type of management has an effect on the provision of these services. Often the demands for these services can lead to conflict – wood harvest can negatively impact biodiversity and climate change mitigation capacity. Although forest management differences are important, spatially explicit data is lacking, in particular on a global scale. We present here a first systematic approach which integrates existing data to map forest management globally through downscaling national and subnational forest data. In our forest management classification, we distinguished between two levels of forest management, with three categories each. Level 1 comprised primary, naturally regrown and planted forests. Level 2 distinguished between different forest uses. We gathered documented locations, where these forest categories were observed, from the literature and a database on ecological diversity. We then performed multinomial logit regression and estimated the effect of 21 socio-economic and bio-physical predictor variables on the occurrence of a forest category. Model results on significance and effect direction of predictor variables were in line with findings of previous studies. Soil and environmental properties, forest conditions and accessibility are important determinants of the occurrence of forest management types. Based on the model results, likelihood maps were calculated and used to spatially allocate national extents of level 1 and level 2 forest categories. When compared to previous studies, our maps showed higher agreement than random samples. Deviations between observed and predicted plantation locations were mostly below 10 km. Our map provides an estimation of global forest management patterns, enhancing previous methodologies and making the best use of data available. Next to having multiple applications, for example within global conservation planning or climate change mitigation analyses, it visualizes the currently available data on forest management on a global level.",
keywords = "Global, Forest management, Land use mapping, Global forest resource assessment, FRA 2015, Location factors, Multinomial logit model",
author = "K. Schulze and Ziga Malek and P.H. Verburg",
year = "2019",
month = "1",
day = "15",
doi = "10.1016/j.foreco.2018.10.001",
language = "English",
volume = "432",
pages = "776--785",
journal = "Forest Ecology and Management",
issn = "0378-1127",
publisher = "Elsevier",

}

Towards better mapping of forest management patterns: A global allocation approach. / Schulze, K.; Malek, Ziga; Verburg, P.H.

In: Forest Ecology and Management, Vol. 432, 15.01.2019, p. 776-785.

Research output: Contribution to JournalArticleAcademicpeer-review

TY - JOUR

T1 - Towards better mapping of forest management patterns: A global allocation approach

AU - Schulze,K.

AU - Malek,Ziga

AU - Verburg,P.H.

PY - 2019/1/15

Y1 - 2019/1/15

N2 - Forests provide numerous ecosystem services, such as timber yields, biodiversity protection and climate change mitigation. The type of management has an effect on the provision of these services. Often the demands for these services can lead to conflict – wood harvest can negatively impact biodiversity and climate change mitigation capacity. Although forest management differences are important, spatially explicit data is lacking, in particular on a global scale. We present here a first systematic approach which integrates existing data to map forest management globally through downscaling national and subnational forest data. In our forest management classification, we distinguished between two levels of forest management, with three categories each. Level 1 comprised primary, naturally regrown and planted forests. Level 2 distinguished between different forest uses. We gathered documented locations, where these forest categories were observed, from the literature and a database on ecological diversity. We then performed multinomial logit regression and estimated the effect of 21 socio-economic and bio-physical predictor variables on the occurrence of a forest category. Model results on significance and effect direction of predictor variables were in line with findings of previous studies. Soil and environmental properties, forest conditions and accessibility are important determinants of the occurrence of forest management types. Based on the model results, likelihood maps were calculated and used to spatially allocate national extents of level 1 and level 2 forest categories. When compared to previous studies, our maps showed higher agreement than random samples. Deviations between observed and predicted plantation locations were mostly below 10 km. Our map provides an estimation of global forest management patterns, enhancing previous methodologies and making the best use of data available. Next to having multiple applications, for example within global conservation planning or climate change mitigation analyses, it visualizes the currently available data on forest management on a global level.

AB - Forests provide numerous ecosystem services, such as timber yields, biodiversity protection and climate change mitigation. The type of management has an effect on the provision of these services. Often the demands for these services can lead to conflict – wood harvest can negatively impact biodiversity and climate change mitigation capacity. Although forest management differences are important, spatially explicit data is lacking, in particular on a global scale. We present here a first systematic approach which integrates existing data to map forest management globally through downscaling national and subnational forest data. In our forest management classification, we distinguished between two levels of forest management, with three categories each. Level 1 comprised primary, naturally regrown and planted forests. Level 2 distinguished between different forest uses. We gathered documented locations, where these forest categories were observed, from the literature and a database on ecological diversity. We then performed multinomial logit regression and estimated the effect of 21 socio-economic and bio-physical predictor variables on the occurrence of a forest category. Model results on significance and effect direction of predictor variables were in line with findings of previous studies. Soil and environmental properties, forest conditions and accessibility are important determinants of the occurrence of forest management types. Based on the model results, likelihood maps were calculated and used to spatially allocate national extents of level 1 and level 2 forest categories. When compared to previous studies, our maps showed higher agreement than random samples. Deviations between observed and predicted plantation locations were mostly below 10 km. Our map provides an estimation of global forest management patterns, enhancing previous methodologies and making the best use of data available. Next to having multiple applications, for example within global conservation planning or climate change mitigation analyses, it visualizes the currently available data on forest management on a global level.

KW - Global

KW - Forest management

KW - Land use mapping

KW - Global forest resource assessment

KW - FRA 2015

KW - Location factors

KW - Multinomial logit model

U2 - 10.1016/j.foreco.2018.10.001

DO - 10.1016/j.foreco.2018.10.001

M3 - Article

VL - 432

SP - 776

EP - 785

JO - Forest Ecology and Management

T2 - Forest Ecology and Management

JF - Forest Ecology and Management

SN - 0378-1127

ER -