Remote sensing approach for spatial planning of land management interventions in West African savannas

Forest management, agroforestry and tree planting are some of the key approaches to sustainable rural development, and climate change adaptation and mitigation in West African savannas. However, the planning of land management interventions is hindered by the lack of information at relevant spatial...

Full description

Bibliographic Details
Main Authors: Heiskanen, J., Liu, J., Valbuena, R., Aynekulu, Ermias, Packalen, P., Pellikka, P.
Format: Journal Article
Language:Inglés
Published: Elsevier 2017
Subjects:
Online Access:https://hdl.handle.net/10568/99295
_version_ 1855524782938783744
author Heiskanen, J.
Liu, J.
Valbuena, R.
Aynekulu, Ermias
Packalen, P.
Pellikka, P.
author_browse Aynekulu, Ermias
Heiskanen, J.
Liu, J.
Packalen, P.
Pellikka, P.
Valbuena, R.
author_facet Heiskanen, J.
Liu, J.
Valbuena, R.
Aynekulu, Ermias
Packalen, P.
Pellikka, P.
author_sort Heiskanen, J.
collection Repository of Agricultural Research Outputs (CGSpace)
description Forest management, agroforestry and tree planting are some of the key approaches to sustainable rural development, and climate change adaptation and mitigation in West African savannas. However, the planning of land management interventions is hindered by the lack of information at relevant spatial resolution. We examined predictive models for mapping various tree, soil and species diversity attributes with a comparison of RapidEye and Landsat imagery. The field data was collected in the vicinity of the community-managed forest in southern Burkina Faso, where the main environmental threats are agricultural expansion and fuelwood extraction. The modelling was done using Random Forest algorithm. According to our results, tree crown cover and correlated attributes, such as basal area and tree species richness, were predicted most accurately. High spatial resolution RapidEye imagery provided the best accuracy but difference to medium resolution Landsat imagery was negligible for most attributes. Burn scar masked Landsat time series performed similar to dry season single date Landsat imagery, but the former avoids image selection and mosaicking, and could be less sensitive to artifacts due to the burn scars. The presented approach provides valuable information on important tree, soil and species diversity attributes for spatial planning of land management interventions.
format Journal Article
id CGSpace99295
institution CGIAR Consortium
language Inglés
publishDate 2017
publishDateRange 2017
publishDateSort 2017
publisher Elsevier
publisherStr Elsevier
record_format dspace
spelling CGSpace992952025-02-19T13:42:22Z Remote sensing approach for spatial planning of land management interventions in West African savannas Heiskanen, J. Liu, J. Valbuena, R. Aynekulu, Ermias Packalen, P. Pellikka, P. canopy cover carbon stocks biodiversity rapideye landsat Forest management, agroforestry and tree planting are some of the key approaches to sustainable rural development, and climate change adaptation and mitigation in West African savannas. However, the planning of land management interventions is hindered by the lack of information at relevant spatial resolution. We examined predictive models for mapping various tree, soil and species diversity attributes with a comparison of RapidEye and Landsat imagery. The field data was collected in the vicinity of the community-managed forest in southern Burkina Faso, where the main environmental threats are agricultural expansion and fuelwood extraction. The modelling was done using Random Forest algorithm. According to our results, tree crown cover and correlated attributes, such as basal area and tree species richness, were predicted most accurately. High spatial resolution RapidEye imagery provided the best accuracy but difference to medium resolution Landsat imagery was negligible for most attributes. Burn scar masked Landsat time series performed similar to dry season single date Landsat imagery, but the former avoids image selection and mosaicking, and could be less sensitive to artifacts due to the burn scars. The presented approach provides valuable information on important tree, soil and species diversity attributes for spatial planning of land management interventions. 2017-05 2019-02-06T11:59:49Z 2019-02-06T11:59:49Z Journal Article https://hdl.handle.net/10568/99295 en Limited Access Elsevier Heiskanen, J.; Liu, J.; Valbuena, R.; Aynekulu, E.; Packalen, P.; Pellikka, P. 2017. Remote sensing approach for spatial planning of land management interventions in West African savannas. Journal of Arid Environments 140, 29-41.
spellingShingle canopy cover
carbon stocks
biodiversity
rapideye
landsat
Heiskanen, J.
Liu, J.
Valbuena, R.
Aynekulu, Ermias
Packalen, P.
Pellikka, P.
Remote sensing approach for spatial planning of land management interventions in West African savannas
title Remote sensing approach for spatial planning of land management interventions in West African savannas
title_full Remote sensing approach for spatial planning of land management interventions in West African savannas
title_fullStr Remote sensing approach for spatial planning of land management interventions in West African savannas
title_full_unstemmed Remote sensing approach for spatial planning of land management interventions in West African savannas
title_short Remote sensing approach for spatial planning of land management interventions in West African savannas
title_sort remote sensing approach for spatial planning of land management interventions in west african savannas
topic canopy cover
carbon stocks
biodiversity
rapideye
landsat
url https://hdl.handle.net/10568/99295
work_keys_str_mv AT heiskanenj remotesensingapproachforspatialplanningoflandmanagementinterventionsinwestafricansavannas
AT liuj remotesensingapproachforspatialplanningoflandmanagementinterventionsinwestafricansavannas
AT valbuenar remotesensingapproachforspatialplanningoflandmanagementinterventionsinwestafricansavannas
AT aynekuluermias remotesensingapproachforspatialplanningoflandmanagementinterventionsinwestafricansavannas
AT packalenp remotesensingapproachforspatialplanningoflandmanagementinterventionsinwestafricansavannas
AT pellikkap remotesensingapproachforspatialplanningoflandmanagementinterventionsinwestafricansavannas