Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine
South Africa is reported to experience timber shortages as a result of growing timber demands and pulp production, coupled with the government’s reluctance to grant new forestry permits. Rampant timber theft in the country makes these circumstances worse. The emergence of cloud-based platforms, such...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Forests |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4907/11/12/1283 |
| _version_ | 1797546264192090112 |
|---|---|
| author | Sifiso Xulu Nkanyiso Mbatha Kabir Peerbhay Michael Gebreslasie |
| author_facet | Sifiso Xulu Nkanyiso Mbatha Kabir Peerbhay Michael Gebreslasie |
| author_sort | Sifiso Xulu |
| collection | DOAJ |
| description | South Africa is reported to experience timber shortages as a result of growing timber demands and pulp production, coupled with the government’s reluctance to grant new forestry permits. Rampant timber theft in the country makes these circumstances worse. The emergence of cloud-based platforms, such as Google Earth Engine (GEE), has greatly improved the accessibility and usability of high spatial and temporal Sentinel-1 and -2 data, especially in data-poor countries that lack high-performance computing systems for forest monitoring. Here, we demonstrate the potential of these resources for forest harvest detection. The results showed that Sentinel-1 data are efficient in detecting clear-cut events; both VH and VV backscatter signals decline sharply in accordance with clear-cutting and increase again when forest biomass increases. When correlated with highly responsive NDII, the VH and VV signals reached the best accuracies of 0.79 and 0.83, whereas the SWIR1 achieved –0.91. A Random Forest (RF) algorithm based on Sentinel-2 data also achieved over 90% accuracies for classifying harvested and forested areas. Overall, our study presents a cost-effective method for mapping clear-cut events in an economically important forestry area of South Africa while using GEE resources. |
| first_indexed | 2024-03-10T14:27:34Z |
| format | Article |
| id | doaj.art-d04bfe2325324fb0999d095c19c75db6 |
| institution | Directory Open Access Journal |
| issn | 1999-4907 |
| language | English |
| last_indexed | 2024-03-10T14:27:34Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Forests |
| spelling | doaj.art-d04bfe2325324fb0999d095c19c75db62023-11-20T22:53:13ZengMDPI AGForests1999-49072020-11-011112128310.3390/f11121283Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth EngineSifiso Xulu0Nkanyiso Mbatha1Kabir Peerbhay2Michael Gebreslasie3Department of Geography, University of the Free State, Phuthaditjhaba 9869, South AfricaDepartment of Geography and Environmental Studies, University of Zululand, KwaDlangezwa 3886, South AfricaSchool of Agricultural, Earth and Environmental Sciences, University of KwaZulu-Natal, Westville Campus, Durban 4000, South AfricaSchool of Agricultural, Earth and Environmental Sciences, University of KwaZulu-Natal, Westville Campus, Durban 4000, South AfricaSouth Africa is reported to experience timber shortages as a result of growing timber demands and pulp production, coupled with the government’s reluctance to grant new forestry permits. Rampant timber theft in the country makes these circumstances worse. The emergence of cloud-based platforms, such as Google Earth Engine (GEE), has greatly improved the accessibility and usability of high spatial and temporal Sentinel-1 and -2 data, especially in data-poor countries that lack high-performance computing systems for forest monitoring. Here, we demonstrate the potential of these resources for forest harvest detection. The results showed that Sentinel-1 data are efficient in detecting clear-cut events; both VH and VV backscatter signals decline sharply in accordance with clear-cutting and increase again when forest biomass increases. When correlated with highly responsive NDII, the VH and VV signals reached the best accuracies of 0.79 and 0.83, whereas the SWIR1 achieved –0.91. A Random Forest (RF) algorithm based on Sentinel-2 data also achieved over 90% accuracies for classifying harvested and forested areas. Overall, our study presents a cost-effective method for mapping clear-cut events in an economically important forestry area of South Africa while using GEE resources.https://www.mdpi.com/1999-4907/11/12/1283forestSentinel-1Sentinel-2harvestRF classificationremote sensing |
| spellingShingle | Sifiso Xulu Nkanyiso Mbatha Kabir Peerbhay Michael Gebreslasie Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine Forests forest Sentinel-1 Sentinel-2 harvest RF classification remote sensing |
| title | Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine |
| title_full | Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine |
| title_fullStr | Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine |
| title_full_unstemmed | Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine |
| title_short | Detecting Harvest Events in Plantation Forest Using Sentinel-1 and -2 Data via Google Earth Engine |
| title_sort | detecting harvest events in plantation forest using sentinel 1 and 2 data via google earth engine |
| topic | forest Sentinel-1 Sentinel-2 harvest RF classification remote sensing |
| url | https://www.mdpi.com/1999-4907/11/12/1283 |
| work_keys_str_mv | AT sifisoxulu detectingharvesteventsinplantationforestusingsentinel1and2dataviagoogleearthengine AT nkanyisombatha detectingharvesteventsinplantationforestusingsentinel1and2dataviagoogleearthengine AT kabirpeerbhay detectingharvesteventsinplantationforestusingsentinel1and2dataviagoogleearthengine AT michaelgebreslasie detectingharvesteventsinplantationforestusingsentinel1and2dataviagoogleearthengine |