| Literature DB >> 32410311 |
Maliwan Namkhan1, George A Gale1, Tommaso Savini1, Naruemon Tantipisanuh1,2.
Abstract
Despite containing extraordinary levels of biodiversity, lowland (<200 m asl) tropical forests are extremely threatened globally. Southeast Asia is an area of high species richness and endemicity under considerable anthropogenic threat with, unfortunately, scant focus on its lowland forests. We estimated extent of lowland forest loss from 1998 to 2018, including inside protected areas and determined the vulnerability of this remaining forest. Maximum likelihood classification techniques were used to classify Landsat images to estimate lowland forest cover in 1998 and 2018. We used Bayesian belief networks with 20 variables to evaluate vulnerability of the forest that remained in 2018. Analyses were conducted at two spatial scales: landscape patch (analogous to ecoregion) and country level. Over 20 years, >120,000 km2 of forest (50% of forest present in 1998) was lost. Of the 14 lowland forest patches, 6 lost >50% of their area. At the country scale, Cambodia had the greatest deforestation (>47,500 km2 ). In 2018, 18% of the lowlands were forested, and 20% of these forests had some formal protection. Approximately 50% of the lowland forest inside protected areas (c. 11,000 km2 ) was also lost during the study period. Most lowland forest remaining is highly vulnerable; eight landscape patches had >50% categorized as such. Our results add to a growing body of evidence that the presence of protected areas alone will not prevent further deforestation. We suggest that more collaborative conservation strategies with local communities that accommodate conservation concessions specifically for lowland forests are urgently needed to prevent further destruction of these valuable habitats.Keywords: Bayesian belief networks; Landsat image classification; Landsat 陆地卫星图像分类; clasificación de imágenes Landsat; forest loss; forest vulnerability; manejo de áreas protegidas; protected area management; pérdida del bosque; redes bayesianas de opinión; vulnerabilidad del bosque; 保护区管理; 森林丧失; 森林脆弱性; 贝叶斯信念网络
Mesh:
Year: 2020 PMID: 32410311 DOI: 10.1111/cobi.13538
Source DB: PubMed Journal: Conserv Biol ISSN: 0888-8892 Impact factor: 6.560