Mapping Prosopis juliflora in West Somaliland with Landsat 8 Satellite Imagery and Ground Information. (14th October 2016)

• Record Type: Journal Article
• Authors: Meroni, Michele; Ng, Wai‐Tim; Rembold, Felix; Leonardi, Ugo; Atzberger, Clement; Gadain, Hussein; Shaiye, Muse
• Is Part Of: Land degradation & development. Volume 28:Number 2(2017)
• Page Start: 494
• Page End: 506
• Publication Date: 2016-10-14
• Subjects: invasive alien species -- Prosopis juliflora -- mapping -- remote sensing -- Landsat 8 -- Somaliland; Land degradation -- Periodicals; Soil conservation -- Periodicals; Reclamation of land -- Periodicals; Land use -- Periodicals; Economic development -- Environmental aspects -- Periodicals; 333.7315
• DOI: 10.1002/ldr.2611
• Languages: English
• ISSNs: 1085-3278

Abstract

Abstract: Prosopis juliflora is a drought‐tolerant fast‐growing tree species originating from South and Central America with a high invasion potential in arid and semi‐arid areas in Africa. It was introduced in Somaliland in the 1980s and is reported to have spread vigorously since. Despite being recognized as a serious issue in the country, the actual scale of the problem is unknown. In this study, we mapped the species in a study area that includes the capital, Hargeisa, using Landsat 8 satellite imagery. During a field campaign in 2015, we collected canopy‐level spectral signatures of P . juliflora and native trees to analyse the potential use of spectral data in discriminating the invasive species. P. juliflora was found to be generally distinguishable because of its greater vigour during the dry season. We tested the accuracy of the random forest classifier and different classification set‐ups, varying the spatial resolution (original 30 m vs pan‐sharpened 15 m) and image acquisition dates (during the wet season, the dry season and a combination of the two). Best overall accuracy (84%) was achieved by using pan‐sharpened data from the two seasons. About 30 years since its introduction, the invasive species was detected in 9% of the total investigated area with highest occurrence in the proximity of human settlements and along seasonal water courses. © 2016 The Authors. Land Degradation and Development published by John Wiley & Sons, Ltd.