Mapping Prosopis glandulosa (mesquite) in the semi-arid environment of South Africa using high-resolution WorldView-2 imagery and machine learning classifiers

Adam, Elhadi; Mureriwa, Nyasha; Newete, Solomon W.

doi:10.1016/j.jaridenv.2017.05.001

“…As such, remote sensing (RS) offers cost-efficient means to assess the distribution of alien plant species and monitor their spread [20] while allowing monitoring areas that are inaccessible. Very High Resolution (VHS) sensors, which collect data at metric or sub-metric resolution, can provide very high accuracy data for landscape change analysis and can distinguish Prosopis from native tree species, such as Vachellia tortilis [17,[21][22][23]. However, these data can't be used to study historical changes of Prosopis cover, as the data are tasked, and the archives do not date back to early introduction periods (e.g., to the 1980s).…”

Section: Introductionmentioning

confidence: 99%

Spatial Evolution of Prosopis Invasion and its Effects on LULC and Livelihoods in Baringo, Kenya

Mbaabu

¹

,

Ng

²

,

Schaffner

³

et al. 2019

View full text Add to dashboard Cite

Woody alien plant species have been deliberately introduced globally in many arid and semi-arid regions, as they can provide services and goods to the rural poor. However, some of these alien trees and shrubs have become invasive over time, with important impacts on biodiversity, ecosystem services, and human well-being. Prosopis was introduced in Baringo County, Kenya, in the 1980s, but since then, it has spread rapidly from the original plantations to new areas. To assess land-use and land-cover (LULC) changes and dynamics in Baringo, we used a combination of dry and wet season Landsat satellite data acquired over a seven-year time interval between 1988–2016, and performed a supervised Random Forest classification. For each time interval, we calculated the extent of Prosopis invasion, rates of spread, gains and losses of specific LULC classes, and the relative importance of Prosopis invasion on LULC changes. The overall accuracy and kappa coefficients of the LULC classifications ranged between 98.1–98.5% and 0.93–0.96, respectively. We found that Prosopis coverage increased from 882 ha in 1988 to 18,792 ha in 2016. The highest negative changes in LULC classes were found for grasslands (−6252 ha; −86%), irrigated cropland (−849 ha; −57%), Vachellia tortilis-dominated vegetation (−3602 ha; −42%), and rainfed cropland (−1432 ha; −37%). Prosopis invasion alone directly accounted for over 30% of these negative changes, suggesting that Prosopis invasion is a key driver of the observed LULC changes in Baringo County. Although the management of Prosopis by utilization has been promoted in Baringo for 10–15 years, the spread of Prosopis has not stopped or slowed down. This suggests that Prosopis management in Baringo and other invaded areas in East Africa needs to be based on a more integrated approach.

show abstract

“…This method of coverage assessment can readily be scaled up to broader spatial scales (several km 2 ) using both rotary and fixed wing unmanned aerial vehicles (UAV) and we have recently applied the technique to airborne surveys of microbial mats on landscape spatial scales in terrestrial Antarctic deserts ( Bollard-Breen et al, 2015 ). In deserts the UAV approach could also be applied to vascular plants that display characteristic patterns of colonization related to resource limitation ( Rietkerk et al, 2004 ), since although satellite multispectral imaging has been used to map vascular plants in deserts the resolution remains limited ( Adam et al, 2017 ).…”

Section: Resultsmentioning

confidence: 99%

Advanced Photogrammetry to Assess Lichen Colonization in the Hyper-Arid Namib Desert

Hinchliffe

¹

,

Bollard-Breen

²

,

Cowan

³

et al. 2017

View full text Add to dashboard Cite

The hyper-arid central region of the Namib Desert is characterized by quartz desert pavement terrain that is devoid of vascular plant covers. In this extreme habitat the only discernible surface covers are epilithic lichens that colonize exposed surfaces of quartz rocks. These lichens are highly susceptible to disturbance and so field surveys have been limited due to concerns about disturbing this unusual desert feature. Here we present findings that illustrate how non-destructive surveys based upon advanced photogrammetry techniques can yield meaningful and novel scientific data on these lichens. We combined ‘structure from motion analysis,’ computer vision and GIS to create 3-dimensional point clouds from two-dimensional imagery. The data were robust in its application to estimating absolute lichen cover. An orange Stellarangia spp. assemblage had coverage of 22.8% of available substrate, whilst for a black Xanthoparmelia spp. assemblage coverage was markedly lower at 0.6% of available substrate. Hyperspectral signatures for both lichens were distinct in the near-infra red range indicating that Xanthoparmelia spp. was likely under relatively more moisture stress than Stellarangia spp. at the time of sampling, and we postulate that albedo effects may have contributed to this in the black lichen. Further transformation of the data revealed a colonization preference for west-facing quartz surfaces and this coincides with prevailing winds for marine fog that is the major source of moisture in this system. Furthermore, a three-dimensional ‘fly through’ of the lichen habitat was created to illustrate how the application of computer vision in microbiology has further potential as a research and education tool. We discuss how advanced photogrammetry could be applied in astrobiology using autonomous rovers to add quantitative ecological data for visible surface colonization on the surface of Mars.

show abstract

“…The early detection and mapping of invasive species are essential to formulating effective management strategies and preventing further spread into non-invaded lands [19]. As such, remote sensing (RS) offers cost-efficient means to assess the distribution of alien plant species and monitor their spread [20] while allowing monitoring areas that are inaccessible.Very High Resolution (VHS) sensors, which collect data at metric or sub-metric resolution, can provide very high accuracy data for landscape change analysis and can distinguish Prosopis from native tree species, such as Vachellia tortilis [17,[21][22][23]. However, these data can't be used to study historical changes of Prosopis cover, as the data are tasked, and the archives do not date back to early introduction periods (e.g., to the 1980s).…”

mentioning

confidence: 99%

“…Very High Resolution (VHS) sensors, which collect data at metric or sub-metric resolution, can provide very high accuracy data for landscape change analysis and can distinguish Prosopis from native tree species, such as Vachellia tortilis [17,[21][22][23]. However, these data can't be used to study historical changes of Prosopis cover, as the data are tasked, and the archives do not date back to early introduction periods (e.g., to the 1980s).…”

mentioning

confidence: 99%

Spatial Evolution of Prosopis Invasion and its Effects on LULC and Livelihoods in Baringo, Kenya

Rima¹,

Ng²,

Schaffner³

et al. 2019

Preprint

0

View full text Add to dashboard Cite

Woody alien plant species have been deliberately introduced globally in many arid and semi-arid regions, as they can provide services and goods to the rural poor. However, some of these alien trees and shrubs have become invasive over time, with important impacts on biodiversity, ecosystem services, and human well-being. Prosopis was introduced in Baringo County, Kenya, in the 1980s, but since then, it has spread rapidly from the original plantations to new areas. To assess land-use and land-cover (LULC) changes and dynamics in Baringo, we used a combination of dry and wet season Landsat satellite data acquired over a seven-year time interval between 1988-2016, and performed a supervised Random Forest classification. For each time interval, we calculated the extent of Prosopis invasion, rates of spread, gains and losses of specific LULC classes, and the relative importance of Prosopis invasion on LULC changes. The overall accuracy and kappa coefficients of the LULC classifications ranged between 98.1-98.5% and 0.93-0.96, respectively. We found that Prosopis coverage increased from 882 ha in 1988 to 18,792 ha in 2016. The highest negative changes in LULC classes were found for grasslands (−6252 ha; −86%), irrigated cropland (−849 ha; −57%), Vachellia tortilis-dominated vegetation (−3602 ha; −42%), and rainfed cropland (−1432 ha; −37%). Prosopis invasion alone directly accounted for over 30% of these negative changes, suggesting that Prosopis invasion is a key driver of the observed LULC changes in Baringo County. Although the management of Prosopis by utilization has been promoted in Baringo for 10-15 years, the spread of Prosopis has not stopped or slowed down. This suggests that Prosopis management in Baringo and other invaded areas in East Africa needs to be based on a more integrated approach. ecosystems and human welfare [1,2]. These species are a major threat to the environment, because they can: (1) suppress or replace native biodiversity, (2) alter ecosystem functions and services, and (3) cause significant economic damage, costing economies millions of dollars [3].African arid and semi-arid lands (ASALs) were severely degraded by the prolonged Sahelian drought of the 1970s [4][5][6], which prompted the prioritization of tree planting. In Kenya, species from the South and Central American genus Prosopis were selected for screening because they had shown potential in the rehabilitation of quarries [7,8]. Prosopis spp., also known as Mesquite and locally as 'Mathenge' or 'Promi', are perennial, multi-stemmed shrubby or single-canopy trees [9] which are nitrogen-fixing and tolerant to arid conditions [10]. In the 1980s, Prosopis juliflora (Sw.) DC., Prosopis pallida (Willd.), and Prosopis chilensis (Molina) Stuntz were planted in Baringo County [11]. The planted Prosopis trees were initially appreciated due to their ability to grow in degraded and barren landscapes where native vegetation could not grow, thus reducing soil erosion and dust storms, while providing shade and pods that served as fodder for lives...

show abstract

Mapping Prosopis glandulosa (mesquite) in the semi-arid environment of South Africa using high-resolution WorldView-2 imagery and machine learning classifiers

Cited by 33 publications

References 44 publications

Spatial Evolution of Prosopis Invasion and its Effects on LULC and Livelihoods in Baringo, Kenya

Spatial Evolution of Prosopis Invasion and its Effects on LULC and Livelihoods in Baringo, Kenya

Advanced Photogrammetry to Assess Lichen Colonization in the Hyper-Arid Namib Desert

Spatial Evolution of Prosopis Invasion and its Effects on LULC and Livelihoods in Baringo, Kenya

Contact Info

Product

Resources

About