Work

Geospatial approaches to mapping invasive alien plants: the neglected role of soil in remote sensing and GIS-based distribution models

Year: 2026

Type: article

Abstract: Introduction Invasive alien plants (IAPs) pose significant ecological and socio‐economic threats by disrupting native biodiversity, ecosystem functioning, and land productivity, underscoring the need for robust spatial tools to guide management interventions. Using a systematic literature review approach guided by the PRISMA framework, this study examines geospatial techniques for modelling Invasive Alien Plant (IAP) distributions, focusing on how climatic, land use/land cover (LULC), and soil variables are integrated to explain habitat suitability. Methods Across reviewed studies, remote sensing and Geographic Information Systems (GIS) emerged as central tools for multi‐scale monitoring and predictor integration. Medium‐resolution satellite imagery (Sentinel‐2, Landsat) commonly supported regional assessments of vegetation change and land-use dynamics, while very‐high‐resolution imagery enabled finer species‐ and structure-level discrimination, though at greater cost. These data streams were increasingly coupled with machine learning and statistical approaches to improve classification accuracy, though performance remained dependent on input data quality and adequate representation of key environmental drivers. Results A notable methodological imbalance emerged in predictor selection. Among the 24 studies included in the soil‐variable integration analysis, 71% (n = 17) excluded soil variables, relying instead on spectral, thermal, and structural proxies, land surface temperature, and canopy characteristics. Only 29% (n = 7) incorporated measured edaphic variables, such as soil pH, texture, and soil classification, which typically require field integration if observations, soil survey data, and digital soil maps within GIS-based frameworks. Discussion This omission stemmed largely from remote sensing’s inability to directly observe below‐ground properties and the logistical challenges of acquiring spatially explicit soil data. The review suggests that the limited integration of soil variables may constrain the ecological realism of geospatial invasion models, particularly in environmentally heterogeneous landscapes. Greater integration of soil, climate and LULC frameworks has the potential to improve ecological realism and predictive capability and warrant further investigation, particularly in under‐studied regions such as Sub-Saharan Africa.

Source: Frontiers in Remote Sensing

DOI: 10.3389/frsen.2026.1980112

Authors: Mosa Koloko, Thabang Maphanga, Siyabonga Benett Madonsela, Karabo Malakane

Institutions: University of the Western Cape, Cape Peninsula University of Technology, University of Limpopo

Language: English

Topic: Species Distribution and Climate Change

Subfield: Ecological Modeling

Field: Environmental Science

Domain: Physical Sciences

SDG: Life on land

Open Access status: gold

Estimated APC paid: $3,077