Knowledge based multi-source, time series classification: A case study of central region of Kenya

Mwaniki, Mercy W.; Möller, S. Matthias

doi:10.1016/j.apgeog.2015.03.005

Cited by 11 publications

(5 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the increase in OA happened in both models (XGBoost and RF) signify the general importance of the indices in the classification. This result confirms previous studies that have shown that the addition of indices to the spectral bands results in classification improvements [89]. For example, Matongera et al [43] achieved a 19.94% improvement in classification accuracy after the addition of vegetation indices; Mudereri et al [78] had a 1% improvement in detection of Striga weed; while Hurskainen et al [88] achieved an increase of 16.5 percentage points after the addition of auxiliary features which included vegetation and topographic indices in a heterogeneous savanna landscape landcover classification.…”

Section: Model Evaluation and Spatial Coveragesupporting

confidence: 92%

“…Additionally, DEM and topographic variables such as slope, aspect [85], and Terrain Wetness Index (TWI) [86], which are known to aid improved decision making for classification tasks, were also included [87][88][89]. Furthermore, these topographic variables do affect temperature, precipitation, radiation regimes, and moisture demands attributes that indirectly affect vegetation dynamics and microclimates [90].…”

Section: Extracting Predictor Variables For Classificationmentioning

confidence: 99%

See 1 more Smart Citation

Mapping Opuntia stricta in the Arid and Semi-Arid Environment of Kenya Using Sentinel-2 Imagery and Ensemble Machine Learning Classifiers

et al. 2021

View full text Add to dashboard Cite

Globally, grassland biomes form one of the largest terrestrial covers and present critical social–ecological benefits. In Kenya, Arid and Semi-arid Lands (ASAL) occupy 80% of the landscape and are critical for the livelihoods of millions of pastoralists. However, they have been invaded by Invasive Plant Species (IPS) thereby compromising their ecosystem functionality. Opuntia stricta, a well-known IPS, has invaded the ASAL in Kenya and poses a threat to pastoralism, leading to livestock mortality and land degradation. Thus, identification and detailed estimation of its cover is essential for drawing an effective management strategy. The study aimed at utilizing the Sentinel-2 multispectral sensor to detect Opuntia stricta in a heterogeneous ASAL in Laikipia County, using ensemble machine learning classifiers. To illustrate the potential of Sentinel-2, the detection of Opuntia stricta was based on only the spectral bands as well as in combination with vegetation and topographic indices using Extreme Gradient Boost (XGBoost) and Random Forest (RF) classifiers to detect the abundance. Study results showed that the overall accuracies of Sentinel 2 spectral bands were 80% and 84.4%, while that of combined spectral bands, vegetation, and topographic indices was 89.2% and 92.4% for XGBoost and RF classifiers, respectively. The inclusion of topographic indices that enhance characterization of biological processes, and vegetation indices that minimize the influence of soil and the effects of atmosphere, contributed by improving the accuracy of the classification. Qualitatively, Opuntia stricta spatially was found along river banks, flood plains, and near settlements but limited in forested areas. Our results demonstrated the potential of Sentinel-2 multispectral sensors to effectively detect and map Opuntia stricta in a complex heterogeneous ASAL, which can support conservation and rangeland management policies that aim to map and list threatened areas, and conserve the biodiversity and productivity of rangeland ecosystems.

show abstract

Section: Model Evaluation and Spatial Coveragesupporting

confidence: 92%

Section: Extracting Predictor Variables For Classificationmentioning

confidence: 99%

Mapping Opuntia stricta in the Arid and Semi-Arid Environment of Kenya Using Sentinel-2 Imagery and Ensemble Machine Learning Classifiers

et al. 2021

View full text Add to dashboard Cite

show abstract

“…A study by Wang et al, (2018), in the Xitiaoxi River Basin, analyzed the spatial patterns of LULC changes from 1985 to 2008 and reported that dominant trend of land-use conversion was between forest-grass land and agricultural land, and the diminishing portion of forest-grass land and agricultural land contributed to the expansion of urban land during the period 1985-2008. Similar trend was also pointed out by Birhanu et al, (2019), in the Gumara catchment, Ethiopia, between 1986and 2015 where the area under forest and grass land was about 11 and 18%, respectively, in 1986, which reduced to 5 and 10%, respectively, in 2015. Also noted was that cultivated land increased from 70% in 1986 to 82% in 2015.…”

Section: Introductionsupporting

confidence: 84%

“…The main reason for the above changes could be attributed the rising human population and economic growth in the area. A study by Mwaniki and Möller (2015) , reported similar results in some sections of Central Kenya about a massive forest on the decline between 1995 and 2002, and a slight rise of forest areas between 2002 and 2010. At this time the Kenyan government had addressed the issue of deforestation and was putting measures to curb the problem for example through the introduction of the Nyayo Tea Zones which created as a "tea boundary" at the national park (Willkomm et al, 2016).…”

Section: Spatial and Temporal Variations In Land Use And Land Cover Changesmentioning

confidence: 54%

Spatial and temporal variations in land use and land cover in the Njoro and Kamweti River catchments, Kenya

Koskey

Ogendi

M’Erimba

et al. 2021

EAJSTI

View full text Add to dashboard Cite

The Njoro and Kamweti River catchments are productive catchments that have and continue to experience major land-use changes with consequences on land cover and the associated environmental resources. It is, therefore, crucial to understand the type of changes occurring, spatial patterns, and the rates at which these changes are occurring. In this study, we quantified the changes in land use and land cover that occurred between 1988 and 2019 identifying areas of change and the average annual rate of change. Thematic Mappers (TM) and Enhanced Thematic Mappers Plus (ETM +) and Sentinel images were obtained for 1988 and 2019. Ground truthing was carried out to enable us to verify the accuracy of the remotely sensed data using in-situ observations to refine the classification output. The results obtained indicated that both catchments have experienced intense land-use changes but at different levels. Njoro River catchment’s forest cover and shrubland had decreased at a rate of 6.06 Km2/year and 0.92 Km2/year respectively and the most increase was recorded in farmlands (3.11 Km2/year) as the other land use classes also increased. In the Kamweti River catchment, forest cover showed a decrease at a rate of 0.21 Km2/year, and farmlands also a slight decrease of 0.1 Km2/year while the other land cover classes increased in area coverage during the period 1988-2019. The changes in land use and land cover were attributed to increased demand for food and housing and thus continued degrading the two catchments especially the Njoro River catchment. Results obtained indicated that anthropogenic activities were the major contributing factors to the changes in Land Use Land Cover experienced in both catchments. We recommend continued analysis of the trends and rates of land cover conversions owing to their potential use by development planners.

show abstract

“…This transformation is often accompanied by losses in forestlands, grasslands, wetlands, and other features of ecological importance (Lark et al 2020;Zeng et al 2018). Empirical evidence suggests that human actions are central to LULCCs (Mwaniki and Möller 2015). These changes vary across diverse spatial scales and magnitudes based on underlying biophysical and climatic conditions.…”

Section: Introductionmentioning

confidence: 99%

Modelling cropland expansion and its drivers in Trans Nzoia County, Kenya

Kipkulei

Bellingrath-Kimura

Lana

et al. 2022

Model. Earth Syst. Environ.

View full text Add to dashboard Cite

Population growth and increasing demand for agricultural production continue to drive global cropland expansions. These expansions lead to the overexploitation of fragile ecosystems, propagating land degradation, and the loss of natural diversity. This study aimed to identify the factors driving land use/land cover changes (LULCCs) and subsequent cropland expansion in Trans Nzoia County in Kenya. Landsat images were used to characterize the temporal LULCCs in 30 years and to derive cropland expansions using change detection. Logistic regression (LR), boosted regression trees (BRTs), and evidence belief functions (EBFs) were used to model the potential drivers of cropland expansion. The candidate variables included proximity and biophysical, climatic, and socioeconomic factors. The results showed that croplands replaced other natural land covers, expanding by 38% between 1990 and 2020. The expansion in croplands has been at the expense of forestland, wetland, and grassland losses, which declined in coverage by 33%, 71%, and 50%, respectively. All the models predicted elevation, proximity to rivers, and soil pH as the critical drivers of cropland expansion. Cropland expansions dominated areas bordering the Mt. Elgon forest and Cherangany hills ecosystems. The results further revealed that the logistic regression model achieved the highest accuracy, with an area under the curve (AUC) of 0.96. In contrast, EBF and the BRT models depicted AUC values of 0.86 and 0.77, respectively. The findings exemplify the relationships between different potential drivers of cropland expansion and contribute to developing appropriate strategies that balance food production and environmental conservation.

show abstract

Knowledge based multi-source, time series classification: A case study of central region of Kenya

Cited by 11 publications

References 43 publications

Mapping Opuntia stricta in the Arid and Semi-Arid Environment of Kenya Using Sentinel-2 Imagery and Ensemble Machine Learning Classifiers

Mapping Opuntia stricta in the Arid and Semi-Arid Environment of Kenya Using Sentinel-2 Imagery and Ensemble Machine Learning Classifiers

Spatial and temporal variations in land use and land cover in the Njoro and Kamweti River catchments, Kenya

Modelling cropland expansion and its drivers in Trans Nzoia County, Kenya

Contact Info

Product

Resources

About