Long-Term Land Use/Land Cover Change Assessment of the Kilombero Catchment in Tanzania Using Random Forest Classification and Robust Change Vector Analysis

Abstract

Information about land use/land cover (LULC) and their changes is useful for different stakeholders to assess future pathways of sustainable land use for food production as well as for nature conservation. In this study, we assess LULC changes in the Kilombero catchment in Tanzania, an important area of recent development in East Africa. LULC change is assessed in two ways: first, post-classification comparison (PCC) which allows us to directly assess changes from one LULC class to another, and second, spectral change detection. We perform LULC classification by applying random forests (RF) on sets of multitemporal metrics that account for seasonal within-class dynamics. For the spectral change detection, we make use of the robust change vector analysis (RCVA) and determine those changes that do not necessarily lead to another class. The combination of the two approaches enables us to distinguish areas that show (a) only PCC changes, (b) only spectral changes that do not affect the classification of a pixel, (c) both types of change, or (d) no changes at all. Our results reveal that only one-quarter of the catchment has not experienced any change. One-third shows both, spectral changes and LULC conversion. Changes detected with both methods predominantly occur in two major regions, one in the West of the catchment, one in the Kilombero floodplain. Both regions are important areas of food production and economic development in Tanzania. The Kilombero floodplain is a Ramsar protected area, half of which was converted to agricultural land in the past decades. Therefore, LULC monitoring is required to support sustainable land management. Relatively poor classification performances revealed several challenges during the classification process. The combined approach of PCC and RCVA allows us to detect spatial patterns of LULC change at distinct dimensions and intensities. With the assessment of additional classifier output, namely class-specific per-pixel classification probabilities and derived parameters, we account for classification uncertainty across space. We overlay the LULC change results and the spatial assessment of classification reliability to provide a thorough picture of the LULC changes taking place in the Kilombero catchment.

Monitoring of Urban Sprawl and Densification Processes in Western Germany in the Light of SDG Indicator 11.3.1 Based on an Automated Retrospective Classification Approach

“…Following the cloud mask, we calculated spectral indices, namely the Normalized Difference Vegetation Index (NDVI) [44,45] and the Normalized Difference Water Index (NDWI) [46], as well as Tasseled cap wetness, greenness and brightness (Table 1) [47]. We implemented a pixel-based compositing approach before the classification procedure and calculated temporal metrics [48][49][50][51] (e.g., 10th, 25th, 50th, 75th, 80th and 90th percentiles of annual time series, as well as a minima and maxima for both annual and growing season time series), which resulted in composites for each target year. We extracted eleven types of image features from each compositing result, including six surface reflectance (green, blue, red, NIR, SWIR1 and SWIR2) bands and the spectral indices (NDVI, NDWI, TC greenness, wetness, brightness).…”

“…Furthermore, the population data at a district level were used to compare the rates of urban growth and population change and calculate the ratio of land consumption rate (LCR) to population growth rate (PGR). It is based on the ratio of Land Consumption Rate and Population Growth Rate (LCRPGR) (Equations ( 1)-( 3)) [8,51,71],…”

Complex Socio-Ecological Systems: Translating Narratives into Future Land Use and Land Cover Scenarios in the Kilombero Catchment, Tanzania

“…As a consequence of the high land pressure, the LULC of the Kilombero catchment has considerably changed in recent decades [5,19,27,[34][35][36]. By 2014, around 18% of the catchment's natural vegetation had been converted to cropland [36].…”

“…As a consequence of the high land pressure, the LULC of the Kilombero catchment has considerably changed in recent decades [5,19,27,[34][35][36]. By 2014, around 18% of the catchment's natural vegetation had been converted to cropland [36]. Wildlife populations declined considerably [20,37] and former wildlife corridors (Figure 1) are inactive [18,19,23,34,37,38].…”

“…In total, the protected area categories considered in this study cover an area of 1,670,296 ha (41% of the catchment) and are characterized by differing conservation goals and levels of enforcement (Table 1) [18,21,[39][40][41]. In the Ramsar wetland, for instance, rice now constitutes the dominant LULC category [36] although agriculture and grazing are legally prohibited [40] (section 32 (2) and section 34 (1b)). Facing these ongoing anthropogenic disturbances, the long-term viability of the ecosystem and the Ramsar status of the wetland are endangered [21,42].…”