Assessment of badland susceptibility and its governing factors using a random forest approach. Application to the Upper Llobregat River Basin and Catalonia (Spain).

Badlands are considered hotspots of sediment production, contributing to large fractions of the sediment budget of catchments and river basins. The erosion rates of these areas can exceed 100 t ha-1 y-1, leading to significant environmental and economic impacts. This research aims to assess badland susceptibility and the relevance of its governing factors at different spatial scales using the well-known machine learning approach random forest (RF). The Upper Llobregat River Basin (ULRB, approx. 500 km2) and Catalonia (approx. 32,000 km2) have been selected as study areas. Previous studies stated that the RF approach is successful at making predictions for the same area where it has been trained, but the results of testing it in a different area remains unexplored. This work aims to evaluate the feasibility of upscaling to the large region of Catalonia a RF model trained in the small ULRB area. Two badland datasets of both small and large regions and a total of eleven governing factors have been used to determine the areas susceptible to badlands. Models performance has been analyzed through three different evaluation metrics: overall accuracy, kappa coefficient and area under receiver operating characteristic curve (AUC). The outcomes of this work confirmed that RF is a powerful tool for badland susceptibility analysis, specially when predictions are made in the same scale and spatial context where the model has been trained. Upscaling a RF model defined in the ULRB to the large area of Catalonia has been possible, but improved results have been obtained when the training of the models has directly been performed in the large region. Our final RF modelling results have facilitated the development of a large scale (32,000 km2) Badland Susceptibility Map for the full extension of Catalonia with a predictive overall accuracy of 97%, which strongly emphasizes lithology and Normalized Difference Vegetation Index (NDVI) as the main conditioning factors of badland distribution.

Assessing vegetation recovery in reclaimed opencast mines of the Teruel coalfield (Spain) using Landsat time series and boosted regression trees.

Opencast mining is an activity that caters to many economic sectors; however, it has a large impact on society and the environment. After mining, the major concern is to restore the previous land cover, which was generally a natural vegetation cover. Establishing permanent vegetation cover can restore landscape connectivity and previous ecosystem functions, enhance aesthetic values and prevent off-side effects associated with post-mining landscapes. Opencast mining reclamation deals with these issues with several strategies that aim to develop a vegetation cover after mining activity has stopped. However, not all reclamation actions are effective, and assessing their efficiency by monitoring vegetation development at reclaimed sites is a time-consuming task because it usually involves extensive field work. In this study, we present a semi-automatic approach based on analysing satellite data (Landsat) time series and using a machine learning technique to identify suitable conditions for vegetation development at reclaimed opencast mines. We analysed the Teruel coalfield (Aragón, central-eastern Spain). This area is a representative Mediterranean-Continental region that is of particular interest due the diversity of reclamation actions that have been applied and the increase in drier conditions during the last decades. Conditions were described with topography derived variables, technical reclamation features and drought-occurrence variables as potential explanatory factors. The implemented approach allowed us to identify the main abiotic filters for vegetation of this geographic region: the water availability and soil retention (both controlled by the topographic slope), and the proximity to seed sources. The analysis evidenced the negative influence of drought occurrence on vegetation development, and different responses were found depending on the timescale at which drought is calculated. Our results indicate that the reclamation landform model is the main key factor influencing vegetation development. A model such as the smooth berm-slope increases water availability and controls soil erosion, and hence, improves vegetation development. In addition, we found that further than 500-600 m from the mine, the effect of seed source declines dramatically. Therefore, all these issues should be considered in future reclamation designs in a Mediterranean-Continental environment. Our methodology could be adapted to other geographic regions where spatial environmental data are available.

Seed-bank structure and plant-recruitment conditions regulate the dynamics of a grassland-shrubland Chihuahuan ecotone.

Large areas of desert grasslands in the southwestern United States have shifted to sparse shrublands dominated by drought-tolerant woody species over the last 150 yr, accompanied by accelerated soil erosion. An important step toward the understanding of patterns in species dominance and vegetation change at desert grassland-shrubland transitions is the study of environmental limitations imposed by the shrub-encroachment phenomenon on plant establishment. Here, we analyze the structure of soil seed banks, environmental limitations for seed germination (i.e., soil-water availability and temperature), and simulated seedling emergence and early establishment of dominant species (black grama, Bouteloua eriopoda, and creosotebush, Larrea tridentata) across a Chihuahuan grassland-shrubland ecotone (Sevilleta National Wildlife Refuge, New Mexico, USA). Average viable seed density in soils across the ecotone is generally low (200-400 seeds/m2 ), although is largely concentrated in densely vegetated areas (with peaks up to 800-1,200 seeds/m2 in vegetated patches). Species composition in the seed bank is strongly affected by shrub encroachment, with seed densities of grass species sharply decreasing in shrub-dominated sites. Environmental conditions for seed germination and seedling emergence are synchronized with the summer monsoon. Soil-moisture conditions for seedling establishment of B. eriopoda take place with a recurrence interval ranging between 5 and 8 yr for grassland and shrubland sites, respectively, and are favored by strong monsoonal precipitation. Limited L. tridentata seed dispersal and a narrow range of rainfall conditions for early seedling establishment (50-100 mm for five to six consecutive weeks) constrain shrub-recruitment pulses to localized and episodic decadal events (9-25 yr recurrence intervals) generally associated with late-summer rainfall. Re-establishment of B. eriopoda in areas now dominated by L. tridentata is strongly limited by the lack of seeds and decreased plant-available soil moisture for seedling establishment.

Assessing landscape structure and pattern fragmentation in semiarid ecosystems using patch-size distributions.

Spatial vegetation patterns are recognized as sources of valuable information that can be used to infer the state and functionality of semiarid ecosystems, particularly in the context of both climate and land use change. Recent studies have suggested that the patch-size distribution of vegetation in drylands can be described using power-law metrics, and that these scale-free distributions deviate from power-law linearity with characteristic scale lengths under the effects of increasing aridity or human disturbance, providing an early sign of desertification. These findings have been questioned by several modeling approaches, which have identified the presence of characteristic scale lengths on the patch-size distribution of semiarid periodic landscapes. We analyze the relationship between fragmentation of vegetation patterns and their patch-size distributions in semiarid landscapes showing different degree of periodicity (i.e., banding). Our assessment is based on the study of vegetation patterns derived from remote sensing in a series of semiarid Australian Mulga shrublands subjected to different disturbance levels. We use the patch-size probability density and cumulative probability distribution functions from both nondirectional and downslope analyses of the vegetation patterns. Our results indicate that the shape of the patch-size distribution of vegetation changes with the methodology of analysis applied and specific landscape traits, breaking the universal applicability of the power-law metrics. Characteristic scale lengths are detected in (quasi) periodic banded ecosystems when the methodology of analysis accounts for critical landscape anisotropies, using downslope transects in the direction of flow paths. In addition, a common signal of fragmentation is observed: the largest vegetation patches become increasingly less abundant under the effects of disturbance. This effect also explains deviations from power-law behavior in disturbed vegetation which originally showed scale-free patterns. Overall, our results emphasize the complexity of structure assessment in dryland ecosystems, while recognizing the usefulness of the patch-size distribution of vegetation for monitoring semiarid ecosystems, especially through the cumulative probability distributions, which showed high sensitivity to fragmentation of the vegetation patterns. We suggest that preserving large vegetation patches is a critical task for the maintenance of the ecosystem structure and functionality.