Terrain analysis, erosion simulations, and sediment fingerprinting: a case study assessing the erosion sensitivity of agricultural catchments in the border of the volcanic plateau of Southern Brazil

Abstract

Erosion and its spatial distribution in three agricultural headwater catchments were assessed in the border of the volcanic plateau in Southern Brazil. We analyzed terrain, hydrological processes and land use influence to provide a comprehensive assessment of the catchments' sensitivity to erosion. MethodsTopographic attributes were acquired from a digital elevation map, WaterSed model was parametrized to simulate runoff, diffuse erosion and sediment yield, and sediment source contributions were estimated using sediment fingerprinting based on near-infrared spectroscopy. Results 2According to the modeled results, areas covered by crop fields, grasslands and those adjacent to the drainage network are the most sensitive to erosion. Short distances from the source to the river network and the occurrence of high magnitude rainfall events (80 mm) promoted increases in connectivity for runoff/sediment transfer. Erosion simulations show that areas of low infiltration, as unpaved roads, were important runoff generators during lower volume rainfall events (25 mm). Sediment fingerprinting provided satisfactory results to quantify the contributions of unpaved roads to sediment (~39%).Topsoil and stream channels were also significant sediment sources for the set of analyzed samples, corresponding to average contributions of 38 and 23%, respectively. ConclusionAreas sharing geomorphological similarities did not lead to similar sediment contributions. Vegetation cover controlled erosion in topographically sensitive areas.Unpaved roads provide a significant sediment source, followed by topsoil and stream channels. The complementary results provide useful insights to better coordinate planning environmental conservation strategies in these fragile landscapes.

Combining sediment source tracing techniques with traditional monitoring: The “Arvorezinha catchment” experience

“…(2020), Valente et al (2020) and Dambroz et al (2022). The first two studies (Ramon et al, 2020;Valente et al, 2020) and Goiana (2857 km 2 ) catchments, respectively.…”

“…Subsequently, some of these techniques, which were initially tested and validated in the Arvorezinha catchment, were also applied in other river catchments in southern Brazil, for example, Ramon et al (2020), Valente et al (2020) and Dambroz et al (2022). The first two studies (Ramon et al, 2020; Valente et al, 2020) provided original approaches; they compared the results of spectroscopic tracers with those obtained with conventional methods (e.g., elemental concentrations, radionuclides).…”

“…Gradually, the catchments' sizes have increased, and studies have spread to other Brazilian states (Minas Gerais, Pernambuco and Distrito Federal). After 2014, some studies were also developed at an intermediate scale (catchment areas ranging from 10 1 to 10 2 km 2 ) (Batista et al, 2019; Dambroz et al, 2022; Miguel et al, 2014). In addition, between 2014 and 2020, sediment fingerprinting studies were carried out in catchments with a total area of between 800 and 1200 km 2 (Batista et al, 2019; Franz et al, 2014; Ramon et al, 2020; Tiecher et al, 2018).…”

“…In addition, the authors also successfully introduced magnetic properties to improve sediment source discrimination. More recently, Dambroz et al (2022) in southern Brazil and Amorim et al (2021) in a semi‐arid region of northeastern Brazil used spectroscopy alone to investigate the contribution of sediment sources in the Guarda Mor (18.5 km 2 ) and Goiana (2857 km 2 ) catchments, respectively. Therefore, the lessons and tests carried out with alternative tracers in the Arvorezinha catchment were fundamental to validating the use of these reflectance spectroscopy techniques as a reliable, fast and low‐cost method without generating chemical residues to be used for quantifying the contribution of sediment sources in catchments across Brazil.…”

“…In the same line of research, Dambroz et al (2022) used topographic attributes acquired from a digital elevation map and a spatially distributed erosion model (WaterSed) to simulate runoff, erosion, SY and sediment source contributions estimated using sediment fingerprinting based on near‐infrared spectroscopy. The authors reported that combining different environmental diagnosis techniques effectively outlined the fragility of those areas where overland flow may accumulate and lead to accelerated soil degradation processes.…”