Date Published: February 9, 2018
Publisher: Public Library of Science
Author(s): Sandra Isay Saad, Jonathan Mota da Silva, Marx Leandro Naves Silva, João Luis Bittencourt Guimarães, Wilson Cabral Sousa Júnior, Ricardo de Oliveira Figueiredo, Humberto Ribeiro da Rocha, Julia A. Jones.
The choice of areas for nature conservation involves the attempt to maximize the benefits, whether by carrying out an economic activity or by the provision of Ecosystem Services. Studies are needed to improve the understanding of the effect of the extent and position along the watershed of restored areas on soil and water conservation. This study aimed to understand how different restoration strategies might reflect in soil conservation and sediment retention. Using InVEST tool, sediment transport was simulated in a small 12 km2 watershed (Posses River, in Southeast Brazil), where one of first Brazilian Payment for Ecosystem Services (PES) projects is being carried out, comparing different hypothetical restoration strategies. With 25% of restoration, sediment export decreased by 78% for riparian restoration, and 27% for the steepest slopes restoration. On the other hand, the decrease in soil loss was lower for riparian restoration, with a 16% decrease, while the steepest slopes restoration reduced it by 21%. This mismatch between the reduction of sediment export and soil loss was explained by the fact that forest not only reduces soil loss locally but also traps sediment arriving from the upper parts of the watershed. While the first mechanism is important to provide soil stability, decreasing the risk of landslip, and to maintain agricultural productivity, the second can improve water quality and decrease the risk of silting, with positive effects on the water reservoirs at the outlet of the watershed. This suggests that Riparian and the Steepest Slopes restoration strategies are complementary in the sense of preventing sediments from reaching the water bodies as well as protecting them at their origin (with the reduction of erosion), so it will be advisable to consider the two types of restoration.
While the conservation of natural resources in not valued in traditional economics , there is a growing perception of the strong human dependence on Ecosystem Services (ES) . Attempts to avoid ecosystem degradation face great challenges: on one hand, most of policies to avoid deforestation have been inefficient , and, on the other, ecosystem restoration is costly, and funds are generally limited . Around the world, billions of dollars have been spent on ecosystem restoration programs which have not always been successful [5,6]. In fact, there is a need to choose priority zones for nature conservation that consider both environmental and social-economic issues as conservation priorities are needed for planning and decision making . When it comes to achieving the economic viability of Payment for Ecosystem Services (PES) projects, a key factor is to choose potential areas for large production of ES  as targeted restoration can be much more efficient than random reforestation . The cost of restoration can be used as a criterion for prioritization , and many times passive restoration can be employed to reduce costs due to the relatively good cost-effectiveness .
Calibration was performed in terms of sediment export, but not of soil loss, due to the lack of soil loss monitoring in the Posses watershed so far, which could be achieved by plot studies (which is beyond the scope of this research). To work around this limitation, we found not one but several possibilities of soil loss for current land use, due to different calibration combinations. And this uncertainty is also reflected in the other land use scenarios evaluated. For the current land-use scenario, our soil loss estimate was between 12.8 and 22.3 Mg ha-1y-1 with a sediment export rate of 1.4 Mg ha-1y-1. The difference in soil loss between the land use scenarios was small in comparison to the estimated uncertainties in each scenario. This was not the case of sediment export, for which the prediction for each restoration strategy was very distinct from the others. However, there might be other uncertainties that were not considered in this study, like the “observed” sediment export, which in an ideal situation would require more measures especially under higher discharge events, as well as direct measures of sediment export, and not only turbidity. Studies using model comparisons would also be advisable for a better estimation of uncertainties as we estimated only the uncertainties related to parameter choice and input data using the InVEST model.
Under different restoration strategies, we evaluated changes in two components of sediment transport across the Posses watershed: soil loss and sediment export. The simulated soil loss in each part of the watershed is routed downslope and downstream, until reaching the water bodies (which characterizes sediment export). Part of the sediment is trapped by vegetation before it reaches streams and reservoirs. We suggested that the decrease in sediment export is controlled not only by the decrease in soil loss, but mostly by the efficiency with which each land use type traps sediments. This is clear when comparing two opposite restoration strategies: one which prioritizes riparian zones, and one that prioritizes the steepest slopes. The Steepest Slopes restoration was the strategy that promoted the greater decrease in soil loss, due to the higher soil loss on the steepest slopes, and Riparian restoration promoted the greater decrease in sediment export, due to the proximity of the vegetation buffer to the rivers, attenuating sediment transported even from the upper parts of the watershed. In the scenarios with an area of 25% of forest (the area achieved with the PES project) soil loss decreased from 18.5 Mg ha-1y-1 in the Anthropized Scenario (no forest) to 14.6 (21% decrease) and to 15.6 Mg ha-1y-1 (16% decrease) for restoration of steepest area and riparian, respectively. Sediment export decreased from 3.4 Mg ha-1y-1 in the Anthropized Scenario to 2.5 (27% decrease) and 0.7 Mg ha-1y-1 (78% decrease) for Restoration of the Steepest Area and Riparian Restoration, respectively. Despite the apparent superiority of riparian restoration in comparison to the steepest slope restoration in actions to protect river water quality and avoid silting, we highlight the fact that forests can have a limited capacity of trapping, due to the possibility of trapping saturation over time during rainfall events. Thus, the steepest slopes, with higher erosion rates, should also be treated as priorities to avoid the arrival of sediments in rivers.