Solve pollution. Save lives. Protect the planet.



Research brief written by

Dr. Francisco Román, Restaura Amazonía (RAMAZ), Puerto Maldonado, Madre de Dios, Perú; Centro de Innovación Científica Amazónica (CINCIA), Puerto Maldonado, Madre de Dios, Perú.

France Cabanillas, Pure Earth – Blacksmith Institute, New York, USA; Centro de Innovación Científica Amazónica (CINCIA), Puerto Maldonado, Madre de Dios, Perú; Restaura Amazonía (RAMAZ), Puerto Maldonado, Madre de Dios, Perú.

Edwing Arapa, Independent Consultant in Hydrology, Lima, Perú.

Charles Espinosa, Pure Earth – Blacksmith Institute, New York, USA.


Since 2017, Pure Earth has collaborated with CINCIA, a local research center that develops ways to restore degraded land in the Peruvian Amazon. Using CINCIA’s cutting-edge methodology, Pure Earth has been working hand-in-hand with formalized gold miners to reforest their degraded mining concessions and fulfill mine closure requirements.

Artisanal and Small-Scale Gold Mining (ASGM) is responsible for the largest fraction of forest loss and disturbance in the Amazon region of Madre de Dios, Peru . ASGM is unique compared to other drivers of deforestation due to the severity of its environmental impacts as well as the health and social impacts caused by mercury contamination. Primary and secondary forests near gold mining areas receive extremely high emissions of mercury and experience elevated levels of total mercury and methylmercury in the atmosphere, canopy foliage, and soils.

Previous studies on abandoned mining sites in the tropics, specifically in Madre de Dios, suggest that forest recovery after ASGM is slower and qualitatively inferior compared to natural regeneration following agricultural land uses. Compared with other land uses, ASGM leads to the lowest residual forest and soil carbon and the highest loss of ecosystem services due to fine sediment removal, defaunation, poor water quality, and mercury contamination in soil, water, and air. Compared to normal soils in the region, substrates after mining can present extreme challenges for plant colonization and the formation of self-sustaining ecosystems.

In recent years, reforestation projects have been implemented in areas degraded by gold mining in Madre de Dios; these projects have helped to generate new knowledge about the survival, growth, and fertilization requirements of a wide variety of species, as well as the complementary potential of natural regeneration.

This knowledge provides the potential for designing interventions that more effectively rehabilitate and restore areas degraded by ASGM. However, the estimation and quantification of the potential benefits of forest restoration, in terms of sediment and mercury retention, have been little studied. This document presents the results of a study estimating the potential effects of different reforestation and deforestation scenarios on the retention of sediments and mercury by vegetation cover in a mining area in Madre de Dios.

Effects of reforestation on streamflow and sediment retention

The results suggest that there is no visible variation in streamflow due to changes in coverage in the different projected scenarios. However, in the case of sediment production, we observe a sediment reduction effect for the Restoration scenario and an opposite effect of sediment increase for the Degradation scenario. Likewise, there is a marked seasonality in the resuls given that the main effects are seen during the months with the highest rainfall.

In the Restoration scenario, where the 539 ha of deforested land are reforested, sediment production may be reduced by 0.64 t/ha/year (5.7% less sediment with respect to the Baseline scenario), which represents an annual sediment retention of 345 t/year. In contrast, in the Degradation scenario, where the 571 ha of remaining forest are deforested, sediment production may increase by 0.84 t/ha/year (7.6% more than the Baseline scenario), which would increase the annual sediment production to 480 t/year.

Seasonal variation of the streamflow and sediment production in the AMATAF concessions located in the Malinowski River Basin for the different scenarios projected.


Based on the average sediment retention values projected in this study and on previous reports on the concentration of mercury in sediments in the Malinowski River Basin, we estimate that restored vegetation cover could retain up to 15.36 mg/ha/year of mercury. Thus, if the 539 currently degraded ha of the AMATAF concessions are reforested, up to 8.3 kg/year of mercury could be prevented from reaching the river. On the other hand, in the Degradation scenario, loss of vegetation cover could emit up to 20.16 mg/ha/year of mercury. In other words, if the 571 ha of forest remaining in the concessions are deforested, up to 11.5 kg of mercury could be displaced into the river annually.


Hydrological Modeling

For this study, we applied the Soil and Water Assessment Tool (SWAT; htps://, a precipitation runof hydrological model, to analyze the efects of land cover change on water supply and sediment transpor in the AMATAF concessions located in the Malinowski River Basin. To run the SWAT model, the following information was collected: a) slope or digital elevation model (ALOS-PALSAR satellite, res. 12.5 m); b) vegetation cover or ecosystems (MINAM, 2019); c) soils (FAO); d) precipitation (PISCO-SENAMHI 1981-2016); e) baseflow at the Amaru Mayu station (SENAMHI).


Location of study area (orange boxes) in the Malinowski River Basin (boundaries in yellow).

Research conclusions
  • The results of this case study show the potential role of vegetation in capturing sediments and mercury, estimating that restored vegetation cover could help retain up to 15.36 mg/ha/year of mercury, preventing its flux into the Malinowski River Basin.
  • Another benefit of reducing sedimentation and mercury content in rivers through the restoration and conservation of riparian vegetation in ASGM areas is the improvement of water quality, beneficial both for productive use as well as the improvement of food security and public healh through the consumption of mercury-free fish.
  • The hydrological modeling and estimates of mercury content in sediments performed in this study are unique to the Malinowski River Basin. Previous studies show important variations in the amount of sediment and mercury found in different rivers of the Madre de Dios region due to differences in geomorphology and intensity of nearby mining activity. Therefore, the extrapolation of these results to other river basins in the region is not recommended.
  • The estimates made in this study should be used with caution since they are a result of projections based on secondary information which will need to be confirmed and adjusted over time through field evaluations and measurements with research projects and ecohydrological monitoring.
The Toxic Truth

Sediment and Mercury Retention Traps:
The potential role of reforestation in mining areas of Madre de Dios