Countries>Peru

MAAP #141: Protected Areas & Indigenous Territories Effective Against Deforestation in the Western Amazon

Posted on by dcadmin
Base Map. Primary forest loss across the western Amazon, with magnified visualization of the data. Click to enlarge. See Methodology for data sources.

As deforestation continues to threaten primary forest across the Amazon, key land use designations are one of the best hopes for the long-term conservation of critical remaining intact forests.

Here, we evaluate the impact of two of the most important land use designations: protected areas and indigenous territories.

Our study area focused on the four mega-diverse countries of the western Amazon (Bolivia, Colombia, Ecuador, & Peru), covering a vast area of over 229 million hectares (see Base Map).

We calculated primary forest loss over the past four years (2017-2020) across the western Amazon and analyzed the results across three major land use categories:

1) Protected Areas (national and state/department levels), which covered 43 million hectares as of 2020.

2) Indigenous Territories (official), which covered over 58 million hectares as of 2020.

3) Other (that is, all remaining areas outside protected areas and indigenous territories), which covered the remaining 127 million hectares as of 2020.

In addition, we took a deeper look at the Peruvian Amazon and also included long-term forestry lands.

In summary, we found that, averaged across all four years, protected areas had the lowest primary forest loss rate, closely followed by indigenous territories (see Figure 1). Outside of these critical areas, the primary forest loss rate was more than double.

Below, we describe the key results in greater detail, including a detailed look at each country.

 

Key Findings – Western Amazon

Figure 1. Primary forest loss rates in the western Amazon.

Overall, we documented the loss of over 2 million hectares of primary forests across the four countries of the western Amazon between 2017 and 2020. Of the four years, 2020 had the most forest loss (588,191 ha).

Of this total, 9% occurred in protected areas (179,000 ha) and 15% occurred in indigenous territories (320,000 ha), while the vast majority (76%) occurred outside key these land use designations (1.6 million ha).

To standardize these results for the varying area coverages, we calculated primary forest loss rates (loss/total area of each category). Figure 1 displays the combined results for these rates across all four countries.

From 2017-19, protected areas (green) had the lowest primary forest loss rates across the western Amazon (less than 0.10%).

Indigenous territories (brown) also had low primary forest loss rates from 2017-18 (less than 0.11%), but this rose in 2019 (0.18%) due to fires in Bolivia.

In the intense COVID pandemic year of 2020, this overall pattern flipped, with elevated primary forest loss in protected areas, again largely due to major fires in Bolivia. Thus, indigenous territories had the lowest primary forest loss rate followed by protected areas (0.15% and 0.19%, respectively) in 2020.

Averaged across all four years, protected areas had the lowest primary forest loss rate (0.11%), closely followed by indigenous territories (0.14%). Outside of these critical areas (red), the primary forest loss rate was more than double (0.30%). The lowest primary forest loss rates (less than 0.10%) occurred in the protected areas of Ecuador and Peru (0.01% and 0.03%, respectively), and indigenous territories of Colombia (0.07%).

Country Results

Figure 2. Primary forest loss rates in the Colombian Amazon.

Colombian Amazon

Colombia had, by far, the highest primary forest loss rates outside protected areas and indigenous territories (averaging 0.67% across all four years).

By contrast, Colombian indigenous territories had one of the lowest primary forest loss rates across the western Amazon (averaging 0.07% across all four years).

The primary forest loss rates for protected areas were on average nearly double that of indigenous territories (mostly due to the high deforestation in Tinigua National Park), but still much lower than non-protected areas.

 

 

 

 

 

Figure 3. Primary forest loss rates in the Ecuadorian Amazon.

Ecuadorian Amazon

Overall, Ecuador had the lowest primary forest loss rates across all three categories.

Protected areas had the lowest primary forest loss rate of any category across the western Amazon (averaging 0.01% across all four years).

Indigenous territories also had relatively low primary forest loss rates, averaging half that of outside protected areas and indigenous territories (0.10% vs 0.21%, respectively).

 

 

 

 

 

 

Figure 4. Primary forest loss rates in the Bolivian Amazon.

Bolivian Amazon

Bolivia had the most dynamic results, largely due to intense fire seasons in 2019 and 2020. Indigenous territories had the lowest primary forest loss rates, with 2019 being the only exception, due to large fires in the Santa Cruz department that affected the Monte Verde indigenous territory.

Protected areas had the lowest primary forest loss rate in 2019, but in extreme contrast, the highest the following year in 2020, also due to large fires in the Santa Cruz department that affected Noel Kempff Mercado National Park.

Overall, primary forest loss was highest outside protected areas and indigenous territories (averaging 0.33% across all four years).

 

 

 

Figure 5a. Primary forest loss rates in the Peruvian Amazon. Data: UMD.

Peruvian Amazon

Following Ecuador, Peru also had relatively low primary forest loss rates, particularly in protected areas (averaging 0.03% across all four years).

Primary forest loss in indigenous territories (that is, combined data for native communities and Territorial/Indigenous Reserves for groups in voluntary isolation) was surprisingly high, similar to that of areas outside protected areas across all four years. For example, in 2020, elevated primary forest loss was concentrated in several titled native communities in the regions of Amazonas, Ucayali, Huánuco, and Junín.

 

 

 

 

 

Figure 5b. Deforestation rates in the Peruvian Amazon. Data: MINAM/Geobosques.

As noted above, we conducted a deeper analysis for the Peruvian Amazon, using deforestation data produced by the Peruvian government and adding the additional category of long-term forestry lands (known as Permanent Production Forests, or BPP in Spanish) (see Annex map).

We also separated the data for indigenous territories into native communities and Territorial/Indigenous Reserves for groups in voluntary isolation, respectively.

These data also show that deforestation was lowest in the remote Territorial/Indigenous Reserves, closely followed by protected areas (0.01% vs 0.02% across all four years, respectively). Deforestation in titled native communities was 0.21% across all four years. Surprisingly, deforestation was higher in the forestry lands than areas outside protected areas and indigenous territories (0.30% vs 0.27% across all four years).

 

 

 

 

Annex – Peruvian Amazon

The following map shows added detail for Peru, most notably the inclusion of long-term forestry lands (known as Permanent Production Forests, or BPP in Spanish).

 

 

 

 

 

 

 

 

 

 

 

 

*Methodology

To estimate deforestation across all three categories, we used annual forest loss data (2017-20) from the University of Maryland (Global Land Analysis and Discovery GLAD laboratory) to have a consistent source across all four countries (Hansen et al 2013).

We obtained this data, which has a 30-meter spatial resolution, from the “Global Forest Change 2000–2020” data download page. It is also possible to visualize and interact with the data on the main Global Forest Change portal.

It is important to note that these data include both human-caused deforestation and forest loss caused by natural forces (landslides, wind storms, etc…).

We also filtered this data for only primary forest loss, following the established methodology of Global Forest Watch. Primary forest is generally defined as intact forest that has not been previously cleared (as opposed to previously cleared secondary forest, for example). We applied this filter by intersecting the forest cover loss data with the additional dataset “primary humid tropical forests” as of 2001 (Turubanova et al 2018). For more details on this part of the methodology, see the Technical Blog from Global Forest Watch (Goldman and Weisse 2019).

Thus, we often use the term “primary forest loss” to describe the data.

Data presented as primary forest loss or deforestation rate is standardized per the total area covered of each respective category. For example, to properly compare raw forest loss data in areas that are 100 hectares vs 1,000 hectares total size respectively, we divide by the area to standardize the result.

Our geographic range included four countries of the western Amazon and consists of a combination of the Amazon watershed limit (most notably in Bolivia) and Amazon biogeographic limit (most notably in Colombia) as defined by RAISG. See Base Map above for delineation of this hybrid Amazon limit, designed for maximum inclusion.

Additional data sources include: National and state/deprartment level protected areas: RUNAP 2020 (Colombia), SNAP 2017 & RAISG 2020 (Ecuador), SERNAP & ACEAA 2020 (Bolivia), and SERNANP 2020 (Peru).

Indigenous Territories: RAISG 2020 (Colombia, Ecuador, and Bolivia), and MINCU & ACCA 2020 (Peru). For Peru, this includes titled native communities and Indigenous/Territorial Reserves for indigenous groups in voluntary isolation.

For the additional analysis in Peru, we used deforestation data from MINAM/Geobosques (note this is actual deforestation and not primary forest loss) and BPP data from SERFOR. We also separated data from titled native communities and Territorial/Indigenous Reserves for groups in voluntary isolation.

Acknowledgements

We thank M. MacDowell (AAF) A. Folhadella (ACA), J. Beavers (ACA), S. Novoa (ACCA), and D. Larrea (ACEAA) for their helpful comments on this report.

This work was supported by the Andes Amazon Fund (AAF), Norwegian Agency for Development Cooperation (NORAD), and International Conservation Fund of Canada (ICFC).

 

Citation

Finer M, Mamani N, Silman M (2021) Protected Areas & Indigenous Territories Effective Against Deforestation in the Western Amazon. MAAP: 141.

MAAP #140: Detecting illegal gold mining in rivers with specialized satellites

Posted on by dcadmin
Image: Skysat (Planet). Analysis: MAAP/Amazon Conservation.

Illegal gold mining is a widespread problem in the southern Peruvian Amazon (Madre de Dios region), where it has caused river contamination and the deforestation of more than 100,000 hectares.

This activity has also emerged in the northern Peruvian Amazon (Loreto region), where it is not yet causing deforestation and the main threat is the contamination of rivers and important water resources.

This type of gold mining activity in water bodies (such as rivers) is illegal in Peru (see the “Legal situation” section below).

Identifying this type of mining (that is, in rivers and not causing deforestation) is difficult because the small mining boats (known as dragas) are mobile and imperceptible to medium and high-resolution satellites.

In this report, we test a novel technique based on tasking very high-resolution specialized satellites (in this case, Skysat with a spatial resolution of 0.5 meters) to detect illegal gold mining in the rivers of Loreto.

Below, we demonstrate how we used Skysat to detect illegal mining boats in the Nanay and Pintuyacu rivers, the main sources of drinking water for the city of Iquitos (capital of Loreto).

It is important to emphasize that this new technique has great potential for public institutions (national and regional levels) and local actors to detect and respond to illegal activity in real-time with appropriate measures (see the “Conclusion” section below).

Base Map: Mining in Nanay and Pintuyacu Rivers (Loreto region)

The Base Map below shows the precise points where illegal gold mining activity was found during 2020 and 2021 in the Nanay and Pintuyacu rivers of the Loreto region. For context, the map also includes the two nearby protected areas (one national, Allpahuayo Mishana, and another regional, Alto Nanay-Pintuyacu-Chambira). All the identified mining activity is illegal because there are no mining concessions in the area, in addition to the fact that it is occurring in bodies of water.

The yellow triangles indicate the illegal activity detected in 2020, based on field data or observations corroborated by specialists.

Based on this information, between March and May 2021, we tasked and analyzed very high-resolution satellite images (Skysat from the company Planet) for various strategic locations along both rivers. For images with possible mining activity, we consulted with field specialists for confirmation. The red triangles indicate the locations of illegal gold mining detected by Skysat and confirmed by the experts.

Base Map. Data: FEMA, MAAP, SERNANP.

Very High-resolution Satellite Images (Skysat)

Next, we show a series of striking images of illegal gold mining detected by Skysat and confirmed by experts. Note that with the very high resolution (0.5 meters), one can actually visualize the detail of a small mining boat. Image 1 shows several mining boats together in the Nanay River (near the town of Puca Urco). There are previous examples from the field of mining boats lining up together during their illegal activity (see the “Annex” section below).

Image 1. Mining boats in the Nanay River. Image: Skysat (Planet). Analysis: MAAP/Amazon Conservation.

Images 2-4 show other examples of likely mining boats in the Nanay River, this time within a national protected area (Alto Nanay-Pintuyacu-Chambira Regional Conservation Area). Note that these cases are also characterized by the presence of several boats lined up together.

Image 2. Mining boats in the Nanay River, in the Alto Nanay-Pintuyacu-Chambira Regional Conservation Area. Image: Skysat (Planet), Analysis: MAAP/Amazon Conservation.
Image 3. Mining boats in the Nanay River, in the Alto Nanay-Pintuyacu-Chambira Regional Conservation Area. Image: Skysat (Planet), Analysis: MAAP/Amazon Conservation.
Image 4: Mining boats in the Nanay River. Image: Skysat (Planet), Analysis: MAAP/Amazon Conservation.

Image 5 shows the presence of illegal mining boats in the Pintuyacu River.

Image 5: Mining boats in the Pintuyacu River. Image: Skysat (Planet), Analysis: MAAP/Amazon Conservation.

Conclusion

Unlike the dire situation in the southern Peruvian Amazon (Madre de Dios region), the illegal gold mining in northern Peru (Loreto region) does not cause deforestation and is caused by small mining boats in the rivers, making it practically invisible to medium and high-resolution satellites. This report presents a novel technique based on the strategic tasking of very high-resolution satellite images (Skysat) to detect this type of river-based illegal mining in real-time. With these images, we demonstrate the unprecedented capability to detect and visualize illegal activity in vast and remote areas, even down to the level of a small mining boat.

This new technique may allow public institutions and local actors to better respond to illegal activity in real-time with appropriate monitoring and control protocols. For example, key actors, such as the Peruvian Special Environmental Prosecutor’s Office (FEMA) can use this type of imagery in the planning and execution of their field interventions.

It is also important to highlight that the neighboring countries of Colombia and Bolivia experience the same problem of gold mining in rivers, so there is potential to replicate this model in other countries of the Amazon.

Annex

Here we show a photo from the field (Nanay River) of how the mining boats may line up during their illegal activity. This photo is for reference only and does not directly correspond to the cases described above.

Reference image of mining boats lined up during illegal activity. Source: ACRANPC.

Situación legal (in Spanish only)

El Decreto Legislativo N.° 1100 prohíbe, en el ámbito de la pequeña minería y minería artesanal, el uso de dragas y otros artefactos similares en todos los cursos de agua, ríos, lagos, lagunas, cochas, espejos de agua, humedales y aguajales. Por lo tanto, toda actividad enmarcada en este supuesto es considerada minería ilegal.

Mediante el Decreto Supremo N.° 150-2020-PCM se declara en emergencia varios distritos de Loreto por la inminente contaminación hídrica del río Nanay. A raíz de esto, se creó una comisión, cuyas actividades giraban en torno a varios operativos conjuntos, entre la Fiscalía Especializada en Materia Ambiental (FEMA), la Policía Nacional del Perú (PNP), la Dirección Regional de Energía y Minas (DREM) y la Autoridad Regional Ambiental (ARA), con el objetivo final de encontrar dragas en dicho río.

Mediante la Ordenanza Regional N.°  006-2003-GR, el Gobierno Regional de Loreto declaró la cuenca del río Nanay “zona de exclusión para actividades de extracción minera y para aquellas que alteren la cobertura vegetal.”

Acknowledgments

We thank Wendy Pineda from Rainforest US and Paul Lopez from the Satellite Monitoring Unit of the Loreto Environmental Specialized Prosecutor’s Office for their technical opinions regarding the confirmation of mining boats identified in the very high-resolution Skysat images.

We also thank Z. Romero (ACCA), G. Palacios (ACA), and G. Ribadeneyra, D. Torres, A. Felix, K. Nielsen, O. Liao and J. Carlos Guerra from USAID’s PREVENT Project, and J. Jara for their helpful comments on this report.

This report was conducted with technical assistance from USAID, via the Prevent project. Prevent is an initiative that is working with the Government of Peru, civil society, and the private sector to prevent and combat environmental crimes in Loreto, Ucayali and Madre de Dios, in order to conserve the Peruvian Amazon.

This publication is made possible with the support of the American people through USAID. Its content is the sole responsibility of the authors and does not necessarily reflect the views of USAID or the US government.

Citation

Finer M, Novoa S, Paz L, Saurez D, Mamani N (2021) Detecting illegal gold mining in rivers with specialized satellites. MAAP: 140.

MAAP #139: Using Satellites to Detect Illegal Logging in Peruvian Amazon

Posted on by dcadmin
Image 1. Illegal logging camp. Data: Skysat, MAAP/ACCA.

Illegal logging, in addition to larger-scale deforestation, is a major problem impacting the Peruvian Amazon.

In 2019, a Global Witness report, based on official information from the Peruvian government, estimated that at least 60% of the inspected timber over the past 10 years had an illegal origin. This problem not only directly affects the forest and its biodiversity, but also contributes to carbon loss (Qin et al, 2021) and forest degradation.

Illegal logging often involves the selective cutting of high-value trees in prohibited areas (whereas deforestation clears an entire area).

While numerous satellites can detect deforestation, only specialized satellites that are very high-resolution (less than one meter) can detect illegal logging.

In this report, we present a new emblematic case of illegal logging in the southern Peruvian Amazon.

It is based on a novel technique of tasking and analyzing very high-resolution images (in this case, with the Skysat satellite fleet from Planet) for a specific target area. Thanks to this new technique, we can tackle the problem of illegal logging in real-time, previously one of the biggest obstacles (see “Conclusion” section below).


Emblematic case

We refer to this as an emblematic case given the strong indicators of illegality (see Legal Status section, below) combined with likely significant impacts on an area of Amazon primary forest important for both indigenous peoples and biodiversity.

First, it is often difficult to confirm illegal logging given the frequent lack of updated technical and administrative information. This case study overcomes both obstacles.

Second, the illegal activity would not only be affecting a forestry concession (operated by the company Wood Tropical Forest), but also threatening important surrounding areas. Adjacent to the concession (to the west) is the Madre de Dios Territorial Reserve, a critical area that protects the territory of indigenous peoples in voluntary isolation. And to the south is the renowned Los Amigos Conservation Concession, a key area for the conservation of biodiversity.

Base Map

As part of our core work of continually monitoring the Los Amigos Conservation Concession, we acquired a series of very high-resolution images that also covered the surrounding Wood Tropical Forest forestry concession. These images, taken between February and April 2021, were obtained by the Skysat constellation (with a spatial resolution of 0.5 meters), operated by the satellite company Planet.

Our analysis revealed a serious situation of probable illegal logging: at least 3 active logging camps and 37 recently cut trees within the Wood Tropical Forest concession and close to both the neighboring Territorial Reserve and Conservation Concession (see Base Map).

Base Map. Data: MAAP/ACCA.

Very High-Resolution Skysat Images

The following images show some of the major findings made by our analysis of the Skysat data. Images 1-2 show examples of the logging camps, and Images 3-5 show examples of the likely selective illegal logging of high-value trees.

Image 2. Logging camp. Data: Skysat, MAAP/ACCA.
Image 3. Illegal logging. Data: Skysat, MAAP/ACCA.
Image 4. Illegal logging. Data: Skysat, MAAP/ACCA.
Image 5. Illegal logging. Data: Skysat, MAAP/ACCA.

Conclusion

This report presents a novel technique, based on the strategic capture of very high-resolution images (in this case, Skysat) and rapid analysis to detect selective illegal logging in real-time. Previously, one of the biggest obstacles to effectively addressing illegal logging was the inability of traditional monitoring methods to detect such small-scale, but widespread, illegal activity in the field. In this report, we demonstrate an important new capablility of detecting illegal logging activity in vast and remote areas in unprecedented detail, down to the level of a logging camp or individual cut trees.

Legal Situation (in Spanish)

La concesión forestal con Contrato N.° 17-TAM/C-J-007-02 fue otorgada en el año 2002 a la Empresa Shihuahuaco Timber S.A.C. y cedió su posición contractual a la empresa Wood Tropical Forestal en el año 2010, quien es titular del contrato de concesión hasta la actualidad.

La presunción de ilegalidad de la tala selectiva, evidenciada por nuestras imágenes satelitales, se debe a que la concesión forestal no se encontraría realizando actividades de aprovechamiento forestal enmarcadas en planes de manejo aprobados por el Gobierno Regional de Madre de Dios.

En efecto, tras realizar la consulta al Gobierno Regional de Madre de Dios, en su calidad de Autoridad Regional Forestal y de Fauna Silvestre (ARFFS), se advierte que la concesión se encuentra vigente. No obstante, no ha presentado a la ARFFS planes operativos para el aprovechamiento forestal desde hace más de ocho años. Incluso, desde el 30 de enero de 2020, cuenta con una resolución de la ARFFS que aprueba la suspensión del derecho de obligaciones contractuales (Resolución de Gerencia Regional N.° 065-2020-GOREMAD/GRFFS).

En ese sentido, y en tanto no se han presentado planes operativos en los últimos años, podemos inferir que en el área de la concesión forestal posiblemente no se estén realizando actividades lícitas de tala, por lo menos, desde hace ocho años.

Aunado a ello, en el Informe de Supervisión N.° 007-2019-OSINFOR/08.1.1, de acuerdo a una supervisión a la concesión para verificar obligaciones contractuales, el OSINFOR da cuenta de que la concesionaria presentó diversas denuncias entre los años 2016 al 2018 a la  Fiscalía Especializada en Materia Ambiental (FEMA), al Organismo de Supervisión de los Recursos Forestales y de Fauna Silvestre (OSINFOR) y al Gobierno Regional de Madre de Dios, en las cuales advirtió la presencia de terceras personas al interior de la concesión que estarían realizando tala ilegal, deforestación, instalación de campamentos ilegales, entre otros.

Acknowledgments

We thank E. Ortiz (AAF), Z. Romero (ACCA), G. Palacios (ACA), and A. Felix, J. Carlos Guerra, K. Nielsen, O. Liao, and R. Suarez from USAID’s PREVENT Project, and J. Jara for their helpful comments on this report.

This report was conducted with technical assistance from USAID, via the Prevent project. Prevent is an initiative that is working with the Government of Peru, civil society, and the private sector to prevent and combat environmental crimes in Loreto, Ucayali and Madre de Dios, in order to conserve the Peruvian Amazon.

This publication is made possible with the support of the American people through USAID. Its content is the sole responsibility of the authors and does not necessarily reflect the views of USAID or the US government.

This work was also supported by NORAD (Norwegian Agency for Development Cooperation), ICFC (International Conservation Fund of Canada), and EROL Foundation.

Citation

Finer M, Yupanqui O, Suarez D, Novoa S (2021) Using Satellites to Detect Illegal Logging in Peruvian Amazon. MAAP: 139.

MAAP #137: New Illegal Gold Mining Hotspot in Peruvian Amazon – Pariamanu

Posted on by dcadmin
Image 1. Very high-resolution image of the recent gold mining deforestation (10 hectares) in the new hotspot around the Pariamanu river. Data: Planet (Skysat)

In 2019, the Peruvian government launched Operation Mercury to confront the illegal gold mining crisis in the southern Amazonian area known as La Pampa (Madre de Dios region).

As a result, deforestation decreased 90% in this critical area (MAAP# 130).

Some illegal gold mining, however, has moved to several new hotspots (Image 1), although at much lower levels.

The most emblematic hotspot is located along the Pariamanu River, northeast of La Pampa in the Madre de Dios region (see Base Map, below).

We have documented the gold mining deforestation of 204 hectares (504 acres) in the Pariamanu area from 2017 to the present

This mining activity is clearly illegal because it is located within Brazil-nut forestry concessions, and is outside the permitted mining zone (commonly called the “mining corridor”).

Fortunately, a series of timely actions by the Peruvian Government has minimized the irreversible damage along the Pariamanu (see below).

The objective of this report is to present Pariamanu as an emblematic case that links technology with the rapid response action of public entities to address illegal activity in the Amazon.

It also represents a concrete case of strategic collaboration between civil society and the government to try and achieve zero illegal deforestation (and avoided deforestation).

Pariamanu

Base Map. Illegal gold mining deforestation along the Pariamanu river, in the context of La Pampa. Data: MAAP.

Base Map

The Base Map shows the location of illegal gold mining along the Pariamanu River, in the southern Peruvian Amazon (Madre de Dios region).

For context, La Pampa (the previous epicenter of illegal mining) and the regional capitol city of Puerto Maldonado are inlcuded. We also show another new illegal mining hotspot next to La Pampa, known as Apaylon.

In total, we have documented the deforestation of 204 hectares (504 acres) of primary forest caused by illegal gold mining in Pariamanu since 2017, indicated in red.

Note that this deforestation is located within Brazil nut forestry concessions and outside the “mining corridor,” thus clearly indicating its illegality.

Satellite Video: Illegal Gold Mining Deforestation in Pariamanu

We present a satellite image video showing an example of illegal gold mining in the Pariamanu area. These images show the deforestation of 71 hectares (175 acres) between 2016 (first image) and 2021 (last image), in the area indicated by the white inset box in the Base Map above. Note that each image is from July of each year (2016-20), with the exception of 2021 which shows January and March. Press the “play” button (lower left) to start the video. Click on the box (lower right) to view in full screen.

Satellite image video. Data: Planet.

Planet link: https://www.planet.com/stories/illegal-gold-mining-in-southern-peruvian-amazon-pa-6DfO4KuGg

MAAP Reports & Government Action

Operativo en Pariamanu, septiembre del 2020. Foto: FEMA Madre de Dios.

The first MAAP report about Pariamanu was published in November 2016, where we described “the start of mining in a new area” (MAAP #50). We found the mining-caused deforestation of 69 hectares (170 acres) on the banks of the Pariamanu river.

In January 2020, we published the second MAAP report about Pariamanu, documenting that the mining deforestation increased to 99 hectares (245 acres) (MAAP # 115). In this report, we warned that there were indications that some miners displaced by Operation Mercury (in February 2019) have moved to this area.

In response to this situation, the Peruvian Government, led by the Special Prosecutor for Environmental Matters (known as FEMA), carried out a series of field operations in 2020 (May, August and September, respectively), as an extension of Operation Mercury focused on cracking down on the illegal mining in Pariamanu.

The operations were effective in destroying mining equipment and sending a strong message that the government was engaged in this area.

However, we found that gold mining deforestation continued in several small areas between October 2020 and March 2021 (see Image 2), reaching the new total of 204 hectares (504 acres).

Fortunately, the government continues to respond effectively. Most recently (March 19, 2021), FEMA and the Peruvian Coast Guard carried out a new operation in Pariamanu, finding an illegal mining camp and equipment.

As mentioned above, the objective of this section (and this report) is to present Pariamanu as an emblematic case that links technology with the rapid response action of public entities to address illegal activity in the Amazon. It also represents a concrete case of strategic collaboration between civil society and the government to try and achieve zero illegal deforestation (and avoided deforestation).

Image 2. Data: Planet, MAAP.

Acknowledgments

We thank S. Novoa (ACCA), G. Palacios (ACA), and A. Felix, K. Nielsen, A. Caceres, I. Canelo, J. Carlos Guerra, O. Liao, y H. Che Piu from USAID’s PREVENT Project, for their helpful comments on this report.

This report was conducted with technical assistance from USAID, via the Prevent project. Prevent is an initiative that is working with the Government of Peru, civil society, and the private sector to prevent and combat environmental crimes in Loreto, Ucayali and Madre de Dios, in order to conserve the Peruvian Amazon.

This publication is made possible with the support of the American people through USAID. Its content is the sole responsibility of the authors and does not necessarily reflect the views of USAID or the US government.

This work was also supported by NORAD (Norwegian Agency for Development Cooperation), ICFC (International Conservation Fund of Canada), and EROL Foundation.

Citation

Finer M, Mamani N (2021) New Illegal Gold Mining Hotspot in Peruvian Amazon – Pariamanu. MAAP: 137.

MAAP #136: Amazon Deforestation 2020 (Final)

Posted on by dcadmin

Base Map. Forest loss hotspots across the Amazon in 2020. Data: Hansen/UMD/Google/USGS/NASA, RAISG, MAAP. The letters A-E correspond to the zoom examples below.

*To download the report, click “Print” instead of “Download PDF” at the top of the page.

In January, we presented the first look at 2020 Amazon deforestation based on early warning alert data (MAAP #132).

Here, we update this analysis based on the newly released, and more definitive, annual data.*

The Base Map illustrates the final results and indicates the major hotspots of primary forest loss across the Amazon in 2020.

We highlight several key findings:

  • The Amazon lost nearly 2.3 million hectares (5.6 million acres) of primary forest loss in 2020 across the nine countries it spans.
    g
  • This represents a 17% increase in Amazon primary forest loss from the previous year (2019), and the third-highest annual total on record since 2000 (see graph below).
    j
  • The countries with the highest 2020 Amazon primary forest loss are 1) Brazil, 2) Bolivia, 3) Peru, 4) Colombia, 5) Venezuela, and 6) Ecuador.
    h
  • 65% occurred in Brazil (which surpassed 1.5 million hectares lost), followed by 10% in Bolivia, 8% in Peru, and 6% in Colombia (remaining countries all under 2%).
    k
  • For Bolivia, Ecuador, and Peru, 2020 recorded historical high Amazon primary forest loss. For Colombia, it was the second highest on record.

In all of the data graphs, orange indicates the 2020 primary forest loss and red indicates all years with higher totals than 2020.

For example, the Amazon lost nearly 2.3 million hectares in 2020 (orange), the third highest on record behind only 2016 and 2017 (red).

Note that the three highest years (2016, 2017, and 2020) had one major thing in common: uncontrolled forest fires in the Brazilian Amazon.

See below for country-specific graphs, key findings, and satellite images for the top four 2020 Amazon deforestation countries (Brazil, Bolivia, Peru, and Colombia).

 

 

 

Brazilian Amazon

2020 had the sixth-highest primary forest loss on record (1.5 million hectares) and a 13% increase from 2019.

Many of the 2020 hotspots occurred in the Brazilian Amazon, where massive deforestation stretched across nearly the entire southern region.

A common phenomenon observed in the satellite imagery through August was that rainforest areas were first deforested and then later burned, causing major fires due to the abundant recently-cut biomass (Image A). This was also the pattern observed in the high-profile 2019 Amazon fire season. Much of the deforestation in these areas appears to associated with expanding cattle pasture areas.

In September 2020 (and unlike 2019), there was a shift to actual Amazon forest fires (Image B). See MAAP #129 for more information on the link between deforestation and fire in 2020.

Note that the three highest years (2016, 2017, and 2020) had one major thing in common: uncontrolled forest fires in the Brazilian Amazon.

Image A. Deforestation in Brazilian Amazon (Amazonas state) of 2,540 hectares between January (left panel) and November (right panel) 2020. Data: Planet.

Image B. Forest fire in Brazilian Amazon (Para state) that burned 9,000 hectares between March (left panel) and October (right panel) 2020. Data: Planet.

Bolivian Amazon

2020 had the highest primary forest loss on record in the Bolivian Amazon, surpassing 240,000 hectares.

Indeed, the most intense hotspots across the entire Amazon ocurred in southeast Bolivia, where fires raged through the drier Amazon forests (known as the Chiquitano and Chaco ecosystems).

Image C shows the burning of a massive area (over 260,000 hectares) in the Chiquitano dry forests (Santa Cruz department).

 

 

 

 

Image C. Forest fire in Bolivian Amazon (Santa Cruz) that burned over 260,000 hectares between April (left panel) and November (right panel) 2020. Data: ESA.

Peruvian Amazon

2020 also had the highest primary forest loss on record in the Peruvian Amazon, surpassing 190,000 hectares.

This deforestation is concentrated in the central region. On the positive, the illegal gold mining that plagued the southern region has decreased thanks to effective government action (see MAAP #130).

Image D shows expanding deforestation (over 110 hectares), and logging road construction (3.6 km), in an indigenous territory south of Sierra del Divisor National Park in the central Peruvian Amazon (Ucayali region). The deforestation appears to be associated with an expanding small-scale agriculture or cattle pasture frontier.

 

 

Image D. Deforestation and logging road construction in Peruvian Amazon (Ucayali region) between March (left panel) and November (right panel) 2020. Data: Planet.

Colombian Amazon

2020 had the second-highest primary forest loss on record in the Colombian Amazon, nearly 140,000 hectares.

As described in previous reports (see MAAP #120), there is an “arc of deforestation” concentrated in the northwest Colombian Amazon. This arc impacts numerous protected areas (including national parks) and Indigenous Reserves.

For example, Image E shows the recent deforestation of over 500 hectares in Chiribiquete National Park. Similar deforestation in that sector of the park appears to be conversion to cattle pasture.

 

 

 

Image E. Deforestation in Colombian Amazon of over 500 hectares in Chiribiqete National Park between January (left panel) and December (right panel) 2020. Data: ESA, Planet.

*Notes and Methodology

To download the report, click “Print” instead of “Download PDF” at the top of the page.

The analysis was based on 30-meter resolution annual data produced by the University of Maryland (Hansen et al 2013), obtained from the “Global Forest Change 2000–2020” data download page. It is also possible to visualize and interact with the data on the main Global Forest Change portal.

Importantly, this data detects and classifies burned areas as forest loss. Nearly all Amazon fires are human-caused. Also, this data does include some forest loss caused by natural forces (landslides, wind storms, etc…).

Note that when comparing 2020 to early years, there are several methodological differences from the University of Maryland introduced to data after 2011. For more details, see “User Notes for Version 1.8 Update.”

It is worth noting that we found the early warning (GLAD) alerts to be a good (and often conservative) indicator of the final annual data.

Our geographic range includes nine countries and consists of a combintion of the Amazon watershed limit (most notably in Bolivia) and Amazon biogeographic limit (most notably in Colombia) as defined by RAISG. See Base Map above for delineation of this hybrid Amazon limit, designed for maximum inclusion. Inclusion of the watershed limit in Bolivia is a recent change incorporated to better include impact to the Amazon dry forests of the Chaco.

We applied a filter to calculate only primary forest loss. For our estimate of primary forest loss, we intersected the forest cover loss data with the additional dataset “primary humid tropical forests” as of 2001 (Turubanova et al 2018). For more details on this part of the methodology, see the Technical Blog from Global Forest Watch (Goldman and Weisse 2019).

To identify the deforestation hotspots, we conducted a kernel density estimate. This type of analysis calculates the magnitude per unit area of a particular phenomenon, in this case forest cover loss. We conducted this analysis using the Kernel Density tool from Spatial Analyst Tool Box of ArcGIS. We used the following parameters:

Search Radius: 15000 layer units (meters)
Kernel Density Function: Quartic kernel function
Cell Size in the map: 200 x 200 meters (4 hectares)
Everything else was left to the default setting.

For the Base Map, we used the following concentration percentages: Medium: 7-10%; High: 11-20%; Very High: >20%.

 

Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyukavina, D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy, A. Egorov, L. Chini, C. O. Justice, and J. R. G. Townshend. 2013. “High-Resolution Global Maps of 21st-Century Forest Cover Change.” Science 342 (15 November): 850–53.

Acknowledgements

We thank E. Ortiz (AAF), M. Silman (WFU), M. Weisse (WRI/GFW) for their helpful comments on this report.

This work was supported by NORAD (Norwegian Agency for Development Cooperation) and ICFC (International Conservation Fund of Canada).

Citation

Finer M, Mamani N (2020) Amazon Deforestation Hotspots 2020 (Final). MAAP: 136.

MAAP #134: Agriculture and Deforestation in the Peruvian Amazon

Posted on by dcadmin

Peru’s first National Agricultural Area Map. Source: MIDAGRI.

For the first time, Peru has a detailed National Agricultural Area Map.

This unique map, produced with high-resolution satellite imagery, was published by the Peruvian Ministry of Agrarian Development (MIDAGRI) in January.*

This map reveals that the agricultural area at the national level is 11.6 million hectares, as of 2018.

Here, we analyze this new information in relation to annual forest loss data, generated by the Peruvian Environment Ministry (Geobosques).

The goal is to better understand the critical link between agriculture and deforestation in the Peruvian Amazon.

Specifically, we analyze the agricultural area of 2018 in relation to the preceding forest loss between 2001 and 2017.

Below are two main sections:

First, we present our Base Map that illustrates the major results.

Second, we show a series of zoomed images of select areas to illustrate key results in detail. These areas include major deforestation events related to oil palm, cacao, and other crops.

 

 

 

 

 

Base Map showing our major results. Data: MAAP, MIDAGRI, MINAM/Geobosques. Double click to enlarge.

Major Results

  • We found that 43% (4.9 million hectares) of Peru’s total agricultural area in 2018 was located in the Amazon basin.
    j
  • Of these Amazonian agricultural areas, more than 1.1 million hectares (24%) came from forest lost between 2001 and 2017 (indicated in red on the Base Map).
    k
  • Expressed another way, over half (56%) of the forest loss in the Peruvian Amazon between 2001 and 2017 corresponds to an agricultural area in 2018.
    l
  • The Base Map also shows, in brown, the agricultural area that is not linked to recent forest loss. The vast majority is located outside the Amazon basin (western Peru).
    l
  • Finally, the Base Map shows, in black, the recent forest loss not linked to agriculture. Much of this loss corresponds to gold mining (southeastern Peru), logging roads, and natural loss such as landslides.

 

 

 

 

 

 

Zooms of Key Areas

A. United Cacao (Loreto)

Image A shows the large-scale deforestation associated with the company United Cacao between 2013 and 2016, in the Loreto region  (MAAP # 128). The clearing, as the name indicates, was for the installation of Peru’s first and only industrial-style cacao plantation. In total, the deforestation for the plantation reached 2,380 hectares.

Zoom A. United Cacao (Loreto region). Data: MAAP, MIDAGRI, MINAM/Geobosques.

B. Oil Palm (Shanusi, Loreto)

Image B shows the large-scale deforestation of more than 16,800 hectares associated with oil palm plantations between 2006 and 2015, along the border of the Loreto and San Martin regions (MAAP #116). Of this total, the deforestation of 6,975 hectares was linked to two plantations managed by the company Grupo Palmas company. The remainder occurred in the private areas surrounding the company’s plantations.

Zoom B. Oil palm deforestation around Shanusi (Loreto region). Data: MAAP, MIDAGRI, MINAM/Geobosques.

C. Oil Palm (Ucayali)

Image C shows the large-scale deforestation of more than 12,000 hectares for two oil palm plantations between 2011 and 2015, in the Ucayali region (MAAP #41).

Zoom C. Oil palm deforestation (Ucayali region). Data: MAAP, MIDAGRI, MINAM/Geobosques.

D. Iberia (Madre de Dios)

Image D shows the expanding agriculture-related deforestation around the town of Iberia, near the border with Brazil and Bolivia (MAAP #75). The major cause, according to local sources, is the increase in corn, papaya, and cacao plantations. We have documented the deforestation of more than 3,000 hectares in this area since 2014.

Zoom D. Agriculture related deforestation around Iberia (Madre de Dios region). Data: MAAP, MIDAGRI, MINAM/Geobosques.

E. Zona Minera (Madre de Dios)

Finally, Image E shows deforestation in the gold mining hotspot known as La Pampa, in the Madre de Dios region. The non-agricultural deforestation in the center is the major illegal gold mining front. Around that area, and along the Interoceanic Highway, there is extensive agriculture-related deforestation.

Zoom E. Mining and agriculture deforestation in southern Peru (Madre de Dios region). Data: MAAP, MIDAGRI, MINAM/Geobosques.

*Notes and Methodology

According to MIDAGRI, the National Agricultural Area Map was “generated based on satellite images from RapidEye and later updated with satellite images from Sentinel-2 and the Google Earth platform, which allowed the mapping and precise measurement of the agricultural surface throughout the national territory.”

The data include “agricultural land with cultivation and without cultivation.” We assume that these data include cattle pasture.

The identification and quantification of deforested areas (2001-2017) that correspond to agricultural area in 2018 results from the analysis carried out in GIS by the superposition of both geospatial layers (MINAM and MIDAGRI).

Amazonian agricultural areas that came from forest lost between 2001 and 2017 = 1,185,722 hectares (indicated in red on the Base Map).

Acknowledgments

We thank E. Ortiz (AAF), S. Novoa (ACCA) and G. Palacios for their helpful comments on this report.

This work was supported by NORAD (Norwegian Agency for Development Cooperation), ICFC (International Conservation Fund of Canada), and EROL Foundation.

Citation

Vale Costa H, Finer M (2021) Agriculture and Deforestation in the Peruvian Amazon. MAAP: 134.

MAAP #132: Amazon Deforestation Hotspots 2020

Posted on by dcadmin

Base Map. Forest loss hotspots across the Amazon in 2020. Data: UMD/GLAD, RAISG, MAAP. The letters A-G correspond to the zoom examples below.

We present a first look at the major hotspots of primary forest loss across the Amazon in 2020 (see Base Map).*

There are several major headlines:

  • We estimate over 2 million hectares (5 million acres) of primary forest loss across the nine countries of the Amazon in 2020.*
    p
  • The countries with the highest 2020 primary forest loss are 1) Brazil, 2) Bolivia, 3) Peru, 4) Colombia, 5) Venezuela, and 6) Ecuador.
    p
  • The majority of the hotspots occurred in the Brazilian Amazon, where massive deforestation stretched across nearly the entire southern region. Many of these areas were cleared in the first half of the year and then burned in July and August. In September, there was a shift to actual forest fires (see MAAP #129).
    p
  • Several of the most intense hotspots were in the Bolivian Amazon, where fires raged through the dry forests (known as the Chiquitano) in the southeast region.
    p
  • There continues to be an arc of deforestation in the northwestern Colombian Amazon, impacting numerous protected areas.
    p
  • In the Peruvian Amazon, deforestation continues to impact the central region. On the positive, the illegal gold mining that plagued the southern region has decreased thanks to effective government action (see MAAP #130).

Below, we show a striking series of high-resolution satellite images that illustrate some of the major deforestation events across the Amazon in 2020 (indicated A-G on the Base Map).

Widespread Deforestation in the Brazilian Amazon

Zooms A-C show examples of a troublingly common phenomenon in the Brazilian Amazon: large-scale deforestation events in the first half of the year that are later burned in July and August, causing major fires due to the abundant recently-cut biomass. Much of the deforestation in these areas appears to associated with clearing rainforests for cattle pastures. The three examples below show the striking loss of over 21,000 hectares of primary forest in 2020.

Zoom A. Deforestation in the Brazilian Amazon (Amazonas state) of 3,400 hectares between April (left panel) and November (right panel) 2020. Data: ESA, Planet.

Zoom B. Deforestation in Brazilian Amazon (Amazonas state) of 2,540 hectares between January (left panel) and November (right panel) 2020. Data: Planet.

Zoom C. Deforestation in Brazilian Amazon (Para state) of 15,250 hectares between January (left panel) and October (right panel) 2020. Data: Planet.

Forest Fires in the Brazilian Amazon

In September, there was a shift to actual forest fires in the Brazilian Amazon (see MAAP #129). Zoom D and E show examples of these major forest fires, which burned over 50,000 hectares in the states of Pará and Mato Grosso. Note both fires impacted indigenous territories (Kayapo and Xingu, respectively).

Zoom D. Forest fire in Brazilian Amazon (Para state) that burned 9,000 hectares between March (left panel) and October (right panel) 2020. Data: Planet.

Zoom E. Forest fire in Brazilian Amazon (Mato Grosso state) that burned over 44,000 hectares between May (left panel) and October (right panel) 2020. Data: Planet.

Forest Fires in the Bolivian Amazon

The Bolivian Amazon also experienced another intense fire season in 2020. Zoom F shows the burning of a massive area (over 260,000 hectares) in the Chiquitano dry forests (Santa Cruz department).

Zoom F. Forest fire in Bolivian Amazon (Santa Cruz) that burned over 260,000 hectares between April (left panel) and November (right panel) 2020. Data: ESA.

Arc of Deforestation in the Colombian Amazon

As described in previous reports (see MAAP #120), there is an “arc of deforestation” concentrated in the northwest Colombian Amazon. This arc impacts numerous protected areas (including national parks) and Indigenous Reserves. For example, Zoom G shows the recent deforestation of over 500 hectares in Chiribiquete National Park. Similar deforestation in that sector of the park appears to be conversion to cattle pasture.

Zoom G. Deforestation in Colombian Amazon of over 500 hectares in Chiribiqete National Park between January (left panel) and December (right panel) 2020. Data: ESA, Planet.

Deforestation in the central Peruvian Amazon

Finally, Zoom H shows expanding deforestation (over 110 hectares), and logging road construction (3.6 km), in an indigenous territory south of Sierra del Divisor National Park in the central Peruvian Amazon (Ucayali region). The deforestation appears to be associated with an expanding small-scale agriculture or cattle pasture frontier.

Zoom H. Deforestation and logging road construction in Peruvian Amazon (Ucayali region) between March (left panel) and November (right panel) 2020. Data: Planet.

*Notes and Methodology

The analysis was based on early warning forest loss alerts known as GLAD alerts (30-meter resolution) produced by the University of Maryland and also presented by Global Forest Watch. It is critical to highlight that this data represents a preliminary estimate and more definitive data will come later in the year. For example, our estimate does include some forest loss caused by natural forces. Note that this data detects and classifies burned areas as forest loss. Our estimate includes both confirmed (1,355,671 million hectares) and unconfirmed (751,533 ha) alerts.

Our geographic range is the biogeographic boundary of the Amazon as defined by RAISG (see Base Map above). This range includes nine countries.

We applied a filter to calculate only primary forest loss. For our estimate of primary forest loss, we intersected the forest cover loss data with the additional dataset “primary humid tropical forests” as of 2001 (Turubanova et al 2018). For more details on this part of the methodology, see the Technical Blog from Global Forest Watch (Goldman and Weisse 2019).

To identify the deforestation hotspots, we conducted a kernel density estimate. This type of analysis calculates the magnitude per unit area of a particular phenomenon, in this case forest cover loss. We conducted this analysis using the Kernel Density tool from Spatial Analyst Tool Box of ArcGIS. We used the following parameters:

Search Radius: 15000 layer units (meters)
Kernel Density Function: Quartic kernel function
Cell Size in the map: 200 x 200 meters (4 hectares)
Everything else was left to the default setting.

For the Base Map, we used the following concentration percentages: Medium: 7-10%; High: 11-20%; Very High: >20%.

Acknowledgements

We thank E. Ortiz (AAF), M.E. Gutierrez (ACCA), and S. Novoa for their helpful comments on this report.

This work was supported by NORAD (Norwegian Agency for Development Cooperation) and ICFC (International Conservation Fund of Canada).

Citation

Finer M, Mamani N (2020) Amazon Deforestation Hotspots 2020. MAAP: 132.

MAAP #131: Power of Free High-resolution Satellite Imagery from Norway Agreement

Posted on by dcadmin

Image 1. Monthly Planet basemap for October 2020 across the Amazon, as seen on Global Forest Watch.

This report demonstrates the powerful application of freely available, high-resolution satellite imagery recently made possible thanks to an agreement between the Government of Norway and several satellite companies.*

This unprecedented agreement will bring commercial satellite technology, previously out of reach to many, to all working in tropical forest conservation around the world.

Here we show how MAAP (an initiative of Amazon Conservation) will use this information to enhance our real-time monitoring program and quickly share timely findings to partners in the field.

Specifically, we highlight the importance of the monthly basemaps (4.7-meter Planet imagery) available under the Norway agreement.* For example, Image 1 shows the stunning, nearly cloud-free October 2020 basemap across the Amazon.

l
Moreover, we show the power of this imagery visualized on Global Forest Watch, where it can be combined with early warning forest loss alerts.
p
Below, we highlight three examples where we combined this data to quickly detect and confirm deforestation in the Colombian, Ecuadorian, and Peruvian Amazon, respectively.

Colombian Amazon

First, we detected recent forest loss alerts (known as GLAD alerts), in the northwestern sector of Chiribiquete National Park. Image 2 is a screen shot of our monitoring search in Global Forest Watch (link here).

Second, we investigated the alerts with the freely available monthly Planet basemaps. Images 3-5 show the basemaps from October to December 2020. These images confirm that the area was covered in intact (likely primary) Amazon rainforest in October, and then experienced a major deforestation event (225 hectares) in November and December. Similar deforestation in the area appears to be conversion to cattle pasture. Note the crosshairs (+) represent the same point in all four images.

Image 2. Forest loss alerts in Chiribiquete National Park

Image 3. Monthly Planet basemap for October 2020 in Chiribiquete National Park.

Image 4. Monthly Planet basemap for November 2020 in Chiribiquete National Park.

Image 5. Monthly Planet basemap for December 2020 in Chiribiquete National Park.

Peruvian Amazon

Similarly, we detected recent forest loss alerts in an illegal gold mining area in the southern Peruvian Amazon known as Pariamanu (Image 6). Images 7 & 8 show the monthly basemaps confirming the expansion of illegal mining deforestation between October and December (see yellow arrows). Global Forest Watch link here.

Image 6. Forest loss alerts in illegal gold mining zone (Pariamanu).

Image 7. Monthly Planet basemap for October 2020 in Pariamanu.

Image 8. Monthly Planet basemap for October 2020 in Pariamanu.

Ecuadorian Amazon

Finally, we detected recent forest loss alerts of 100 hectares in an indigenous territory (Kichwa) surrounding an oil palm plantation in the Ecuadorian Amazon (Image 9). Images 10 & 11 show the monthly basemaps confirming large-scale deforestation between September and December, likely for the expansion of the plantation. Note the crosshairs (+) represents the same point in all three images. Global Forest Watch link here.

Image 9. Forest loss alerts in the Ecuadorian Amazon.

Image 10. Monthly Planet basemap for September 2020 in Ecuadorian Amazon.

Image 11. Monthly Planet basemap for December 2020 in Ecuadorian Amazon.

Summary

In summary, we show a major advance for free and real-time deforestation monitoring thanks to an agreement between the Government of Norway and satellite companies.* A key aspect of this agreement is making publically available (such as on Global Forest Watch) monthly basemaps created by the innovative satellite company Planet. Thus, users can now freely visualize recent forest loss alerts and then investigate them with high-resolution monthly basemaps on On Global Forest Watch. MAAP illustrated this process with three examples in the Colombian, Peruvian, Ecuadorian Amazon, respectively.

*Notes 

In September 2020, Norway’s Ministry of Climate and Environment entered into a contract with Kongsberg Satellite Services (KSAT) and its partners Planet and Airbus, to provide universal access to high-resolution satellite monitoring of the tropics in order to support efforts to stop the destruction of the world’s rainforests. This effort is led by Norway’s International Climate and Forest Initiative (NICFI). The basemaps are mosaics of the best cloud-free pixels each month. In addition to viewing the monthly basemaps on Global Forest Watch, users can sign up with Planet directly at this link: https://www.planet.com/nicfi/

Acknowledgements

We thank M. Cohen (ACA), M. Weisse (WRI/GFW), E. Ortiz (AAF) and G. Palacios for their helpful comments on this report.

This work was supported by NORAD (Norwegian Agency for Development Cooperation).

Citation

Finer M, Mamani N (2020) Power of Freely Available, High-resolution Satellite Imagery from Norway Agreement. MAAP: 131.

MAAP #130: Illegal Gold Mining Down 78% in Peruvian Amazon, But Still Threatens Key Areas

Posted on by dcadmin

Image 1. Very high resolution image of recent gold mining deforestation along the Pariamanu River. Data: Planet (Skysat).

As part of USAID’s Prevent Project (dedicated to combating environmental crimes in the Amazon), we conducted an updated analysis of illegal gold mining deforestation in the southern Peruvian Amazon.

In early 2019, the Peruvian government launched Operation Mercury, an unprecedented crackdown on the rampant illegal gold mining in the region.

The Operation initially targeted an area known as La Pampa, the epicenter of the illegal mining. In 2020, it expanded to surrounding critical areas.

In this report, we compare rates of gold mining deforestation before vs after Operation Mercury at six key sites (see Base Map and Methodology below).

We report four major results:

1) Gold mining deforestation decreased 90% in La Pampa (the most critical mining area) following Operation Mercury.

2) Gold mining deforestation increased in three key areas –Apaylon, Pariamanu, and Chaspa – indicating that some miners expelled from La Pampa moved to surrounding areas. The Peruvian government, however, has recently carried out major interventions in all three of these areas.

3) Overall, gold mining deforestation decreased 78% across all six sites following Operation Mercury.

4) Illegal mining does persist, however. We documented 1,115 hectares of gold mining deforestation across all six sites since Operation Mercury (but, compared to 6,490 hectares before the Operation).

Below, we provide a more detailed breakdown of the major results across all six sites. We also present a series of very high resolution satellite images (Skysat) of the recent gold mining deforestation.

Base Map – 6 Major Illegal Gold Mining Sites

The Base Map illustrates the results across the six major gold mining fronts in the southern Peruvian Amazon. Red indicates gold mining deforestation post Operation Mercury (March 2019 – October 2020), while yellow indicates the pre Operation baseline (January 2017 – February 2019).

Base Map. Major gold mining fronts in the southern Peruvian Amazon before (yellow) and after (red) Operation Mercury. Data: MAAP.

In La Pampa, we documented the dramatic loss of 4,450 hectares within the buffer zone of Tambopata National Reserve (Madre de Dios region) prior to Operation Mercury. Following the Operation, we confirmed the loss of 300 hectares. Note the main mining front in the core of the buffer zone has essentially been stopped, with most recent activity further north near the Interoceanic Highway.

In neighboring Alto Malinowski, located in the buffer zone of Bahuaja Sonene National Park (Madre de Dios region), we documented the loss of 1,558 hectares prior to Operation Mercury. Following the Operation, we confirmed the loss of 419 hectares.

In Camanti, located in the buffer zone of Amarakaeri Commuanl Reserve, we documented the loss of 336 hectares prior to Operation Mercury. Following the Operation, we confirmed the loss of 105 hectares.

In Pariamanu, located in the primary forests along the Pariamanu River (Madre de Dios region), we documented the loss of 72 hectares prior to Operation Mercury. Following the Operation, we confirmed the loss of 98 hectares. In response, the government conducted a major intervention in August 2020.

In Apaylon, located in the buffer zone Tambopata National Reserve (Madre de Dios region), we documented the loss of 73 hectares prior to Operation Mercury. Following the Operation, we confirmed the loss of 78 hectares. In response, the government has conducted a series of interventions in the area during 2020.

Chaspa, located in the buffer zone of Bahuaja Sonene National Park (Puno region), represents a unique case of a new gold mining front that appeared following Operation Mercury. Starting in September 2019, we documented the deforestation of 113 hectares impacting the Chaspa River watershed. In response, the government conducted a major intervention in October 2020.

Gold Mining Deforestation Trends

The following chart illustrates that gold mining deforestation fronts decreased following Operation Mercury in the three largest fronts (La Pampa, Alto Malinowski, and Camanti), and increased in three smaller areas (Pariamanu, Apaylon, and Chaspa). Thus, overall gold mining deforestation decreased 78% across all six major sites following Operation Mercury.

Table 1. Rates of gold mining deforestation before (orange) and after (red) Operation Mercury. Data: MAAP.

In La Pampa, the gold mining deforestation averaged 165 hectares per month prior to Operation Mercury. Following the Operation, the deforestation dropped to 17 hectares per month, an overall 90% decrease.

In Alto Malinowski, the gold mining deforestation dropped from 58 hectares per month to 23 hectares per month following Operation Mercury, an overall 60% decrease.

In Camanti, the gold mining deforestation dropped from 12.5 hectares per month to 6 hectares per month following Operation Mercury, an overall 54% decrease.

In Pariamanu, the gold mining deforestation increased from 2.8 hectares per month to 5 hectares per month following Operation Mercury, an overall 87% increase.

In Apaylon, the gold mining deforestation increased from 2.8 hectares per month to 4 hectares per month following Operation Mercury, an overall 43% increase.

Chaspa, located in the buffer zone of Bahuaja Sonene National Park, represents the unique case of a new gold mining front that appeared following Operation Mercury (8.5 hectares per month).

Very High Resolution Satellite Imagery (Skysat)

We recently tasked very high resolution satellite imagery (Skysat, 0.5 meter) for the major illegal gold mining areas. Below, we present a series showing some of the highlights from these images. Note that insets (in the upper corner of each image) show the same area before the mining activity (see red points as a reference).

Pariamanu

The following two images show the expansion of new gold mining areas into the primary rainforests near the Pariamanu River (Madre de Dios region).

Image 2. Expansion of new gold mining areas into the primary rainforests near the Pariamanu River (Madre de Dios region). Data: Planet.

Image 3. Expansion of new gold mining areas into the primary rainforests near the Pariamanu River (Madre de Dios region). Data: Planet.

La Pampa

The following image shows the expansion of a new gold mining area in the northern part of La Pampa.

Image 4. Expansion of a new mining area in the northern part of La Pampa (Madre de Dios region). Data: Planet, Maxar.

Chaspa

The following image shows the sudden appearance of a new gold mining front along the Chaspa River (Puno region).

Image 5. New gold mining front along the Chaspa River (Puno region). Data: Planet (Skysat).

Camanti

The following image shows the recent expansion of gold mining deforestation in the buffer zone of Amarakaeri Communal Reserve (Cusco region).

Image 6. Recent expansion of gold mining deforestation in the buffer zone of Amarakaeri Communal Reserve (Cusco region). Data: Planet (Skysat).

Methodology

We analyzed high-resolution imagery (3 meters) from the satellite company Planet obtained from their interface Planet Explorer. Based on this imagery, we digitized gold mining deforestation across six major sites: La Pampa, Alto Malinowski, Camanti, Pariamanu, Apaylon, and Chaspa. These were identified as the major active illegal gold mining deforestation fronts based on analysis of automated forest loss alerts generated by University of Maryland (GLAD alerts) and the Peruvian government (Geobosques) and additional land use layers. The area referred to as the “mining corridor” is not included in the analysis because the issue of legality is more complex.

Across these six sites, we identified, digitized, and analyzed all visible gold mining deforestation between January 2017 and the present (October 2020). We defined before Operation Mercury as data from January 2017 to February 2019, and after Operation Mercury as data from March 2019 to the present. Given that the former was 26 months and the latter 20 months, during the analysis the data was standardized as gold mining deforestation per month.

The data is updated through October 2020.

Acknowledgments

We thank A. Felix (DAI), S. Novoa (ACCA), and G. Palacios for their helpful comments on this report.

This report was conducted with technical assistance from USAID, via the Prevent project. Prevent is an initiative that is working with the Government of Peru, civil society, and the private sector to prevent and combat environmental crimes in Loreto, Ucayali and Madre de Dios, in order to conserve the Peruvian Amazon.

This publication is made possible with the support of the American people through USAID. Its content is the sole responsibility of the authors and does not necessarily reflect the views of USAID or the US government.

Citation

Finer M, Mamani N (2020) Illegal Gold Mining Down 79% in Peruvian Amazon, But Still Threatens Key Areas. MAAP: 130.

MAAP #129: Amazon Fires 2020 – Recap of Another Intense Fire Year

Posted on by dcadmin

Base Map. Major Amazon fires 2020 (orange dots) within Amazon watershed (blue line). Data: MAAP.

Following the intense Amazon fire season of 2019 that made international headlines, here we report another major fire year in 2020.

Using the novel data from our real-time Amazon Fires Monitoring app,* we documented over 2,500 major fires across the Amazon in 2020 (see Base Map).

The vast majority (88%) of the major fires were in the Brazilian Amazon, followed by the Bolivian Amazon (8%) and Peruvian Amazon (4%). No major fires were detected in the other Amazonian countries.*

We highlight several major headlines:

  • In the Brazilian Amazon, we detected 2,250 major fires. Most (51%) burned recently deforested areas, defined as fires in areas previously cleared between 2018 and 2020. These fires burned an estimated 1.8 million acres, emphasizing the current high deforestation rates in Brazil. In September, there was a major spike in forest fires, impacting vast areas of standing forest (over 5 million acres).
    m
  • In the Bolivian Amazon, we detected 205 major fires. The vast majority (88%) burned in Amazonian savanna and dry forest ecosystems. Notably, a quarter of these fires burned within protected areas.
    ,
  • In the Peruvian Amazon, we detected 116 major fires. There were three major types: 41% burned high elevation grasslands (impacting 26,000 acres), 39% burned recently deforested areas, and 17% burned standing forest (impacting 6,700 acres).
    v
  • The vast majority of the major fires across all three countries were likely human-caused and illegal, in violation of governmental fire management regulations and moratoriums.
    k
  • The app was only fully implemented in 2020, so we do not have comparable data for 2019. However, our extensive analysis of satellite imagery indicates that, in the Brazilian Amazon, both 2019 and 2020 had in common the extensive burning of recently deforested areas. The late season shift to forest fires seemed much more intense in 2020. In the Bolivian Amazon, both 2019 and 2020 had in common the extensive burning of Amazon savannas and dry forests.

See below for additional and more detailed findings for each country. Also, check out Mongabay’s real-time Brazilian Amazon fire tracker based on our analysis.

Brazilian Amazon

Image 1. Major fire burning recently deforested area in Brazilian Amazon (Mato Grosso). Data: Planet.

We emphasize the following additional findings for the Brazilian Amazon:

  • Of the 2,250 major fires, over half (51%) burned recently deforested areas, defined as areas where the forest was previously cleared between 2018 and 2020 prior to burning (Image 1). These fires burned an estimated 1.8 million acres (742,000 hectares), highlighting the current high deforestation rates in Brazil.
    .
  • A striking number (40%) were forest fires, defined here as human-caused fires in standing forest. A rough initial estimate suggests that 5.4 million acres (2.2 million hectares) of Amazon forest burned.
    .
  • Over half (51%) occurred in September, followed by August and October (25% and 15%, respectively). September was also when we documented a major shift from fires in recently deforested areas to forest fires.
    .
  • An important number of major fires (12%) occurred within indigenous territories and protected areas. The most impacted were Xingu and Kayapó Indigenous Territories, Jamanxim National Forest, and Nascentes da Serra do Cachimbo Biological Reserve.
    .
  • The vast majority of the major fires (97%) appear to be illegal, occurring after the Amazon fire moratoriums established in July (the government established a 4-month national fire moratorium starting July 15).
    ,
  • Pará  (38%) and Mato Grosso (31%) states had the most fires, followed by Amazonas (15%), Rondônia (11%), and Acre (4%).

Bolivian Amazon

Image 2. Major fire in Noel Kempff Mercado National Park, in the Bolivian Amazon. Data: Planet.

We emphasize the following additional findings for the Bolivian Amazon:

  • Of the 2015 major fires, many (46%) occurred in Amazon savannas.
    .
  • Another 42% of the fires were located in forests, mostly in the dry forests of the Chiquitano. Note, in November there was a major spike in these fires.
    .
  • Importantly, 25% of the major fires were in protected areas. The most impacted were Noel Kempff Mercado National Park (Image 2), Copaibo Municipal Protected Area, Iténez National Park, Keneth Lee Reserve, Rios Blanco y Negro Wildlife Reserve, and Pampas del Río Yacuma Integrated Management Natural Area.
    k
  • The vast majority of the fires (96%) were likely illegal, occuring after the fire moratoriums (August 3 in Beni and Santa Cruz, followed by October 5 nationally).
    .
  • Most of the fires occurred in the department of Beni (51%), followed by Santa Cruz (46%).
    .
  • August had the most fires (27%) followed closely by each of September, October, and November (24% each).
    h

Peruvian Amazon

Image 3. Major fire in higher elevation grassland of the Peruvian Amazon. Data: Planet.

We emphasize the following additional findings for the Peruvian Amazon:

  • Of the 116 major fires, many (39%) burned recently deforested areas. Although the pattern is similar to the Brazilian Amazon, the burned (and previously deforested) areas are much smaller (4,660 vs 1.8 million acres).
    ,
  • There were also numerous major fires (41%) in higher elevation grasslands across several regions (Image 3). These fires impacted 26,000 acres (10,000 hectares). We likely underestimated the number of these fires because, due to the lack of biomass in these ecosystems, they didn’t always register as a major fire in the app.
    k
  • Another 17% were forest fires, impacting 6,700 acres (2,700 hectares).
    k
  • All of the fires in the Peruvian were likely illegal, according to Peruvian fire management regulations.
    j
  • 15 regions experienced major fires, reflecting the mix of both grassland and forest fires. The regions with the most fires were Madre de Dios (23%), Ucayali (12%) and Junin (11%).
    h
  • November surprisingly had the most major fires (46%), followed by October and September (29% and 22%, respectively).
    j

*Notes and Methodology

The data is based on our analysis of Amazon Conservation’s novel real-time Amazon Fires Monitoring app. We started daily monitoring in May and continued through November. Specifically, he first major fire was detected on May 28 and the data was updated daily through November 30.

The app displays aerosol emissions as detected by the European Space Agency’s Sentinel-5 satellite. Elevated aerosol levels indicate the burning of large amounts of biomass, defined here as a “major fire”. In a novel approach, the app combines data from the atmosphere (aerosol emissions in smoke) and the ground (heat anomaly alerts) to effectively detect and visualize major Amazon fires.

When fires burn, they emit gases and aerosols. A new satellite (Sentinel-5P from the European Space Agency) detects these aerosol emissions (aerosol definition: Suspension of fine solid particles or liquid droplets in air or another gas). Thus, the major feature of the app is detecting elevated aerosol emissions which in turn indicate the burning of large amounts of biomass. For example, the app distinguishes small fires clearing old fields (and burning little biomass) from larger fires burning recently deforested areas or standing forest (and burning lots of biomass). The spatial resolution of the aerosol data is 7.5 sq km. The high values in the aerosol indices (AI) may also be due to other reasons such as emissions of volcanic ash or desert dust so it is important to cross reference elevated emissions with heat data and optical imagery.

We define “major fire” as one showing elevated aerosol emission levels on the app, thus indicating the burning of elevated levels of biomass. This typically translates to an aerosol index of >1 (or cyan-green to red on the app). To identify the exact source of the elevated emissions, we reduce the intensity of aerosol data in order to see the underlying terrestrial heat-based fire alerts. Typically for major fires, there is a large cluster of alerts. The major fires are then confirmed, and burn areas estimated, using high-resolution satellite imagery from Planet Explorer.

Some additional country-specific notes:

Bolivia – As note above, the high values in the aerosol indices (AI) may also be due to other reasons such as emissions of volcanic ash or desert dust. Hence, some areas, such as the Salar de Uyuni, in western Bolivia, often have orange or red tones.

Colombia – Our daily 2020 monitoring took place from May until November, but Colombia’s drier burning season was likely earlier in the year (January – March). We will be monitoring Colombia during this time frame in 2021.

Acknowledgements

The app was developed and updated daily by Conservación Amazónica (ACCA). The data analysis is led by Amazon Conservation in collaboration with SERVIR Amazonia.

We thank E. Ortiz, A. Folhadella, A. Felix, and G. Palacios for their helpful comments on this report.

Citation

Finer M, Villa L, Vale H, Ariñez A, Nicolau A, Walker K (2020) Amazon Fires 2020 – Recap of Another Intense Fire Year. MAAP: 129.