- Years of satellite imagery and analysis reveals that United Cacao, a company once publicly traded on the London Stock Exchange, deforested nearly 2,000 hectares (about 5,000 acres) of primary forest in the Peruvian Amazon.
- The evidence refutes the company’s narrative that farmers had degraded the land before it arrived.
- The deforestation, as well as other legal violations, have led to sanctions against a successor to United Cacao’s Peruvian subsidiary, now called Tamshi SAC.
- But Tamshi is now claiming that Mongabay Latam improperly used the term “deforestation” and has sued for defamation.
The satellite’s eye view from high above northern Peru in mid-2013 laid out a stark reality: Ecologist Matt Finer and his colleagues were watching “blow by blow” in a series of images from NASA’s Landsat satellite as a once-unbroken blanket of green Amazonian rainforest was giving way to neat strips of bare land.
The systematic dismantling of nearly 2,000 hectares (5,000 acres) of decades-old, closed-canopy rainforest was at first thought to be a frontier foray into Peru’s growing oil palm sector at the time. But it soon became clear that the big push was altering the landscape for a huge estate in the remote Amazon by a company — or rather, a chain of companies controlled by self-described “serial plantations entrepreneur” Dennis Melka.
Neither the companies nor Melka were trying to hide the results of their work, at least from the people they hoped would invest in the operation through United Cacao, a publicly traded umbrella company based in the Cayman Islands. In fact, they were celebrating their progress outside the Amazonian hamlet of Tamshiyacu. At times in 2013 and 2014, crews were clearing as much as 100 hectares (250 acres) of forest a week in a bid to make United Cacao “the largest and lowest cost corporate grower of sustainable and traceable cacao beans,” according to a quote from Melka on the company’s website as it existed in 2015.
To drum up early interest ahead of United Cacao’s initial public offering on the London Stock Exchange’s AIM market, Melka as CEO and the company’s directors repeatedly made the case to investors and through media appearances that the developing estate in Peru would be the answer to the projected shortfalls in the raw material used to make chocolate. It would also, they said, address consumer and industry concerns about the sustainability, human rights violations and unpredictable weather in West Africa, which supplies around 70% of the world’s cacao.
But could a company that had wiped out hectare upon hectare of the Amazon forest be considered “sustainable”? No, Finer told Mongabay.
“This is the definition of ‘not sustainable,’” said Finer, the director of the Monitoring of the Andean Amazon Project (MAAP), a program of the organization Amazon Conservation. “From my point of view, it is not acceptable — and certainly not sustainable — to cut down standing forest, let alone primary forest, for large-scale agriculture.”
In its financial disclosures, on its website and in Melka’s public statements, the position of the company was that nothing resembling forest, let alone primary forest, existed by the time the company purchased the land, mostly from title-holding farmers. United Cacao insisted that the farmers had degraded the area since the 1990s. As a result, the company couldn’t have deforested the area because few if any trees remained to be cut, so the thinking went.
“There was no high-conservation-value forest on that land,” Ed Portman, a spokesman for United Cacao with the London-based public relations firm Tavistock, told Mongabay in late 2014.
Their argument implied that only scrubby, secondary forest remained, and that the land was worth more as a plantation that would help slake the world’s chocolate obsession than it was as a slice of the planet’s largest bank of tropical rainforest, providing habitat for wildlife and storage for carbon.
The trouble with that argument is that it directly contradicted the ongoing scientific analysis of the satellite imagery. Over much of the next decade, public relations and legal battles would involve conflicting interpretations of Peru’s evolving land use and forest protection laws. Questions stemming from the company’s practices, including the initial clearance of forest, would ultimately lead to the resignation of United Cacao’s CEO and several other senior executives. Soon after, the London Stock Exchange’s AIM market delisted United Cacao.
Later, in 2019, managers of United Cacao subsidiary Cacao del Peru Norte (CDPN), which is now known as Tamshi SAC, were sentenced to prison time. Tamshi’s general manager got eight years for trafficking illegal timber. Part of the evidence leading to the convictions came from the satellite imagery, like the high-resolution satellite photographs from September 2013 had revealed enormous trees lying prostrate at the plantation site like spilled matchsticks.
The public prosecutor of the Ministry of the Environment, Julio Guzmán Mendoza, said that his team used satellite photographs in the forensic reconstruction of what happened, calling them “the history in pictures of our forest.”
“The images proved that there was a forest in that place and that this forest was removed without authorization,” Guzmán said. He also noted that analyses of the images confirmed the prior presence of “primary forest” in the area that had been cleared.
The court in charge of the case calculated that CDPN had caused more than 156 million soles (about $41.7 million) in damages to the ecosystem and the services it provides. In addition to the sentences, the company was ordered to pay more than 15.7 million soles ($4.2 million) in reparations.
Alberto Yusen Caraza, a prosecutor specializing in environmental issues in the province of Maynas where Tamshiyacu is located, told Mongabay Latam at the time that he expected the effects of the decision to ripple far beyond this single decision.
“For us it is an emblematic case because of the magnitude of deforestation and the amount of 15 million soles of civil reparation,” Caraza said.
The fight continues to this day, though, centering on the question of what constitutes “deforestation,” legally and substantively. It has come to embroil not just the environmental organizations, who contend that the company broke several Peruvian laws, but also a reporter from Mongabay’s Latin America bureau based in Lima, Peru.
But also to this day, the satellite images remain an inescapable record of what United Cacao and CDPN did to the forest outside the town of Tamshiyacu.
The rapid changes that unfolded on the plantation site between May 2013 and August 2014 tracked so meticulously by Finer and his colleagues show clearly that the area was largely undisturbed for decades — in Guzmán’s words, primary forest — right up through 2012. CDPN turned up in mid-2013, and all of a sudden, cookie-cutter swatches of forest begin to disappear. But those early images would merely be the beginning, the first few tendrils of an avalanche of scientific evidence directly contradicting United Cacao’s narrative of how the land looked before its arrival. They left few doubts that the swift disappearance of trees near Tamshiyacu had indeed been deforestation.
In the United Cacao version of the story, the company acquired land from peasant farmers, many of whom had been resettled in this part of Peru after fleeing the terrorism that plagued the country in the 1980s and 1990s. In this interpretation, the company was repurposing land already in use for agriculture, leaving intact the foundations of United Cacao’s claim to being a sustainable cacao producer. In a similar vein, Dennis Melka and company documents maintained that the area had also been “heavily logged of all tropical hardwoods” before the company’s arrival.
Melka obliquely addressed the concerns raised by environmental groups about deforestation for large-scale agriculture in a 2015 interview posted on DirectorsTalk, an investment website. (The post no longer appears on the site.) “I think it’s important to note that by the time the plantation companies get to the land, that land has been logged or clear cut of all tropical hardwoods,” he told the host. “It’s not rainforest. It’s secondary forest. It’s a highly degraded area.”
Here, Melka conflates “clear-cutting,” a recognizable term that entails the wholesale destruction of an area of forest, with selective “logging” for the most economically valuable species of hardwood trees. That kind of selective logging has occurred across the Amazon for decades, if not centuries, but it does not imply the clearance of swaths of forest.
“Much of the tropical world has been selectively logged,” tropical ecologist William Laurance told Mongabay in 2015 in response to Melka’s commentary. “The notion that this is just degraded, valueless land is simply a laughable proposition.”
By definition, selective logging leaves other trees standing. Scientists agree that selective logging, especially on an industrial scale, certainly impacts ecosystems. But these forests often remain vitally important providers of wildlife habitat, carbon sequestration and myriad other ecosystem services, such as water regulation, that robust and intact rainforests provide.
Until he left United Cacao, Melka repeatedly made the point that the forest was worthless for anything other than an agriculture plantation. But this assessment didn’t line up with even the early analyses led by Finer.
After looking at the images immediately before and after the development of the plantation, the team took a deeper dive into the Landsat historical record, going back to 1985. At no point does anything more than a fleck of forest change register over the course of the time series within the plantation boundaries — that is, until 2013, when United Cacao, through CDPN, began cutting down trees.
Such a record suggests that the previous landowners hadn’t done much farming on the land. By Finer’s calculations, the area that United Cacao and CDPN cleared had been 98% covered by closed-canopy rainforest, which is a common definition ecologists use for primary — not secondary, as Melka suggested — forest. The team concluded from the dense canopy’s persistence that the area had hosted neither significant agriculture nor had it been subjected to any recent heavy selective logging.
Finer and his team didn’t stop there, though. They did what’s called an NDVI analysis, short for normalized difference vegetation index. This tool compares the “greenness” signatures in different locations to get an idea of how much biomass they contain. Lowland rainforests, for example, typically hold a lot of biomass; thus, their corresponding NDVI values are quite different from those that come from the sparser greenery found in scrubland or deserts.
Finer used a protected area adjacent to the plantation site as a benchmark for comparison because it had NDVI values typical of primary forest. The NDVI values for the forest that had existed on the United Cacao plantation site were quite similar to those from the protected area.
In addition to the Landsat images and the vegetation analysis, Finer’s team also looked at the carbon held by the trees cleared for the cacao plantation. Ecologist Greg Asner, then at the Carnegie Institution for Science at Stanford University, and scientists from Peru’s Ministry of the Environment had recently developed nuanced maps of the country’s carbon stocks from airborne lidar data. “Lidar” is short for “light detection and ranging” and uses measurements from laser pulses to develop a three-dimensional picture of a forest’s structure.
Asner and his colleagues ground-truthed that data and then in 2014 scaled it up to publish a hectare-by-hectare map of the country’s carbon stocks. Not surprisingly, the Amazon had the greatest stockpiles of carbon, which appear in deep red on the countrywide map. In particular, Asner and his colleagues note that the region of Loreto, home to Tamshiyacu and the cacao plantation, has “very high carbon densities.” The researchers calculated that the forest that existed before it was cleared for the plantation had an average of 122 metric tons of carbon per hectare. Loreto, therefore, offered an important opportunity to keep copious amounts of carbon locked away in the trees by preserving its forests, they wrote.
“These carbon stock values are found only in intact tropical forest in the region, and are among the highest values mapped within Peru during the joint Carnegie-Peruvian Ministry of Environment carbon mapping project,” Asner, now the director of Arizona State University’s Center for Global Discovery and Conservation Science, told Mongabay in 2015. “The logical conclusion from the scientific data is that large-statured, intact forest was removed by this deforestation event.
“That suggests to me that you’ve got a lot of really big trees there,” Asner said, an assertion corroborated by the increasingly sharp Landsat photographs. “If you’ve got a lot of big trees, you’ve got something approaching mature-phase forest.”
New lines of evidence
Several years later, Finer and the MAAP team delved deeper into the carbon data, zooming in on the location using the Asner team’s high-resolution maps. They found that clearing the forest for United Cacao’s plantation led to the release of 291,000 metric tons of carbon. More than 100 metric tons of carbon was frittered away from nearly every hectare of forest cut down for the plantation. Even clearing the few spots of less-densely forested or degraded areas at the plantation site led to the loss of a still-high 50 to 100 metric tons of carbon per hectare. By comparison, most research suggests that a typical oil palm plantation maxes out at storing around 40 metric tons in each hectare.
“This is another line of evidence that the company cleared primary forest, contrary to their claims that the area was already degraded,” the MAAP team wrote.
The team continues to counter United Cacao’s narrative that its activities didn’t constitute “deforestation.” They’ve tracked the continued expansion of the plantation, including during periods when progress was supposed to have been paused so that authorities could investigate what was going on. They noted in 2016 that the Peruvian Ministry of Agriculture and Irrigation classified nearly all — more than 92% — of the land cleared for the plantation as optimal for forest production, not agriculture, based on climate, soil and topography studies of the area. And new images of ever-increasing resolution reinforce the initial case that the company dispatched long-standing, mature rainforest to plant cacao.
The current debate on what qualifies legally as “deforestation” has touched off a lawsuit by the directors of Tamshi SAC. They have accused a Mongabay Latam reporter of defamation, saying that the term was improperly used with respect to the company’s activities and that the company has suffered negative consequences as a result.
But the defamation lawsuit is at odds with yet another recent decision by the Peruvian government. In late 2020, Peru’s Environmental Assessment and Enforcement Agency (OEFA) required that Tamshi cease its work and leave the region of Loreto because it didn’t have the proper environmental permits. The OEFA also noted that Tamshi had illegally gotten rid of hazardous waste in the Amazon. The agency imposed a fine of 129 million soles (nearly $35 million) on the company.
In another twist, a judge tossed out the sentences handed down to several former CDPN managers in late 2020, much to the disappointment of prosecutors like Guzmán.
“This sentence produces frustration. We expected an exemplary sentence,” he told Mongabay. “This is a serious blow to the fight against deforestation.
Legal wrangling aside, the scientists familiar with the case harbor no doubts about what happened near Tamshiyacu, namely, that the destruction of closed-canopy rainforest that stood virtually undisturbed for decades is the definition of deforestation. The satellite data and subsequent analyses leave little room for debate, and the case has proven that companies can no longer hide deforestation — as it is defined by science — in areas far from official scrutiny.
Clinton Jenkins, a conservation ecologist now at Florida International University who collaborated early on with Finer’s team, pointed out in 2013 that companies deforesting the Amazon “should be on notice.”
“We can see everything they do from space,” Jenkins said, “and so they will be exposed when they lie.”
A time-lapse animation reveals the sudden deforestation that occurred near the town of Tamshiyacu beginning in 2013. Image courtesy of Google Timelapse.
Banner image: Global Forest Watch tree cover loss 2001-2021 set on Planet imagery from Jun-Sep 2020.
John Cannon is a staff features writer with Mongabay. Find him on Twitter: @johnccannon
Asner, G.P., Knapp, D. E., Martin, R. E., Tupayachi, R., Anderson, C. B., Mascaro, J., … Neyra, A.F. The High-Resolution Carbon Geography of Perú. Minuteman Press.
Gorelick, N., Hancher, M., Dixon, M., Ilyushchenko, S., Thau, D., & Moore, R. (2017). Google Earth Engine: Planetary-scale geospatial analysis for everyone. Remote Sensing of Environment.
FEEDBACK: Use this form to send a message to the author of this post. If you want to post a public comment, you can do that at the bottom of the page.