Hans Nicholas Jong, Jakarta – Researchers have called out the Indonesian government's claim of having reduced deforestation by 90% over the past decade, pointing to cherry-picked data and a skewed baseline that paint an incomplete picture of the reality on the ground.
In a recently published report, Norwegian and Indonesian researchers at the Norwegian University of Life Sciences (NMBU) compared the official data with other satellite-based deforestation data sets from independent sources, such as Global Forest Watch (GFW). They found that while all the independent data sets also showed a decline, these were lower than the 90% figure that the Indonesian government gives for the period 2015-2021. GFW data, for instance, showed a decline of 69% from the period 2016-2023.
Ahmad Dermawan, a postdoctoral researcher at NMBU and co-author of the report, said a major issue was the Indonesian government's choice of 2015 as the baseline year, i.e., the year against which deforestation rates for subsequent years would be compared.
2015 marked an extreme case of El Nino, which in Indonesia results in hotter and drier conditions than usual. This, combined with burning by plantation companies to clear land, led to some of the worst forest and land fires on record across Indonesia. An estimated 2.6 million hectares (6.4 million acres) – an area four times the size of the island of Bali – burned that year, of which more than 1 million hectares (2.5 million acres) were forest loss.
That constituted one of the highest deforestation rates in the country's history, according to official data. By using 2015 as a reference point, deforestation rates in subsequent years, which were nowhere near as bad, appear as dramatic improvements, Ahmad said.
"Since there is no official baseline year, selecting 2015 as the benchmark results in the largest possible decline in deforestation," Ahmad told Mongabay.
If the government had instead set 2010 as its baseline, for instance, the decline in deforestation would look far less significant, he added.
Strategic decision
Choosing 2015 as the base year was likely a strategic decision, with the government also doing so in its negotiations with Norway for their climate partnership, Ahmad said. That partnership, signed in 2010 and renewed in 2022, commits Norway to paying Indonesia $1 billion if it succeeds in reducing greenhouse gas emissions from deforestation and forest degradation, a mechanism known as REDD+.
By using 2015 as the starting point, Indonesia can maximize its claimed emissions reductions, which strengthens its case for receiving payments. Since 2022, Norway has paid out $216 million under the partnership.
Yet that baseline also is "not necessarily incorrect," Ahmad said, given that it was the first full year of then-president Joko Widodo's first term in office. "So it might be the case that 2015 was used because it marks the beginning of the Jokowi administration," he said, referring to the former president by his popular nickname.
However, Ahmad said it's important to note that forest protection policies the government credited for the deforestation decline – such as a timber legality system, as well as a moratorium on clearing new areas of forests and peatlands – predate the Jokowi administration.
Beyond the baseline selection, another key issue underlying the government's claimed deforestation decline is the perennial one of how it defines what a forest is.
The official Indonesian definition of a forest includes commercial tree plantations such as acacia (grown and felled to make pulp and paper) and rubber. Most independent forest monitoring platforms like GFW, however, don't include tree plantations in their forest loss statistics, instead distinguishing between natural forests and plantations.
What this means is that a pulpwood company can clear a forest, which the government records as deforestation, and then plant the land it with acacia, which then counts as reforestation – i.e., there's virtually no net deforestation recorded, even though a swath of natural forest has been cleared.
"This creates a wide range of figures, opening for cherry picking and selective uses," Ahmad and his colleagues wrote in their report.
They say this underscores the need for full transparency on definitions and data to enable better comparisons and detect underlying trends.
Better comparison
The researchers suggest reducing bias in deforestation trend analysis by adopting a multiyear moving average approach. This calculates average deforestation over multiple consecutive years and then compares it to the average of another multiyear period.
For instance, in 2014 and 2016, Indonesia's official deforestation rate was 400,000 and 630,000 hectares (990,000 and 1.56 million acres). The year in between, 2015, was an outlier, at 1 million hectares.
By taking the average of multiple years, extreme fluctuations like this can be neutralized, Ahmad said.
And to get even better analysis of deforestation trends, it's useful to compare the periods before and after the change, as it yields a more representative figure, he said.
Taking the average for the years 2010-2017 (largely before the disastrous fires of 2015) and 2020-2022 (the period after), the drop in the deforestation rate comes out to 52% using the official data.
Taken together, the researchers conclude that "deforestation has been reduced by at least 50% since ca. 2016," rather than the 90% drop touted by the government.
Drivers of declining deforestation
Regardless of the scale of the decline, the downward trend is something all data sets agree on, the researchers say. Why there's been a decline, however, is harder to say. "There is no straightforward method to answer such a complex question," and "no single, best approach to trace impacts," the researchers wrote in their report.
But since the reduction has occurred across most provinces and direct drivers of deforestation, like palm oil and pulpwood, the researchers suggest that national policies and structural changes might have been a major factor, more than commodity-specific ones like commodity prices and certification.
The researchers suggest the decline being seen now is the result of policies and measures enacted much earlier, given that public policy reforms take time to have an impact on the ground. For instance, the moratorium on clearing new areas of primary forest and peatlands was enacted in 2011 by then-president Susilo Bambang Yudhoyono as part of the REDD+ partnership agreement with Norway. And while its impact is still being debated, several studies indicate it has had a positive conservation effect. Others, however, make the case that the moratorium's impact may be overstated because much of the area it was meant to protect was already under existing conservation status.
Nevertheless, many of the experts interviewed by Ahmad and his colleagues for their report agreed the moratorium has contributed to forest protection by establishing clear boundaries between areas that may and may not be developed. Another important effect of the moratorium has been improved coordination across ministries and levels of government, the researchers noted.
Forest scarcity effect
The researchers also highlighted another likely major reason for the drop in deforestation: that there's simply much less forest left to clear. They estimated that up to a third of the reduction in the deforestation rate can be explained by this so-called forest scarcity effect.
Using a model, the researchers were able to estimate the turning point when deforestation rates in each province start declining. They found that once a province's forest cover dwindles to about 40% of its total land area, the deforestation rate starts to slow down. This was particularly evident in heavily deforested provinces on the island of Sumatra. The corollary, however, is that provinces where forest cover remains high, such as in the eastern region of Papua, will remain subject to potentially high deforestation rates for some time to come, the researchers warned.
They also looked at the role of commodity prices, corporate policies and industry initiatives as factors for the overall decline in deforestation. Commodity prices, they concluded, didn't have a significant influence. But there were some indications that corporate actions, like the adoption of NDPE (no deforestation, no peat, no exploitation) policies, may have produced some reduction in deforestation. Still, there's limited academic research quantifying NDPE's effect, Ahmad noted.
Furthermore, companies with NDPE policies still face challenges in ensuring that the smallholders and indirect suppliers they source from comply with their NDPE policies, the researchers wrote. As such, forest scarcity is the more likely factor, they concluded.
Future deforestation trend
Despite the declining trend over the past decade, there's been a sharp reversal in the last couple of years, the researchers said. GFW data show a 21% increase in deforestation from 2022 to 2023. Data from another independent forest monitoring platform, Nusantara Atlas, suggest an increase in deforestation driven by palm oil (up 36%) and pulpwood plantations (15%), with mining also likely to continue its rise.
Bella Nathania, deputy program director at the Indonesian Center for Environmental Law (ICEL), who was not involved in the analysis, said deforestation is likely to increase under the current administration, given President Prabowo Subianto's target of 8% economic growth.
This will put enormous pressure on forests, she said during a recent online event.
The forest scarcity model suggests present and future increases will be concentrated in parts of eastern Indonesia that still have high forest cover, such as Papua and Sulawesi, the researchers said.
"We need to watch Papua as it is one of the last major tropical forest frontiers," Ahmad said. "It still has high forest cover, and forest loss [there] is still relatively low. But this could change rapidly if deforestation drivers shift eastward."
In Papua, the biggest deforestation threat comes from the government's food estate program. The project aims to clear 3 million hectares (7.4 million acres) of land in the district of Merauke – two-thirds of it for rice fields and the rest for sugarcane plantations – an area 45 times the size of Jakarta.
President Prabowo has categorized the food estate program as one of strategic national importance, having campaigned in the 2024 election on achieving food and energy self-sufficiency.
The researchers noted that such programs are exempt from the forest- and peat-clearing moratorium, and thus pose a threat to the country's remaining forests.
Other major threats to keep an eye on are pulpwood plantations and mining, as both have grown in prominence as drivers of deforestation in recent years, Ahmad said. The researchers identified expanding mining activities for coal and nickel in Borneo and Sulawesi islands respectively.
While mining's overall share in deforestation remains low for now, the researchers said the activity often has large indirect effects through infrastructure development and drawing in workers and their families. These effects don't immediately show up in post-forest land-use figures, they wrote.
"The indirect impact of mining on deforestation may be larger than the direct impact," Ahmad said. "If mining-related deforestation covers 1 million hectares, the secondary effects could be several times greater due to road-building, settlements, and indirect land-use changes."
As coal will likely remain as Indonesia's primary energy source, despite discussions about transitioning to renewables, mining is expected to continue to expand, he warned.
Increased demand for renewable energy and minerals needed to produce electric vehicle batteries, such as nickel, also poses a threat to forests, the researchers wrote. Indonesia is the world's top nickel producer, supplying close to half (48%) of global production in 2023, and the expansion of the nickel industry, heavily pushed by the government, has come at the expense of the country's rainforests. A recent study found that the kind of nickel mining carried out in Indonesia has a land-use footprint about 20 times larger than previously thought.
Role of civil society
As deforestation threats continue to loom, civil society groups play a critical role as watchdogs, especially in an era of contradictory policies, Ahmad said.
For instance, the Ministry of Forestry is pushing for forest protection and reforestation, while the Ministry of Energy and Mineral Resources wants to boost coal production, which could drive further deforestation in the future, Ahmad said.
"This contradiction makes NGO oversight even more important," he said.
He added that environmental issues tend to gain public attention only after civil society groups expose them. These groups' role in forest monitoring is also important as they use independent platforms like MapBiomas and Nusantara Atlas, which offer alternative perspectives on land-use dynamics than official data sources, he said. Nusantara Atlas, for instance, can track commodity-driven deforestation (e.g., from palm oil, pulpwood or mining), a level of detail missing from the government's data set.
"For a more detailed picture of deforestation, we must also examine the commodities driving forest loss," Ahmad said.
Bella of ICEL agreed, saying it's crucial to have a better and unified dataset, supported by improved transparency and coordination to strengthen monitoring and law enforcement.
"Because according to official data we have, deforestation is decreasing. But we're worried that on the ground, there's a lot of legal deforestation happening for energy and large-scale food projects," she said. "So this needs to be reflected in the data – whether [the] deforestation is legal or illegal."
As Indonesia enters a new political era marked by intensified resource extraction and mounting forest threats in the east, having a clear and honest picture of what's happening to its forests is more critical than ever, Bella said.
Citation
Heijlen, W., & Duhayon, C. (2024). An empirical estimate of the land footprint of nickel from laterite mining in Indonesia. The Extractive Industries and Society, 17, 101421. doi:10.1016/j.exis.2024.101421