The research, led by Dhia Zalfa Zahira, Anjar Dimara Sakti, and Ketut Wikantika of Institut Teknologi Bandung, goes beyond simply counting lost trees. It tracks how much carbon those forests actually stored — both above ground in trunks, branches, and leaves, and below ground in root systems — and how that storage changed year by year. Most previous studies of the region offered static snapshots for a single year or focused only on aboveground biomass. By modelling both aboveground and belowground biomass carbon across five time points, the team has produced one of the most dynamic pictures yet of a tropical carbon reservoir in decline.

Defining what counts as an intact forest was itself a technical feat. The researchers started with the Hansen tree cover dataset derived from Landsat imagery, flagging any pixel with at least 30 percent canopy cover in 2000. They then filtered out oil-palm plantations using a global planting-year dataset and removed cropland using the MODIS land cover product. From that baseline, they iteratively subtracted three drivers of loss — stand-replacing deforestation, new cropland, and new oil-palm planting — at five-year intervals, generating intact forest maps for 2000, 2005, 2010, 2015, and 2020. This multi-layer filtering matters because oil-palm plantations, though tree-covered, are not ecologically intact forests; they typically replace natural forest and store far less carbon.

With the forest maps in hand, the team turned to a Random Forest regression model, an ensemble machine learning method that aggregates predictions from hundreds of decision trees. Thirteen environmental predictors fed the model, spanning vegetation indices such as NDVI and NIRv, productivity metrics including gross and net primary productivity, leaf area index, and the fraction of absorbed photosynthetically active radiation, along with climate variables like land surface temperature and precipitation, and topographic features including elevation, slope, and a topographic wetness index. All layers were processed at 300-metre resolution and aggregated to 1.5-by-1.5-kilometre grids to reduce noise and pixel misalignment.

A key design choice set this model apart: rather than training only on forest pixels, the researchers trained across the full gradient of Southeast Asian vegetation — forests, shrublands, grasslands, and cropland alike — using more than two million grid cells from 2010. This ensures the model learned the entire spectrum of carbon density, reducing bias at forest edges and recently disturbed areas. The data was split 80-20 into training and testing sets, with a Kolmogorov-Smirnov test confirming that both subsets covered the region geographically without spatial bias. Against the NASA reference biomass dataset, the model achieved a testing R² of 0.899 for aboveground carbon and 0.874 for belowground carbon, indicating strong predictive skill with limited overfitting.

Independent validation against the European Space Agency’s Climate Change Initiative Biomass product — built from radar observations by Sentinel-1, Envisat, and the ALOS satellites — showed moderate but consistent agreement, with R² values of 0.645 and 0.632 for the two carbon components. Among the thirteen predictors, the fraction of absorbed photosynthetically active radiation emerged as the single most influential variable, which makes physical sense: how much solar energy a canopy captures directly governs how much carbon it can fix into biomass. Land surface temperature and precipitation followed, while the classic vegetation indices contributed surprisingly little, apparently limited in their sensitivity to structural biomass in dense tropical canopies.

The carbon accounting that followed is sobering. In 2000, intact forests in the region held an estimated 26.18 megatonnes of aboveground biomass carbon and 6.61 megatonnes belowground, a combined total of roughly 32.8 megatonnes. By 2020, those figures had fallen to 15.96 and 3.96 megatonnes respectively — a drop of more than 39 percent in twenty years. The steepest losses occurred between 2010 and 2015, when forest area shrank by nearly 384,000 square kilometres and biomass carbon declined at a rate of about 1,496 megatonnes per year in total. The hardest-hit landscapes were Kalimantan and Sumatra in Indonesia, Sarawak and Sabah in Malaysia, and cropland-driven loss zones in northern Thailand, Myanmar, and parts of Papua.

To translate those ecological losses into terms policymakers cannot ignore, the team ran an indicative economic valuation under two carbon pricing scenarios. At the current average price across ASEAN+3 countries — a modest 6.45 US dollars per tonne of CO₂ equivalent — the region’s intact forest carbon was worth about 776 billion dollars in 2000, falling to roughly 471 billion by 2020. Under the globally recommended price of 75 dollars per tonne, aligned with social cost of carbon estimates, the value dropped from about 9.02 trillion dollars to 5.48 trillion. The authors are careful to stress these are illustrative, non-market figures, not credit-eligible carbon trading values, but they make the scale of the liability unmistakable: a loss of nearly 4 trillion dollars in indicative carbon assets in two decades.

The findings land at a delicate moment for climate policy. Southeast Asia harbours nearly 15 percent of the world’s tropical forests, and their continued erosion threatens not only regional biodiversity but global mitigation commitments under the Paris Agreement, REDD+ frameworks, and Sustainable Development Goal targets on sustainable forest management. There is also a real risk of a tipping dynamic: as intact forests degrade, they can flip from carbon sinks to net carbon sources, emitting more than they absorb and accelerating the very warming that stresses them. The spatially explicit maps produced by this study could help governments prioritise conservation zones where avoided deforestation yields the greatest climate benefit, and design community-based forest management schemes that recognise indigenous peoples as the primary stewards of these carbon reservoirs.

The study is candid about its limitations. The 2000 baseline may include some old plantations that the Hansen dataset cannot distinguish from natural forest; the model assumes predictor relationships stable across two decades; and the absence of explicit spatial block cross-validation means the reported accuracy figures may be somewhat optimistic. The economic valuation, by design, omits additionality, leakage, and monitoring requirements. Still, the core signal — a 59 percent collapse in intact forest area and a 39 percent loss of biomass carbon, tracked with low reported uncertainty — is difficult to dispute. Future work integrating radar-based biomass data and country-specific carbon accounting could sharpen the picture further. What this study makes clear is that Southeast Asia’s intact forests are not merely an ecological treasure; they are a quantifiable, and rapidly depreciating, planetary asset.

Subject of Research: Remote sensing and random forest modelling of intact forest biomass carbon dynamics and carbon economics in Southeast Asia from 2000 to 2020

Article Title: Intact forest carbon dynamics and indicative carbon economy in Southeast Asia: A remote sensing and random forest approach

Article References: Zahira, D. Z., Sakti, A. D., & Wikantika, K. (2026). Intact forest carbon dynamics and indicative carbon economy in Southeast Asia: A remote sensing and random forest approach. Environmental Advances, 26, Article 100758. https://doi.org/10.1016/j.envadv.2026.100758

Image Credits: AI Generated

DOI: 10.1016/j.envadv.2026.100758

Keywords: intact forests, Southeast Asia, biomass carbon, remote sensing, random forest, deforestation, oil palm, carbon pricing, REDD+, aboveground biomass, belowground biomass, machine learning

The central challenge the researchers confronted is one of the oldest problems in imaging science. A standard RGB photograph compresses the entire visible spectrum into just three broad bands, discarding the subtle reflectance signatures that betray chlorophyll loss, water stress, or the earliest stages of infection. Hyperspectral sensors capture this information across dozens of narrow wavelength bands, but they are expensive, bulky, and impractical to deploy across thousands of hectares of tropical plantation. Simply training a network to map RGB values directly to full spectra, however, is mathematically ill-posed: a three-dimensional RGB vector cannot uniquely determine a high-dimensional spectral signature, so many different plausible spectra correspond to the same photograph, producing unstable and physically inconsistent reconstructions.

The team’s answer was to stop treating spectral reconstruction as a single regression problem and instead decompose it into a sequence of physically meaningful stages. First, a coarse spectral cube is generated from the RGB input as an initial spectral prior, constraining the enormous space of possible solutions. In parallel, a multi-scale feature cube is extracted using a Multi-scale Atrous Convolutional Residual Network, capturing textures, edges, and structural patterns that provide contextual evidence when similar colors could mean different things. The visible spectrum, spanning 400 to 700 nanometers, is reconstructed first because it is most directly constrained by the RGB observation. The near-infrared bands, from 700 to 950 nanometers, which carry critical information about internal leaf structure and water status, are then estimated using the refined visible representation together with the spatial features. Finally, the visible and near-infrared sub-cubes are concatenated into a complete 3D hyperspectral cube with 31 spectral bands.

But richer data alone was not enough. Like most real-world agricultural datasets, the oil palm leaf collection used in the study, drawn from a public repository of 7,312 images across five health classes, was severely imbalanced. Potassium deficiency accounted for 4,076 samples while brown spot disease had only 475, a majority-to-minority ratio of roughly 8.6 to 1. Conventional fixes like random oversampling or simple geometric augmentation fail here because they cannot capture the complex distribution of high-dimensional spectral data. The researchers therefore built an imbalance-aware 3D Generative Adversarial Network that operates directly on volumetric spectral cubes. Its generator is conditioned on minority-class labels, so synthetic samples are produced preferentially for the rare conditions, while a class-weighted adversarial objective penalizes errors on underrepresented classes more heavily. Signed Distance Function representations add structural regularization, encouraging smooth spatial-spectral transitions and guarding against the twin failure modes of GAN training: overfitting and mode collapse.

The quality of the synthetic samples was assessed quantitatively using two complementary metrics. Spectral Angle Mapper values, which measure the angular similarity between real and generated spectral signatures, averaged just 2.609 degrees across the five classes, ranging from 2.409 degrees for healthy leaves to 2.813 degrees for brown spot. Fréchet Inception Distance scores, which quantify how closely the distribution of synthetic samples matches the real data, averaged 5.080, with healthy leaves again scoring best at 4.088. These consistently low values indicate that the generated cubes preserved the class-specific spectral characteristics rather than drifting into unrealistic territory, and visualization of the learned feature space showed compact, well-separated clusters emerging by the later training epochs.

The balanced, spectrally enriched cubes then feed into the final stage of the pipeline: a 3D Convolutional Neural Network. Unlike conventional 2D networks that process only spatial information, 3D convolutions slide across height, width, and wavelength simultaneously, allowing the model to learn joint spatial-spectral features. The architecture uses three successive convolutional blocks with 3 by 3 by 3 kernels, increasing from 32 to 64 to 128 filters, each followed by max-pooling. The first block captures local intensity variations and short-range spectral relationships, the second assembles these into texture patterns and spatial-spectral interactions, and the third produces abstract, class-discriminative representations that a fully connected layer maps to the five output classes: brown spot, white scale, healthy, nitrogen deficiency, and potassium deficiency.

The results were evaluated against six benchmark models, including Mask R-CNN, a DenseNet-based classifier, a Firefly Algorithm-optimized support vector machine, the CoDet pipeline, the BEiT vision transformer, and the lightweight DLMC-Net. The proposed framework achieved the lowest root mean square error in every class, recording 0.112 for healthy leaves, 0.128 for brown spot, 0.124 for white scale, 0.121 for nitrogen deficiency, and 0.116 for potassium deficiency. Coefficients of determination followed the same pattern, reaching 0.930 for healthy and potassium-deficient leaves, while ratios of performance to deviation peaked at 2.946 for the healthy class, well above the best competitor’s 2.444. An ablation study confirmed that no single component was sufficient on its own: reconstruction, generative balancing, and volumetric classification each contributed, and only their integration delivered the lowest errors and most stable convergence across all leaf conditions.

Statistical testing reinforced the picture. Paired t-tests across the five health classes found that the proposed method’s improvements in accuracy, precision, recall, and F1-score were significant against every benchmark at the 0.05 level, with 95 percent confidence intervals for the mean differences remaining entirely above zero. The margins were largest against Mask R-CNN and DnCNN and narrowest against the strongest competitors, BEiT and DLMC-Net, but even those smaller gaps remained statistically significant. The authors are careful to note an important caveat, however: the benchmarks were originally designed for RGB imagery, so the comparison primarily demonstrates that reconstructed spectral information adds discriminative value over conventional color-based approaches, rather than proving superiority over models fed directly acquired hyperspectral data from real sensors.

That distinction points to the framework’s main limitations. Because the hyperspectral cube is estimated rather than measured, its fidelity is ultimately bounded by the information contained in three RGB channels, and reconstruction errors may propagate into classification. Variations in illumination, camera characteristics, cultivar, geography, and disease presentation could all shift the distribution the model was trained on, and the study did not include dedicated robustness experiments with sensor noise, occlusion, or field lighting changes. The computational cost is also nontrivial, combining spectral reconstruction, adversarial training, and 3D convolutions, which the authors acknowledge may require model compression and lightweight variants before deployment on drones or edge devices becomes practical.

Even with those caveats, the work represents a compelling step toward democratizing spectral agriculture. If a hyperspectral-grade diagnosis can be extracted from an ordinary photograph, the barrier to monitoring vast plantations drops from the cost of specialized optics to the cost of computation. The researchers outline a roadmap that includes validating the reconstruction against directly acquired hyperspectral measurements under matched conditions, extending the dataset to more diseases and environments, exploring diffusion models for even higher-quality synthetic augmentation, and developing edge-optimized versions for real-time UAV and IoT monitoring. For an industry in which early detection of a single disease can save millions of dollars in lost yield, teaching cheap cameras to see what only expensive sensors could see before may prove one of the more consequential ideas in precision farming.

Subject of Research: RGB-driven hyperspectral reconstruction with imbalance-aware 3D GAN and 3D CNN for oil palm health assessment

Article Title: A novel RGB-driven 3D spectral representation learning framework with imbalance-aware 3D GAN and 3D CNN for oil palm health assessment

Article References: Hartono, Zuhanda, M. K., Syah, R., Ongko, E., Kuswardani, R. A., & Suswati (2026). A novel RGB-driven 3D spectral representation learning framework with imbalance-aware 3D GAN and 3D CNN for oil palm health assessment. Smart Agricultural Technology, 15, Article 102590. https://doi.org/10.1016/j.atech.2026.102590

Image Credits: AI Generated

DOI: 10.1016/j.atech.2026.102590

Keywords: oil palm, hyperspectral imaging, spectral reconstruction, 3D GAN, 3D CNN, class imbalance, plant disease detection, precision agriculture, deep learning, RGB imagery, nutrient deficiency, smart farming

The study, published in Plant Cell Reports, centers on a gene called EgKT2-1, which encodes a Shaker-type potassium channel in the African oil palm, Elaeis guineensis. Shaker channels are the workhorses of potassium transport in plants, voltage-dependent pores embedded in the plasma membrane that move massive fluxes of the ion across long distances. Each channel subunit carries six transmembrane segments, including a voltage sensor and a highly selective pore lined by the conserved TXXTXGYGD amino acid motif, plus a large intracellular tail bearing cyclic nucleotide-binding, ankyrin and KHA domains. Four subunits must assemble into a tetramer to form a functional pore, and the identity of those subunits determines whether the channel carries potassium into the cell, out of it, or in both directions depending on membrane voltage.

Using the oil palm reference genome, the researchers catalogued the entire Shaker channel family and found eight members, a count matching date palm and coconut but differing from the nine channels of Arabidopsis and ten of rice. Intriguingly, palms lack the silent, regulatory Group IV subunits found in many model plants, and they possess two members of the AKT2-like Group III subfamily, EgKT2-1 and EgKT2-2, where Arabidopsis and rice each carry only one. Sequence analysis placed EgKT2-1 squarely within the AKT2-like family, channels that typically behave as weak rectifiers, allowing potassium to flow both into and out of the cell. What happened next defied that expectation.

When the team expressed EgKT2-1 in Xenopus laevis oocytes and clamped the membrane voltage across a range from +70 to −170 millivolts, the channel produced large, time-dependent inward currents with the sigmoidal activation kinetics characteristic of plant inward rectifiers. Inward currents reached amplitudes of up to −25 microamperes, while outward currents were essentially negligible even at low external potassium. In other words, despite its family pedigree, EgKT2-1 functions as a one-way valve for potassium uptake, not a bidirectional conduit. The activation threshold sat near −50 millivolts, and current amplitude grew with external potassium concentration, confirming a genuine potassium-dependent conductance. The channel showed strong selectivity, passing rubidium only partially and carrying almost no current with sodium or lithium, and its activity was blocked by the classic potassium channel inhibitors cesium and barium, which suppressed currents by 75 and 80 percent respectively at −170 millivolts.

The atypical behavior has a plausible structural explanation. In rice, the OsAKT2 channel also behaves predominantly as an inward rectifier, and site-directed mutagenesis showed that a single lysine residue in the voltage-sensing S4 domain is a major determinant of that rectification. EgKT2-1 carries a lysine at the corresponding position, K195, whereas Arabidopsis AKT2 has two consecutive arginines there. The comparison suggests that weak rectification is not a universal property of AKT2-type channels but depends on specific amino acids within the voltage sensor. Equally puzzling was what the team did not find: unlike several AKT2 orthologs whose activity is dampened by external acidification, EgKT2-1 was indifferent to pH changes between 5.5 and 7.5, even though the residues previously implicated in proton sensitivity in Arabidopsis are conserved in the palm channel. That insensitivity remains unexplained and will require further structural work.

Where the gene is switched on proved just as revealing as how its protein behaves. Quantitative PCR showed EgKT2-1 transcripts abundant in leaves and shoot meristems of young palms, and in adult trees, expression peaked in leaves and in fruits during early development, between 30 and 60 days after anthesis, a window when fruit size and mass expand rapidly and when potassium content in the mesocarp is known to climb. Expression was low in flowers, roots, rachis, and in fruits at mid-ripening around 120 days after anthesis or maturity around 160 days. In situ hybridization sharpened the picture: in roots, transcripts localized to the cortex, phloem and pericycle cells adjacent to xylem poles, consistent with a role in potassium uptake and radial transport toward the vascular cylinder; in leaves, signals appeared in palisade parenchyma cells of the mesophyll, hinting at a contribution to photosynthetic tissue function; and in fruits, transcripts were detected exclusively in mesocarp cells at 60 days after anthesis, precisely the stage of peak potassium accumulation.

The timing matters because potassium is deeply entwined with the biochemistry of oil. As the most abundant cytosolic cation in plant cells, potassium activates enzymes, sustains turgor, drives phloem sap circulation and regulates stomatal opening. Among the enzymes it supports are pyruvate kinase and pyruvate dehydrogenase, which supply precursors for fatty acid synthesis. Earlier work on oil palm showed that low potassium availability perturbs pyruvate and mesocarp metabolism during fruit development, and that mesocarp potassium content rises through the first 105 days after anthesis before declining as lipid biosynthesis takes over. A channel that loads potassium into mesocarp cells during that early window could therefore help set the metabolic stage for the oil accumulation that follows, and strict inward rectification would be advantageous, favoring influx at negative membrane potentials while limiting losses under depolarized conditions.

The most striking result came from a field comparison. The researchers studied two oil palm progenies, C2 and C3, both derived from the same Deli by La Mé parental cross and grown under identical potassium fertilization at a plantation in Nigeria. Previous agronomic work had established that C3 produces significantly more bunches and more oil than C2 and allocates more potassium to its aerial organs. When the team measured EgKT2-1 expression in developing fruits, they found it nearly fourfold higher in C3 than in C2. The correlation is tantalizing: a channel that ferries potassium into the oil-rich flesh of the fruit is more active in the progeny that yields more oil. It raises the possibility that EgKT2-1 expression could serve as a molecular marker for oil production traits, potentially allowing breeders to screen young palms for high-yield potential before years of field trials.

The authors are careful, however, not to overclaim. The yield association rests on a comparison of just two progenies, and other genetic differences between them could equally explain the phenotypic gap, so the data establish a correlation rather than a causal link between EgKT2-1 and oil accumulation. The oocyte experiments prove the protein is an inward potassium channel, but its impact on oil biosynthesis in living palms remains inferred. Validation in larger, genetically diverse breeding populations, together with genetic approaches such as transgenic or CRISPR-based manipulation of EgKT2-1 expression, will be needed to test whether the channel directly drives yield differences. Still, the study delivers the first functional portrait of a Shaker channel in oil palm and opens a concrete path toward a longstanding goal: breeding palms that squeeze more oil from every kilogram of fertilizer, easing both the economics and the environmental footprint of the world’s most productive oil crop.

Subject of Research: Functional characterization of the Shaker-type potassium channel EgKT2-1 in oil palm and its correlation with potassium allocation and oil yield

Article Title: EgKT2-1, an inwardly rectifying Shaker-type potassium channel in oil palm that correlates with oil yield

Article References: Monder, H., Espeout, S., Zimmermann, S. D., Billotte, N., Verdeil, J.-L., Bocs, S., Ollivier, J., Impens, R., Jacob, F., Gaillard, I., & Cuéllar, T. (2026). EgKT2-1, an inwardly rectifying Shaker-type potassium channel in oil palm that correlates with oil yield. Plant Cell Reports, 45(10), Article 315. https://doi.org/10.1007/s00299-026-03975-7

Image Credits: AI Generated

DOI: 10.1007/s00299-026-03975-7

Keywords: oil palm, Elaeis guineensis, potassium channel, Shaker channels, EgKT2-1, inward rectifier, potassium transport, mesocarp, fruit development, oil yield, electrophysiology, plant breeding

The team set out to answer a deceptively simple question: where, exactly, would the land come from? Indonesia’s B50 mandate requires that half of all diesel fuel consumed nationally be derived from crude palm oil, a policy that has been progressively ratcheted upward over the past decade as the country sought to reduce imported fuel dependence and absorb domestic palm oil surpluses. At the same time, the government has pledged rice self-sufficiency through nationally planned food estates, a program with deep historical roots stretching back to presidential decrees of the 1990s and repeatedly revived by successive administrations. Both policies enjoy strong political momentum, and neither accounts explicitly for the other’s appetite for land.

Methodologically, the study is notable for its combination of an improved high-resolution land cover map with a multicriteria evaluation framework and production scenario modelling. The researchers used the analytic hierarchy process, a structured technique for weighting competing decision criteria first formalized by Thomas Saaty, to rank candidate parcels of land according to suitability for oil palm and rice cultivation. Criteria included agroclimatic conditions, soil characteristics, slope, accessibility and, critically, constraints designed to reflect current policy such as moratoria on new permits in primary natural forests and peatlands. Production scenarios were then modelled to determine how much land would need to be converted under different yield assumptions and land restriction regimes, with results disaggregated by island to capture Indonesia’s enormous geographic heterogeneity.

The headline numbers are stark. Meeting the B50 biodiesel target by 2030 would require converting between 4.85 and 8.55 million hectares of land to oil palm, depending on yield trajectories and the strictness of forest and peatland constraints. Achieving rice self-sufficiency through the planned food estate program could convert up to 2.3 million additional hectares. The modelling revealed substantial spatial overlap between the zones most suitable for food production and those targeted for energy crops, meaning the two flagship programs would compete directly for the same finite and increasingly scarce agricultural frontier. On islands such as Kalimantan and Papua, where much of the remaining forest estate lies, this overlap translates into direct pressure on intact tropical ecosystems.

The carbon consequences are potentially enormous. The study estimates that land conversion for oil palm expansion would release between 360 and 3,753 megatonnes of carbon dioxide equivalent, while conversion for rice cultivation could emit a further 509 to 1,297 megatonnes. The upper bounds of these estimates would surpass historic Indonesian emissions events, including the devastating 2015 fire crisis, when burning peatlands and forests released carbon on a scale that briefly made Indonesia one of the world’s largest national emitters. Satellite-based studies of that crisis estimated CO2 emissions exceeding 1,500 megatonnes in a matter of weeks, and the new analysis suggests that gradual, policy-driven land conversion could ultimately deliver a comparable or larger pulse of greenhouse gases, silently and legally, undermining the enhanced nationally determined contribution that Indonesia has submitted under the Paris Agreement.

The researchers emphasize that the ranges are wide precisely because policy choices matter. Where land conversion is constrained to already degraded or non-forest land, and where yields are improved through intensification rather than expansion, both land requirements and emissions fall dramatically at the lower end of the scenarios. Conversely, if constraints are relaxed, as some political proposals to open 20 million hectares of forest for food and energy production would imply, the upper bounds come into play. The yield gap analysis in the study shows that Indonesia’s existing rice and palm oil lands produce well below their agronomic potential, and that closing this gap through better seed varieties, fertilizer management and replanting of aging palm plantations could substitute for a substantial share of new land conversion.

Peatlands emerge as a particularly dangerous fault line in the analysis. Several of the planned food estates are located on or near deep peat soils, including areas with a troubled history such as the former Mega Rice Project in Central Kalimantan, an earlier attempt at rice self-sufficiency that drained vast peat swamps in the 1990s, produced almost no rice, and left behind a landscape chronically vulnerable to fire. Draining and converting peatlands releases not only the carbon stored in vegetation but also the far larger stocks oxidizing in the soil itself, generating emissions that continue for decades. The study’s emission estimates incorporate these soil carbon dynamics, which is why the upper-bound figures for rice are so severe relative to the land area involved.

The authors argue that the solution lies not in abandoning food or energy security but in integrating the planning of both. They call for policy frameworks that explicitly coordinate land allocation across sectors, prioritize intensification on existing agricultural land, enforce existing forest and peatland moratoria rigorously, and steer any necessary expansion toward degraded lands with low carbon stocks and low biodiversity value. The spatial tools developed in the study, which the team has made publicly available through a GitHub repository alongside the underlying data and code, are designed to support exactly this kind of integrated decision-making, allowing planners to visualize trade-offs before commitments are locked in. The analysis also implicitly speaks to a broader global debate about the land requirements of the energy transition, echoing recent findings from studies of food system efficiency in China and bioenergy expansion in Asia that show how poorly coordinated sectoral targets can multiply environmental damage.

For Indonesia, the stakes extend well beyond carbon. Forest conversion on the scale modelled would fragment some of the world’s richest biodiversity reservoirs, threaten the livelihoods of indigenous and local communities whose land tenure remains insecure, and expose the country to heightened fire and flood risks. The study’s authors, led by Claire V. Squire and Jiehong Lou, conclude that without a more coherent national land-use strategy, Indonesia risks trading short-term gains in fuel blending percentages and rice production statistics for long-term losses to its climate commitments, its forests and the communities that depend on them. Whether the country can reconcile its food, energy and environmental ambitions will be one of the defining sustainability tests of the coming decade, and the numbers in this analysis suggest the window for getting the balance right is rapidly narrowing.

Subject of Research: Spatial optimization of land use for biodiesel production, rice self-sufficiency and forest conservation in Indonesia

Article Title: Land-use optimization for food security, bioenergy and forest conservation in Indonesia

Article References: Squire, C. V., Lou, J., Parker, K. J., Schreier, M. A., Hilde, T. C., Sari, A., Lohff, L. C., Shah, K., & Hultman, N. (2026). Land-use optimization for food security, bioenergy and forest conservation in Indonesia. Nature Sustainability. https://doi.org/10.1038/s41893-026-01923-7

Image Credits: AI Generated

DOI: 10.1038/s41893-026-01923-7

Keywords: Indonesia, land-use optimization, biodiesel, rice self-sufficiency, deforestation, oil palm, food estates, carbon emissions, peatlands, forest conservation, bioenergy, Nature Sustainability