Now Reading
India: We May Have Overestimated How Much Carbon Our Forests Can Hold

India: We May Have Overestimated How Much Carbon Our Forests Can Hold

Train through Lataguri Forest, West Bengal, India. Photo: Somdeep Datta/Wikimedia Commons


  • The climate mitigation potential of Indian forests, by building more carbon sinks, is overestimated, finds a new study.
  • Using a machine learning framework, the researchers defined additional areas suitable for forest restoration and agroforestry avoiding areas that do not support the growth of natural forests, such as grasslands, savannahs and woodlands.
  • The study finds that the additional mitigation potential from forest restoration in India is less than a quarter of the country’s commitments to the Paris Agreement in 2015.

Climate mitigation, by creating additional carbon sinks through forest restoration and agroforestry, would meet less than a quarter of India’s Paris Agreement 2015 goals in the land-use, land-use change and forestry category (LULUCF), says a new study.

India’s Nationally Determined Contribution to the Paris Agreement targets creating an additional carbon sink of 2.5-3 billion tonnes of carbon dioxide equivalent through additional forest and tree cover. India aims to bring 33% of its geographical area under forest cover by 2030 to fulfill its international commitments.

Scientists at the University of Oxford, University of Exeter, University of Hyderabad and Jawaharlal Nehru University used a machine learning framework with over 11,000 GPS points of forests across India, to map where natural forests can be sustained based on climate, soil and topographic factors (bioclimatic envelope) to avoid identifying areas that do not support the growth of natural forests, such as grasslands, savannahs and woodlands.

“This area mapped also includes current forest cover. After mapping this suitable area, we excluded all current natural forest areas (and other exclusions like wetlands, grasslands, savannahs etc.) and the remaining area which exists outside the current forest area is the opportunity that can act as carbon sinks if allowed to naturally regenerate,” study co-author Trisha Gopalakrishna, a DPhil candidate at the School of Geography and the Environment, University of Oxford, told Mongabay-India.

India’s commitments “overestimate the area available and the mitigation potential of forest restoration” and that forest restoration and agroforestry can only be “one of many strategies” fundamental to meet the goals of the Paris Agreement, the authors stress, adding that substantial greenhouse gas emission reductions in others sectors, most importantly, energy, will be needed.

The researchers estimate the additional feasible area for restoration to be only 1.58 Mha (million hectares), cumulatively sequestering 61.3 TgC (terra grams of carbon), which is substantially less than estimates derived from global studies.

The study also finds up to 14.67 Mha of restoration opportunity in the agroforestry sector in existing agricultural areas, delivering up to 98.1 TgC nationally, catering to only 19.5-23.4% of India’s LULUCF pledge to the Paris Agreement, which means that additional areas available for agroforestry “were very small compared to global India-specific estimates and India’s forest restoration commitments at the Paris Summit.”

With 0.26 Mha each, Chhattisgarh and Madhya Pradesh in central India had the highest additional area available for forest restoration that is suitable for sustaining natural forests. In comparison, Goa in western India and Mizoram in northeast India had the least opportunity of 0.002 Mha each; Delhi in north India and Tripura in northeast India showed no restoration opportunity.

The limited opportunity in the northeast reflected the areas available for restoration after excluding the vast forest swathes.

“Since much of the northeast is still forested when compared to other regions, forest restoration is not the best strategy to use. Rather we should continue to protect the already existing forests so that they continue to sequester carbon. With regards to the high opportunity present in the central Indian states, we attribute it to the forest loss that has occurred over many years creating the area to restore,” observed Gopalakrishna.

The criteria of additionality adopted in the study is an important benchmark, especially for carbon credits and the carbon markets. “The concept behind this criteria is that existing forests will continue to sequester carbon (if they are protected and not converted to another land use and/or land cover). And we should be excluding these areas because they already have forests, i.e. not double count carbon stocks accumulation,” she said.

Mitigation potential of forest restoration mapped by total area of each state (MgC/ha) (no agroforestry). Map by Gopalakrishna et al.

The dominant land use land cover spanning 1.58 Mha amenable to forest restoration are degraded forests and scrub. Building on the research, the scientists suggest further research to “investigate the social and cultural dimensions like forest governance and land tenure issues because of historical and current land-use legacies and other factors associated with the political economy of specific forest restoration schemes.”

Indian Forest Service (IFS) officer Pushpendra Rana, who was not associated with the paper, said, global studies like Gopalakrishna et al. can help build the overall perspective while designing afforestation plans. “The critical question is how can we downscale these global findings to suit local contexts and there lies the main challenge. These national-scale studies are highly likely to miss critical social, political, and economic factors specific to local contexts either due to lack of data or other reasons, which can pose problems in utilising the global findings in places where afforestation is actually done. So at the local scale, we will need to find ways to incorporate all these divergent and critical contexts into the restoration opportunity potential to develop effective and informed plantation programmes,” Rana told Mongabay-India.

Bridging research and policy gaps

“One of the main LULCs that has caused a debate is the inclusion or rather, limited exclusion of non-forest ecosystems such as grassy biomes such as grasslands, savannahs and woodlands. Current or historically grassy areas are distinct ecosystems that provide crucial ecosystem and societal benefits. Many global estimates identify these areas suitable for forest restoration, thereby threatening these systems i.e., promoting trees/forests in areas that cannot naturally sustain trees, which is called afforestation,” Gopalakrishna said.

The authors point to the Restoration Opportunities Atlas developed by World Resources Institute (WRI), India, the only existing India-specific study (Chaturvedi et al., 2018). Its novelty lay in identifying potential canopy densities in restorable areas. It also used carbon stocks data of naturally regenerating forests by forest type. The calculations of opportunity for forest and tree cover restoration in the Atlas come to 138 Mha – higher than the calculations in the latest paper by Gopalakrishna and co-authors.

The “relatively higher estimates of mosaic restoration strategy in the Atlas are due to the inclusion of all croplands with less than 40% forest canopy cover,” according to Gopalakrishna and co-authors. In their assessment, Gopalakrishna and co-authors ruled out all croplands and irrigated lands for food security. Further, the Atlas excludes a variety of LULCs from India’s total landmass resulting in higher estimates, Gopalakrishna and co-authors state.

Ruchika Singh, a co-author of the 2018 study, said that the Atlas addressed three critical challenges centred on data gaps, inaccessible and/or fragmented information from too many data sources that hamper evidence-based decision-making.

Shedding light on the methodology adopted in developing the Atlas, Singh and Jayahari K.M., another co-author of the 2018 paper, noted that all irrigated croplands were excluded from the mosaic restoration potential analysis (primarily from agroforestry) to recognise potential trade-offs with food security and livelihoods.

“For remaining crops lands, the maximum, achievable tree cover was capped at 20% and 40% (two scenarios) based on findings of a Planning Commission study (now NITI Aayog),” Singh, the director of sustainable landscapes and restoration at WRI India, told Mongabay-India.

Grassland and shola forest in Jodigere, Biligirirangan Hills. Photo: L. Shyamal/Wikimedia Commons

The present paper argues that the exclusion process (of LULCs), adopted by the Atlas, is the reason for the higher estimation of areas for restoration. Referring to this, Singh and Jayahari explained that the methodology for developing the Atlas, factors in safeguards for natural and fragile ecosystems by precluding these areas from the restoration opportunity assessment. These ecosystems include grasslands, deserts, sand dunes, permanent snow and ice, and scrubland.

“Urban and built-up areas were also excluded from the estimation. The difference in numbers could be because the paper (by Trisha Gopalakrishna and co-authors) excludes forest areas. In contrast, the restoration Atlas has included the restoration potential within forest areas,” added Jayahari.

Addressing the gap identified in the current paper about the Atlas not considering “natural disturbances, (e.g., fire) that prevent ancient open ecosystems (i.e., less dense forests, grasslands, and savannah-forest mosaics) to reach their maximum potential of forest canopy density,” Singh and Jayahari note that there was no need to consider natural disturbances because the Atlas aimed to assess the maximum potential in a condition where the disasters are controlled and prevented through better management.

“We do look at several enabling conditions around tenure, resources rights,” Singh told Mongabay-India, underscoring that the Atlas data is not recommended for developing implementation plans on the ground. “That necessitates a deeper landscape restoration planning at the district and sub-district level paired with developing detailed project reports or collaborative implementation plans at the village level,” Singh said.

This article was originally published on Mongabay-India.

Scroll To Top