Systemica is a company committed to developing projects with a high level of environmental integrity, in addition to generating positive social impact and contributing to the maintenance of biodiversity. For us, the analysis published on January 18, 2023 in The Guardian opts for an extreme criticism, based on less than robust scientific evidence. Our team of scientists rigorously analyzed all the documentation presented, and we concluded that the investigation is a missed opportunity to expand the necessary debate regarding the importance of the voluntary carbon market and possible methodological improvements.
On January 18, The Guardian published an article criticizing the quantification of Greenhouse Gas (GHG) reductions from ongoing projects registered with Verra. Although it is based on reputable research sources, the article fails to present the limitations of the models used by the researchers. A superficial analysis of the scientific studies highlighted shows that the authors are aware of the weaknesses in the conclusions and methodology of their studies, which leads to the inevitable conclusion that the material chosen by The Guardian cannot support the accusations against Verra.
Specifically, the headline that “more than 90% of Verra's carbon credits are worthless” is not in any of the studies mentioned in the article. In fact, this conclusion is based on an analysis of the vehicle itself. They reached this conclusion using the 2020 and 2023 results from the article written by Thales AP West et al, with a sample of 29 REDD+ projects. Extending the same conclusion to the entire VCS program is a blatant distortion, and the premise that the methodology presented in these academic studies provides an empirical analysis data set is even more problematic.
We are committed to the highest level of scientific integrity in the industry projects we work on, and it is imperative that we provide our perspective and contribution to a more accurate understanding of such important issues.
First of all, it is worth highlighting that predicting the future is not an easy task. Developing models that adequately describe the dynamic behavior of complex systems is a demanding and changing field of research, with many scientific and technological considerations. In general, models that predict broad behaviors have limitations in scope and accuracy, which are largely determined by specific assumptions and the complexity of the system in question. This could not be different for models that predict deforestation and that are the basis for creating project methodology that aim to avoid unplanned deforestation and quantify reductions in GHG emissions.
West et al's results are obtained through a modeling approach that uses so-called synthetic control groups to generate counterfactuals that supposedly emulate what would have happened in local areas if REDD+ projects did not exist. These synthetic controls are sample areas distributed across the country where the project is located to match some specific spatial characteristics. To produce a counterfactual, the controls are statistically weighted, so that their average historical deforestation rates follow the same line as the rates observed around the territory of each project.
The predictive capacity of these controls is judged by how well they can reproduce the projects' deforestation curve. If this description is satisfactory, the group is adopted as the standard to be used when determining the base scenario for the project site.
In most of the projects examined, the synthetic control model produces lower carbon offsets than the claims of project proponents, who developed their baselines from Verra methodologies. The carbon credits that correspond to this divergence are called “ghost credits” within The Guardian article.
However, VCU models in projects that prevent deforestation are not based solely on statistical models. Baseline estimates need to be supported by a consistent and robust description of the dynamics underlying deforestation, thinking in terms of local agents and characteristics. Understanding these dynamics is crucial to making concrete predictions, as it allows identifying the relevant variables that explain deforestation patterns in the project region and long-term evolution. Models that use synthetic controls have a significant limitation as they are unable to accurately capture the correlation between deforestation patterns and the complex interactions of local agents in the project areas.
Again, it is important to highlight that the authors of the cited articles, who express legitimate concerns about the integrity of the VCS program, are aware of these limitations.
Journalists also analyzed the results of an alternative counterfactual method described by Guizar-Coutiño et al (2022). The approach is similar to the synthetic control model, but alternatively using a selection of pixels representing forest areas in satellite images, which function as controls with spatial attributes corresponding to those of the REDD+ project's initial forest area. This set of pixels is used as a model to estimate the expected deforestation at the project site if the REDD+ project did not exist. However, as the pixels are spread over a much larger region, this method also fails when making correlations between local agents. Therefore, this issue does not provide a basis to discredit the methodologies adopted by Verra which, despite being imperfect, are always based on locally correlated data.
Guizar-Coutiño and the other authors of the scientific article used by The Guardian do not compare compensation claimed by project proponents calculated using this model. The authors also make clear the disagreement between the results obtained by them (more optimistic) and the results of the 2020 research by West and collaborators when comparing base estimates for 19 projects analyzed in both studies. Even so, they express support for the REDD+ model.
Journalists compared deforestation rates calculated by the model proposed by Guizar Coutiño et al with information available in the Verra database for 32 REDD+ projects, and concluded that predicted forest losses were inflated by an average of 406%. They did not reveal how they arrived at that number. Again, this information was based on a questionable assumption that the pixel-based control method accurately describes what would have happened at a given project location, as opposed to what would have happened using the synthetic model. In fact, both approaches have limitations.
Verra is currently conducting a complete review of its AUD methodologies. This process, supported by experts and public consultations, aims to improve the methodological requirements of deforestation prediction models, increasing the accuracy and reliability of their results. Measures like this show Verra's continued effort to increase the integrity of AUD projects, and are in line with our aspirations for a healthy and prosperous carbon market.
The voluntary carbon market is a valuable tool in the effort to curb global warming. It allows companies and individuals to offset their carbon emissions by investing in projects that reduce/avoid or sequester greenhouse gases. These projects, such as agriculture, renewable energy, forest conservation and reforestation, play an important role in preserving the environment and biodiversity, as well as promoting sustainable development in local communities. It is widely recognized that the REDD+ mechanism plays a crucial role in keeping forests standing, despite the problems identified in the report. REDD+'s ability to reduce emissions from deforestation and forest degradation and its ability to promote sustainable development makes it an important tool in tackling the 17 SDGs.
REDD+ projects have been successful in protecting threatened forests and encouraging biodiversity, as well as promoting sustainable livelihoods for local communities. Furthermore, REDD+ has created successful financial incentives for environmental conservation, essential for protecting forests in developing countries. Although there are indeed challenges associated with all REDD+ projects, it is important to emphasize that these issues are not inherent to REDD+ projects, being common in any type of large-scale effort within the environmental scenario. Proper planning, monitoring and assessments can correct and overcome these obstacles, ensuring that this important tool continues to help conservation and mitigate the effects of climate change.
We also believe it is important to highlight that responsibility for the integrity of the VCS standard does not rest solely with Verra. Yes, Verra requires project developers and Validation and Verification Bodies to follow ISO 14064 and 14065 standards to accurately conduct their projects and assessments. The methodologies were also created by the main market developers, following the guidelines of the United Nations Framework Convention on Climate Change and the guidelines of the LLUCF (Land Use, Land-Use Chance and Forestry, term of the climate policy that organizes emissions released and captured resulting from direct land use, such as settlements and commercial areas). This process allows Verra to implement common principles and ensures that all projects use the correct tools to achieve the best possible results. Even so, there is a great dependence on product developers to correctly interpret and follow these guidelines, making decisions that do justice to the initial spirit of the regulation. This partnership is what keeps forests standing in good preservation conditions, with biodiversity intact and social objectives met.
At Systemica, we are committed to maintaining the environmental and climate integrity of our REDD+ projects. In recent years, we have analyzed thousands of areas in the legal Amazon using different forecast models and various types of methodologies approved by Verra. We understand that protecting the forest and mitigating the effects of climate change are essential for a sustainable future. That's why we make a point of following strict standards and principles in every decision we make. We guarantee that our REDD+ projects are based on realistic estimates, and developed in a socially and environmentally responsible way.
Systemica has been automating our main methodologies and analyzes approved in the Agriculture, Forestry and Land Use (AFOLU) sector. We seek to amplify our projects, bringing agility to monitoring and verification processes and avoiding any type of error when estimating and quantifying emissions reductions. We are committed to transparency and accountability, and ensure that our projects undergo regular monitoring, reporting and verification.
We work side by side with local communities and stakeholders, and we believe that this participation is essential to the success of our projects. We also work closely with third-party and independent verifiers, who ensure that our projects always meet standards of excellence in environmental and social performance.
We have assembled an interdisciplinary team and developed strategic partners to implement our project activities, designed through social technologies that address the most urgent issues faced by communities in their daily lives. We are proud of the positive impact our projects have on the environment, local communities and the fight against climate change.
References
West, TA, Börner, J., Sills, EO and Kontoleon, A., 2020. Overstated carbon emission reductions from voluntary REDD+ projects in the Brazilian Amazon. Proceedings of the National Academy of Sciences, 117(39), pp.24188-24194.
Guizar‐Coutiño, A., Jones, JP, Balmford, A., Carmenta, R. and Coomes, DA, 2022. The global evaluation of the effectiveness of voluntary REDD+ projects at reducing deforestation and degradation in the moist tropics. Conservation Biology, p.e13970.
West, TA, Wunder, S., Sills, EO, Börner, J., Rifai, SW, Neidermeier, AN and Kontoleon, A., 2023. Action needed to make carbon offsets from tropical forest conservation work for climate change mitigation. ArXiv Preprint.
[1] For more details about these concepts visit: https://verra.org/wp content/uploads/2022/12/VCS-Standard-v4.4-FINAL.pdf
