Since there are no currently active contests, we have switched Climate CoLab to read-only mode.
Learn more at
Skip navigation
Share via:


To reduce forest degradation and wildlife loss, we identify ecologically optimal forest remnant networks, prior to planned deforestation.



Chaco province, Argentina;

The Gran Chaco is the second largest tropical forest in South America after the Amazon, spanning 850,000 km² and home to myriad species, including jaguar and lowland tapir. It leads the globe in tropical deforestation rates, principally caused by soy and cattle production (ref. 1). Effective, scalable national and international incentives for forest conservation are lacking, and local efforts to mitigate planned deforestation in the Gran Chaco are largely uncoordinated and ineffective from an ecological standpoint (ref. 2).

Lowland Tapir; World Land Trust image

Agricultural fields separated by narrow forest strips; NASA image

Existing land use planning laws in the Gran Chaco require producers to maintain significant amounts of native forest cover after clearing. However, producers typically leave remnant forest in the form of hedgerows, in a grid-like formation of narrow, long, linear forest strips. These strips are prone to degradation from exposure to climate, solar radiation, herbicides and encroaching cattle, and are of little benefit to wildlife (ref. 3).Ecologically based planning of forest remnants would improve conservation outcomes without affecting farm productivity.

We will develop a model called “Yvyra” -meaning “tree” in Guarani- which will utilize remotely sensed information, national regulations, and stakeholder input to derive maps of landscape level remnant forest configurations while allowing for property level flexibility. Our aim is to avoid cumulative degradation and associated greenhouse gas emissions while promoting overall ecological utility. This effort can quickly and efficiently be implemented with direct stakeholder involvement, requiring no financial incentives, minimal effort for local government, and no additional regulations.

What actions do you propose?

Window of Opportunity - Incentives to Participate:

Given that South America’s Gran Chaco has the world's highest deforestation rate, long-term strategies are needed to reconcile agricultural sector interests with the sustainability of the region’s natural resources (ref. 1). Chaco forest nations- Paraguay, Bolivia, Argentina, and Brazil- have made commitments to the UNFCCC to address climate change adaptation and mitigation in the forestry and agricultural sectors (e.g., national INDCs). However, few approaches exist to realize these commitments.

The agricultural sector is facing mounting criticism for its role in deforestation and climate change at the local and international level. Business-as-usual agricultural clearing using hedgerows is viewed by many as poorly planned, not meeting the needs of wildlife or preserving ecosystem services. The farming sector seeks increased economic and political certainty as they often face slow land use permit processing times and threats of new zero deforestation regulations. As a result, the agricultural sector has demonstrated a newfound interest to improve their business practices and public relations.

A window of opportunity exists to modify practices to ensure continued agricultural productivity while improving the ecological utility forest remnants. In planning at the landscape-scale, ranchers and farmers and government agencies can make important improvements to public perception and environmental conditions, with simple, but cumulative changes. Producers in the Gran Chaco have recently come forward seeking to develop a coordinated forest conservation plan. We are in the process of model development and in-country stakeholder consultations with field crews currently in the field in Paraguay. 

Goal and Objectives:

In response to these challenges, we propose adjustments to the land permitting process. Through implementing the results of a spatially-explicit model, developed by this project, we will identify optimal forest conservation networks within and between properties to preserve and connect forest remnants. Coordinated optimization of remnant forest throughout the Gran Chaco landscape, planned ahead of deforestation, will improve the ecosystem services and protect remaining wildlife habitat and cumulatively achieve significant conservation and climate change mitigation outcomes.

Our goal is to minimize forest degradation and maximize ecological and social co-benefits of mandatory remnant forest areas left after planned deforestation. We will achieve this goal by:

1) Assessing stakeholder acceptance and potential effectiveness of the proposed forest optimization approach;

2) Determining conservation objectives and quantify landscape requirements;

3) Developing and implementing a temporally and spatially explicit optimization model to maximize remnant forest conservation across the Gran Chaco landscape.

We propose an integrated framework to meet these objectives which includes stakeholder surveys in the field, converting these surveys into useable data and mathematical models to be practically applied. 

Assessing Edge Effects- The Ecology:

Current regulations require landowners who deforest to set aside portions of their forest land often in a grid-like formation of long, narrow forest strips (hedgerows) that are typically 30-100 m wide. In some cases, landowners can also set aside forest blocks of varying sizes and dimensions to fulfill these requirements. Depending of jurisdiction, the amount of forest required to be preserved after clearing ranges from 10-40%. Hedgerows, however, suffer from rapid degradation caused by increased exposure to a harsher climate, solar radiation, pesticides/herbicides and encroaching cattle. The disappearance of hedgerows, coupled with ongoing mass deforestation, will lead to a complete and total loss of forests across large landscapes, eliminating all ecosystem services provided by the forest (e.g. micro-climate control, control of pests, water storage and filtration, the provision of food and forage, carbon storage, and flood and erosion control.). This outcome would negate the primary objectives intended by current conservation policies (e.g., forest retention requirements).  

There are numerous ecosystem services that we hope to preserve in the Gran Chaco. One way to effectively encapsulate the majority of ecosystem services within a conservation strategy is to conserve towards one that relies upon the provision of the others(ref. 4). Wildlife rely upon the majority of ecosystem services that this project seeks to conserve. Therefore, conserving wildlife habitat gives us a way to create more focused and quantitative conservation strategies. 

The current hedgerow configuration is largely unsuitable for large-bodied mammal habitat, such as lowland tapir, white-lipped peccary or jaguar, that are generally intolerant to exposure and threats, including human hunting pressure(ref. 2,3,5,6). Implications of continual and cumulative forest loss in the Gran Chaco may include extreme isolation of forest fragments that are left behind on the landscape, leading to eventual genetic isolation of wildlife populations, local species’ extirpation, and ultimately, contribute to species’ extinction. By preventing these outcomes for wildlife, we also maintain the other ecosystem services of interest.

Effects of degradation can be minimized and the utility for wildlife maximized by reducing the edge-to-forest ratio, creating larger forest blocks with fewer hedgerows, and wider corridors between forest blocks, and by prioritizing conservation of ecologically important landscape features, e.g. riparian zones. This would improve benefits for wildlife and support overall landscape connectivity, and help ensure the long-term persistence of the forest cover that is left in reserve (ref. 2,7,8). 

Methods Summary:

1) Data inputs:

Remotely Sensed: Landsat, Sentinel II, MODIS, SMAP (forest cover, existing deforestation maps)

Terrestrial georeferenced: Ecological hotspots, roads, cadastral data

Regulatory Information: International, national (Argentina, Paraguay, Bolivia), local-level

2) Assess stakeholder acceptance and efficacy of proposed conservation approach

We will conduct interviews of soy and cattle producers in Argentina, Bolivia, and Paraguay that have investments and property within the Gran Chaco biome. Further we will contact research and extension service providers, social and environmental NGOs, and applicable government agencies responsible for granting deforestation permits. We will interview these stakeholders to clarify legal parameters/ guidelines of mandatory forest retention laws, evaluate both policy and practice, and to better understand stakeholder perspectives regarding existing forest retention laws and potential ideas for improvement. This information will help determine the deliverables we produce, and feedback will be incorporated into model development, including parameterization and application.

3) Determine conservation objectives and quantify landscape requirements

a) Identification of landscape features

The first step for model parameterization is to establish conservation priorities. The aim of the model is to assess large landscapes, i.e. the entire Gran Chaco, while being applicable on the property level. The scale of the project requires information that can describe various aspects of the environment which can be obtained for the entire study area. We will obtain and map available data (including information on water bodies, forest cover, corridor or protected areas, hotspots, road networks, human settlements, etc.) to identify potentially important distinguishing landscape features that may have an ecological influence. This information will identify locations that should be considered for conservation (e.g., areas in proximity to lagoons or known breeding/nesting grounds) or locations that may be detrimental to ecological function and serve as a deterrent to wildlife (e.g., areas proximate to roads or human settlements).

We will then establish a relative weighting system for each mapped landscape feature to guide the model in its prioritization assessment. The weighting system will be guided by published literature and developed throughout the course of the project.

b) Edge effect assessment

To maximize the permanence of forests, the forest-to-edge ratio must be minimized. By reducing edge effects we also increase habitat suitability for wildlife. We will conduct a meta-analysis of published literature on wildlife tolerance to varying states of forest cover to evaluate the forest-to-edge ratios required for effective wildlife conservation within forest fragments and corridors. This information will establish the parameters for forest block and corridor size, configuration and widths. Various species classes will be evaluated (mammals, birds, reptiles, etc.). The most limiting species, in terms of forest-to-edge requirements, will dictate the parameterization in our model to reduce edge effect. The paucity of data on animal behavior in this specific area may require information on similar species in other areas of the tropics, as proxies. Expert opinion will be incorporated to complement a meta-analysis, including information from biologist stakeholder meetings/consultations to make final determinations for model parameterization.

4. Develop a temporally and spatially explicit model to maximize remnant forest conservation

Using our model, Yvyra, we will perform a randomized forest clustering and connectivity analysis that addresses the spatial and temporal uncertainty of deforestation events (i.e., order of property clearing), land subdivision or policy changes (i.e., changes to forest retention laws). The model offers a range of forest configurations, sizes and locations that are deemed optimal to ensure minimal edge effects and maximum connectivity within and amongst private properties (and abutting public land). The results of the model will allow for land-owner property-level negotiation on exact conservation configurations while maintaining the larger landscape level conservation plan.

A model prototype is currently operational. The following image provides sample outputs of the prototype on a hypothetical landscape:

Explanation of New and Unusual Techniques:

Our project is novel in the following ways: 

1. Remaining flexible: We will develop a framework to optimize forest remnant networks while maintaining property-level flexibility. Landowners can negotiate among a range of forest configurations and locations while optimizing forest connectivity across the landscape;

2. Planning ahead: Our model assesses presently intact forest that is slated for deforestation and plans for conservation -before- development enters a region, resulting in forest structures that can be preserved into perpetuity and providing greater ecological benefits;

3. Addressing uncertainty: Given uncertainty of deforestation dynamics, we will develop a spatially and temporally explicit model that is responsive to socio-political changes on the landscape as they occur (e.g., incorporating areas where new and emerging deforestation may radiate, and factoring in regulatory and policy uncertainty);   

4. Dynamic Conservation: Our model will spatially update according to changes on the landscape and property ownership, thereby allowing the incorporation of new information while preserving the relevance of previously conserved forest remnants

5. Stakeholder driven: Throughout model development we are incorporating transparent stakeholder engagement (surveys and outreach) that will be directly used in model development and implementation; and

6. Working within existing legal and policy framework: Our conservation approach requires no changes to laws that would restrict the agricultural sector’s development. Further no new incentives or international financing are needed to enable the project to succeed and operate. 

Who will take these actions?

We will engage key actors using a bottom-up and top-down approach. Mitigation of forest degradation will be led by large-scale cattle ranchers and soybean producers (bottom up), and will be reviewed and subject to the approval of government agencies across the Gran Chaco (top down). 

Key actors include:

Cattle ranchers and soybean producers (general focus is on large-scale producers averaging 1500 ha+): to assess project scoping requirements, determine feasibility of introducing adjustments to present land use planning laws, and discuss sector’s needs/preferences/attitudes to inform and parameterize model.

Indigenous and mestizo communities: to consult on potential effects proposed changes to land use planning requirements may have on this group, and to ensure full impacts of model development and outcomes are clarified and addressed.

National and provincial government authorities (oversight of ranching and farming sector; forestry; planning/permitting): to clarify legal questions, discuss feasibility of adjustments present legal requirements, and engage in model development and implementation.

Non-governmental organizations and university researchers (including locally based and international entities): to potentially collaborate, access new contacts and data resources.

Core project team members include:

Loretta Moreno, M.Sc., USA: Conservation planner and concept developer. Manager of project design and development. 

Adam Moreno, M.Sc., PhD Candidate, Austria: Ecological modeler with specialization in computer engineering and programming. Scientific advisor and developer of Yvyra model.

Dr. Yann le Polain de Waroux- Stanford University, USA: Land use change scientist with specialization in geography, remote sensing, and geo-informatics. Scientific advisor and co-developer of project.

María del Carmen Fleytas, M.Sc.- Wildlife Conservation Society, Paraguay: Country Program Director acting as key in-country contact and collaborator.  

Where will these actions be taken?

Ultimately, we aim to apply this approach to deforestation frontiers globally, such as in Africa and Asia, where a new wave of deforestation is taking shape given cheap land and a high demand for food products. This particular project focuses on the Gran Chaco biome of South America where we will carry out several pilot projects at the national and sub-national levels in: Argentina, Paraguay and Bolivia.  

How much will emissions be reduced or sequestered vs. business as usual levels?

While it is impossible to estimate degradation rates of remnant forest reserves in the Gran Chaco forest basin at this time, we can assume that within the proximate 100 years, remnant forest will be mostly or completely degraded due to the influence of edge effects coupled with the lack of microclimates necessary to maintain this subtropical forest type. Given deforestation dynamics in the Chaco, planning requirements for remnant forests, and based upon forest carbon estimates for this biome, we have estimated that 2.4 Gt of CO₂ are at risk of being lost through degradation.  This estimate represents current in situ forest carbon stocks and doesn’t take into account the continued carbon sequestered annually, which would be maintained by our project, but lost through business-as usual management practices.  

What are other key benefits?

If we are able to modify the way deforestation practices unfold on the landscape and optimize forest remnants to be larger and better connected across private and public lands, direct social and environmental co-benefits will be realized. These include improved ability of the landscape to deliver flood and soil erosion control, and a better water supply. Further, with less isolation of forest fragments, the threat of genetic isolation of wildlife populations and exposure to hunting pressure, which can lead to local species’ extirpations, will be reduced. Large areas of the Gran Chaco have human populations, including indigenous people, who are dependent on the ecosystem services provided by this forest, whether it’s hunting wild game, micro-climate control, or access to clean drinking water. It is our duty to ensure that agriculture and associated economic growth, while important, do not generate uninhabitable conditions for the occupants of the landscape.  

What are the proposal’s costs?


To achieve the outcomes of our optimization assessment we will include an extensive process of stakeholder consultation to ensure the model we develop and its products are in fact useful to producers and government agencies alike. However, in negotiating landscape-scale changes to well-established agricultural and forestry practices and policies, resistance and concerns related to possible opportunity costs can be expected. Cattle and soybean producers, upon making their initial investment in the deforestation frontier, usually have well developed plans and preferences about where they will to develop their property, given legal constraints. Any adjustment the way in which the law dictates forest remnant retention location, configuration, and size may impinge on these plans and preferences. This is why we will include stakeholders in the early stages of project development to ensure we gain their support. Further, the model and associated maps are equipped to provide a range of conservation options to guide landowners in their selection of forest remnant areas to conserve, given a landscape outlook. This allows the model to incorporate the complex needs and individual circumstances of farmers and ranchers in this deforestation frontier.        

Time line

Yr 5-15: Using the developed, parameterized, and validated “Yvyra” model simulated over the Gran Chaco, landowners will implement our landscape-level forest conservation plan. The region’s stakeholders including farmers, ranchers, indigenous communities, government agencies and NGOs are beginning to form an international consortium to implement this plan. We are working toward fine-tuning the model according to new scientific insights. We begin to plug our plan into international sustainable supply chain efforts, helping this region reorient its marketing campaigns and incentivizing landowners to expand their forest remnant networks. We are transferring our methods and experiences to other parts of world.  

Yr 15-50: After the ratification of an international pact to implement a landscape-level forest remnant conservation plan for the Gran Chaco, there has been a marked reduction of detrimental effects from agricultural expansion on wildlife populations, microclimates, etc. Large areas of native forest are interconnected by a series of corridors, which allows for animal movement, and can be accessed by indigenous groups. Forests that used the old hedgerow system are now seeing the effects of degradation, half the size they were before, thin and spindly, devoid of biodiversity and yielding no ecosystem services to local communities, farmers or wildlife.

Yr 50-100: The United Nations has now implemented a mandatory carbon offset and trading scheme that incentivizes forest conservation. Improved agricultural technology has slowed agricultural expansion and larger forest reserves can be worked into our landscape design. As our plan is designed to be flexible to changes in forest/agriculture policy these new reserves work into the landscape scheme seamlessly. Forests that were remnants from the old hedgerow method are now completely degraded and new afforestation efforts are underway to put new forest into areas fitting the international forest remnant conservation plan.

Related proposals

Not applicable. 



1. Hansen, M. C. et al. High-resolution global maps of 21st-century forest cover change. Science 342, 850–3 (2013).

2. Nunez-Regueiro, M. M. et al. Spatial patterns of mammal occurrence in forest strips surrounded by agricultural crops of the Chaco region, Argentina. Biol. Conserv. 187, 19–26 (2015).

3. Gascon, C., Williamson, G. B. & Fonseca, G. A. B. Receding Forest Edges and Vanishing Reserves. Science (80-. ). 288, 1356–1358 (2000).

4. Chan, K. M. A., Shaw, M. R., Cameron, D. R., Underwood, E. C. & Daily, G. C. Conservation Planning for Ecosystem Services. PLOS 4, 2138–2152 (2006).

5. Henle, K., Davies, K. F., Kleyer, M., Margules, C. & Settele, J. Predictors of species sensitivity to fragmentation. Biodivers. Conserv. 13, 207–251 (2004).

6.Turner, I. M. Species loss in fragments of tropical rain forest: a review of the evidence. J. Appl. Ecol. 33, 200–209 (1996).

7. Piquer-Rodríguez, M. et al. Effects of past and future land conversions on forest connectivity in the Argentine Chaco. Landsc. Ecol. 30, 817–833 (2015).

8. Ribeiro, M. C., Metzger, J. P., Martensen, A. C., Ponzoni, F. J. & Hirota, M. M. The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation. Biol. Conserv. 142, 1141–1153 (2009).