Democratizing Conservation Planning

Google Scholar profile

The main research question that I tackle is: “How can we facilitate and support more sustainable decisions in conservation planning?”. As outlined below this question spans three interconnected topics: i) advancing methodologies for spatial planning; ii) developing tools and approaches that allow non-technical users to meaningfully interact with complex data (i.e. the core of the data democratization), and iii) developing datasets and information that can help inform planning approaches. My work spans a range of fields including remote sensing, spatial statistics, spatial planning, species management, and optimization techniques. I have published my research in a variety of fields by fostering collaborative networks among scientists from academia, government and non-governmental organizations, by having a keen eye for cross-fertilization among disciplines, and by developing broadly applicable, novel analytical techniques.

My vision for the longer-term future of my research covers two very large topics. First, although much of our work to date has focused squarely on conservation planning, the methods we developed are not restricted to that use.  As such, one of my goals is to expand our work to other decision contexts, such as land management and land use planning more broadly for an integrated planning approach. Second, partnering among different knowledge systems is something I care deeply about. I want to expand my work beyond a western science scope to accommodate other forms of knowing. The goal is to overcome barriers between knowledge systems and allow different ways of knowing to be an equal partner in the co-creation of knowledge in embracing two-eyed seeing and gathering in ethical space.

The following is a description of the three main areas of my research, including interdisciplinary projects of interest to faculty members at the Institute of Environmental Science and in associated departments.

  1. Advancing the methodology used in spatial planning for better informed decision-making

I have a strong interest in working on advancing the methodological aspects of spatial conservation planning. To date, I have published 17 peer-reviewed papers on this topic (117).

In the most recent example of this work, I have shown that exact integer linear programming solvers, as implemented in our open source R package prioritizr, outperform simulated annealing, as used in the current generation of conservation planning software (17). We have also recently developed a new method for conservation planning project prioritization that generates optimal solutions to the posed problems, instead of relying on heuristics, as previous methods have done (9). Additional papers related to primary methods development covered return on investment evaluations (12), combining species distribution models with value of information analysis (15), using species traits to help find new species occurrences (16), and identifying optimal highway crossing locations for mammals (2) among others.

A recent Nature Communications paper (8), that I led illustrates the intersection of my work on methods development and applied research well. For this work, we have used information from the community science platform eBird, based on 14 million unique observations across 1.7 million locations to optimize the conservation of 117 migratory bird species throughout their full annual cycle. These findings illustrate that community science data can help us better plan land management decisions.

  1. Tools and approaches that allow non-technical users to meaningfully interact with complex data

A crucial component of realizing the vision of democratizing the planning tools we develop is making these tools accessible to technical and non-technical users. For the technical users we produce R packages such as prioritizr or oppr and every paper I lead has all data and analytical code openly available either as a supplementary document accompanying the paper or in online repositories, such as the Open Science Framework or GitHub. This is an important aspect of my work, which I take very seriously, because I strongly believe that we as scientists have an obligation to share their information with others and make it as easy as possible for others to i) meaningfully use what we produce, and ii) be able to build on our work without huge amounts of effort to re-create what we did.

Another and perhaps even more important aspect of democratizing conservation planning (or even science in general), is the provisioning of tools and frameworks that allow less technically inclined people to implement them. This is a crucial component of my work, which is reflected by the 16 data science tools and software packages I have developed since 2016 and the list of mostly contracts that are included on this in the professional history section of my CV. In addition to the organizations that I also listed in my cover letter (Environment and Climate Change Canada, Fisheries and Oceans Canada, Mikisew Cree First Nation, Inter-Tribal Council of Michigan, Wildlife Conservation Society Canada, and Canadian Parks and Wilderness Society) the following two projects illustrate this aspect of my work.

Modernizing and democratizing Nature Conservancy of Canada’s (NCC) land use planning methods

This is a three-year (2020 – 2023), $308,000 project, funded through NCC to Joe Bennett and me as co-PIs. The primary focus of this work is to help NCC optimally allocate sparse conservation funds, to maximize the return on investment NCC can expect from their land acquisition and management decisions. This furthers their ultimate goal to ensure biodiversity persistence into the future. We are currently developing two sets of tools, one on prioritizing where NCC should work and one on what action it should take to reach its goals. The first tool uses the systematic spatial planning prioritizr R package, that I co-developed and am the maintainer of. The second tool uses the optimal project prioritization oppr R package, that I have co-developed (9). I am developing interactive web apps for both of these tools for easy access by NCC staff, and eventually anyone that has interest, via a public facing instance of the tool framework. We will place NCC’s and publicly-available biodiversity and socio-economic datasets and the latest optimization techniques behind a simple user interface for developing land use prioritization scenarios, allowing NCC staff to identify locations that are priorities for their specific values. Due to the computational efficiency of our techniques, our web-based platform will allow real-time identification of priority actions, areas and strategies for land use planning. The power to do so in real time will facilitate collaborative exploration of options, for example during stakeholder meetings.

Action mapping for essential life support areas (ELSA) – United Nations Development Programme

The goal of this project is to deliver across commitments to the UN Framework Convention on Climate Change, Sustainable Development Goals, and Convention on Biological Diversity. ELSAs are defined as areas that together conserve critical biodiversity and provide humans with essential ecosystem services, such as carbon storage, food, fresh water, water filtration, and disaster risk reduction. Oscar Venter from UNBC and I have developed the analytical planning framework for ELSA, which I built on prioritizr. The user interface is a web application that interested parties can interact with. To date, there are five pilot countries that have held ELSA workshops and for which I have developed the tools for (Costa Rica, Columbia, Kazakhstan, Peru and Uganda). Each country can take full ownership of the framework and analytical approach, including the code base, supported via training and consultation, if they choose to.

  1. Develop datasets and information that help inform these planning approaches

The third main topic of my research is the development of datasets and information that can inform topics 1 and 2 above, both for our own work, but also beyond, up to a global scale. This topic spans the fields of remote sensing, spatial statistics, and model fitting approaches including occupancy models and machine learning methods.

The first paper of my PhD involved a sophisticated modelling effort using detection/non-detection data and occupancy models at regional scale to estimate the probability of occurrence of songbirds(1). I further produced a machine learning approach to advance model fitting and expanded the approach to a Bayesian framework to test model fit and residual spatial autocorrelation (3). Eventually, I extended the scope of this work further to include a total of 73 bird species and an area of 27,250 km2 (12).

More recently, I have focused on continental to global scales for the development of datasets and information. The most recent, published global effort was to identify three global conditions (3Cs) for sustainable use and biodiversity conservation (18). The 3Cs framework evaluates land-use drivers and human pressures to establish a baseline state of three broad terrestrial conditions: cities and farms, shared lands, and large wild areas. In another global effort that’s currently in press in Nature Communications (but available as preprint (19)) we integrate data on observed and inferred human pressures and an index of lost connectivity, to generate the first globally-consistent, continuous index of forest condition as determined by degree of anthropogenic modification. Another global effort that I’m leading (manuscript in preparation) incorporates governance, land-use and climate risk into a prioritization analysis trying to maximize the persistence of 30930 vertebrate species. Finally, we are mapping critical natural assets for people globally. Our approach can inform a broader set of decisions around sustainable development.


  1. R. Schuster, P. Arcese, Using bird species community occurrence to prioritize forests for old growth restoration. Ecography. 36, 499–507 (2013).
  2. R. Schuster, H. Römer, R. R. Germain, Using multi-scale distribution and movement effects along a montane highway to identify optimal crossing locations for a large-bodied mammal community. PeerJ. 1, e189 (2013).
  3. R. Schuster, T. G. Martin, P. Arcese, Bird community conservation and carbon offsets in western North America. PloS one. 9, e99292 (2014).
  4. P. Arcese, R. Schuster, L. Campbell, A. Barber, T. G. Martin, Deer density and plant palatability predict shrub cover, richness, diversity and aboriginal food value in a North American archipelago. Diversity and Distributions. 20, 1368–1378 (2014).
  5. R. Schuster, P. Arcese, Effects of disputes and easement violations on the cost-effectiveness of land conservation. PeerJ. 3, e1185 (2015).
  6. J. A. Lee‐Yaw, H. M. Kharouba, M. Bontrager, C. Mahony, A. M. Csergő, A. M. E. Noreen, Q. Li, R. Schuster, A. L. Angert, A synthesis of transplant experiments and ecological niche models suggests that range limits are often niche limits. Ecology Letters. 19, 710–722 (2016).
  7. R. Schuster, E. A. Law, A. D. Rodewald, T. G. Martin, K. A. Wilson, M. Watts, H. P. Possingham, P. Arcese, Tax Shifting and Incentives for Biodiversity Conservation on Private Lands. Conservation Letters. 11, e12377 (2018).
  8. R. Schuster, S. Wilson, A. D. Rodewald, P. Arcese, D. Fink, T. Auer, J. R. Bennett, Optimizing the conservation of migratory species over their full annual cycle. Nature Communications. 10, 1754 (2019).
  9. J. O. Hanson, R. Schuster, M. Strimas‐Mackey, J. R. Bennett, Optimality in prioritizing conservation projects. Methods in Ecology and Evolution. 10, 1655–1663 (2019).
  10. A. D. Rodewald, M. Strimas-Mackey, R. Schuster, P. Arcese, Beyond canaries in coal mines: Co-occurrence of Andean mining concessions and migratory birds. Perspectives in Ecology and Conservation. 17, 151–156 (2019).
  11. C. Roy, N. Michel, C. Handel, S. Van Wilgenburg, J. Burkhalter, K. Gurney, D. Messmer, K. Princé, C. Rushing, J. Saracco, R. Schuster, A. Smith, P. Smith, P. Sólymos, L. Venier, B. Zuckerberg, Monitoring boreal avian populations: how can we estimate trends and trajectories from noisy data? Avian Conservation and Ecology. 14 (2019), doi:10.5751/ACE-01397-140208.
  12. A. D. Rodewald, M. Strimas-Mackey, R. Schuster, P. Arcese, Tradeoffs in the value of biodiversity feature and cost data in conservation prioritization. Sci Rep. 9, 1–8 (2019).
  13. J. L. McCune, S. R. Colla, L. E. Coristine, C. M. Davy, D. T. T. Flockhart, R. Schuster, D. M. Orihel, Are we accurately estimating the potential role of pollution in the decline of species at risk in Canada? FACETS (2019), doi:10.1139/facets-2019-0025.
  14. S. Wilson, R. Schuster, A. D. Rodewald, J. R. Bennett, A. C. Smith, F. A. La Sorte, P. H. Verburg, P. Arcese, Prioritize diversity or declining species? Trade-offs and synergies in spatial planning for the conservation of migratory birds in the face of land cover change. Biological Conservation. 239, 108285 (2019).
  15. C. V. Raymond, J. L. McCune, H. Rosner‐Katz, I. Chadès, R. Schuster, B. Gilbert, J. R. Bennett, Combining species distribution models and value of information analysis for spatial allocation of conservation resources. Journal of Applied Ecology. 57, 819–830 (2020).
  16. J. L. McCune, H. Rosner‐Katz, J. R. Bennett, R. Schuster, H. M. Kharouba, Do traits of plant species predict the efficacy of species distribution models for finding new occurrences? Ecology and Evolution. 10, 5001–5014 (2020).
  17. R. Schuster, J. O. Hanson, M. Strimas-Mackey, J. R. Bennett, Exact integer linear programming solvers outperform simulated annealing for solving conservation planning problems. PeerJ. 8, e9258 (2020).
  18. H. Locke, E. C. Ellis, O. Venter, R. Schuster, K. Ma, X. Shen, S. Woodley, N. Kingston, N. Bhola, B. B. N. Strassburg, A. Paulsch, B. Williams, J. E. M. Watson, Three global conditions for biodiversity conservation and sustainable use: an implementation framework. Natl Sci Rev. 6, 1080–1082 (2019).
  19. H. S. Grantham, A. Duncan, T. D. Evans, K. R. Jones, H. Beyer, R. Schuster, J. Walston, J. Ray, J. Robinson, M. Callow, Modification of forests by people means only 40% of remaining forests have high ecosystem integrity. bioRxiv (2020).