A 30% by 2030 Protected Areas Plan
for Canada and the United States
What is a 30% by 2030?
The current American and Canadian federal governments have committed to protecting at least 30% of their country's land and ocean by 2030 (Simmons et al., 2021; Skene, 2021); opening the discussion within the conservation industry about how to best achieve this '30x30' goal. Many of the targets surrounding biodiversity conservation are relatively unclear compared to other components of this target and there can be competing goals surrounding its implementation (Simmons et al., 2021). This shows that there is an opportunity for stronger planning for the conservation of biodiversity values within the 30x30 target.
Percent protected by ecoregion (Belle et al., 2018)
In addition, the commitment by both the Canadian and American governments brings the opportunity of collaboration between the two countries; especially considering they share many biodiversity features (Skene, 2021). By planning new protected areas in the context of both countries, lands can be prioritized for protection while considering the entirety of the countries' biodiversity values leading to more effective solutions to the larger issue. This type of planning could also implement better connectivity between protected areas to create more efficient conservation.
The Objective:
The objective of this project is to identify areas of Canada and the United States that would be most suitable to be added to the existing protected area network to achieve the 30% by 2030 goal of the Biden and Trudeau governments.
This project will also look at how a 'start-over' scenario would compare to our current protected areas (i.e. what pieces of land are of the highest priority to conserve in a theoretical situation where we had no existing protected areas).
Methodology:
Zonation:
Zonation Logo ( link )
Zonation (Moilanen, 2014) was used in this project to plan new protected areas for Canada and the United States.
Zonation is a raster-based planning tool that ranks land for the conservation of multiple biodiversity features (Moilanen et al., 2014). Using a specified removal rule, the software continually removes the least important raster cells while minimizing loss of the biodiversity feature values and accounting for their connectivity (Moilanen et al., 2014). The result is a raster whose cells are ranked based on their potential for conserving the inputted biodiversity features.
A large advantage of Zonation's prioritization algorithm is that it accounts for the principle of complementarity (Moilanen et al., 2014). This means that selection of new protected areas should complement (often via containing different features) those that are already established (Justus and Sarkar, 2002), making Zonation results much more optimal than species richness alternatives (Justus and Sarkar, 2002; Moilanen et al., 2014).
Biodiversity Features:
Five different biodiversity feature layers were used as a proxy for conservation in the Zonation analysis (Figure 1.): bird richness (Bateman et al., 2020), tree refugia index (Stralberg et al., 2018), elevation diversity, land facet diversity, and ecotypic diversity (Caroll et al., 2017). These layers are a combination of fine-filter (direct measures of biodiversity; Figure 1. A-B.) and coarse-filter (indirect measures of biodiversity; Figure 1. C-D.) features.
Figure 1. Priority values used in the Zonation scenarios 1 and 2 for Canada and the United States. Darker shades represent higher values. A. Bird richness during summer months (birds per km 2 ) from Bateman et al. (2020) B. Tree refugia index from Stralberg et al. (2018) C. Elevational diversity (m) from Caroll et al. (2017) D. Land facet diversity Gini-Simpson index from Caroll et al. (2017) E. Ecotypic diversity Gini-Simpson index from Caroll et al. (2017)
ArcMap (ESRI, 2019) was used to project all priority value layers into NA Lambert Azimuthal Equal Area ( see here ) with a 5km 2 resolution before input into Zonation. If the data had to be resampled, a cubic resampling technique was used and data were subsequently adjusted to be non-negative. Any null values were set to zero.
Scenarios:
Two scenarios were executed in Zonation (Moilanen, 2014): Scenario 1 ('start-over' scenario) and Scenario 2 ('locked-in protected areas' scenario). Scenario 1 only applied an analysis area mask that limited the ranking to land-based data. In addition to this, Scenario 2 applied a hierarchial removal mask that locked-in existing protected areas before ranking.
Both scenarios were run using the Core Area Zonation (CAZ) removal rule which ranks cells without being directly influenced by a single biodiversity feature's richness (Moilanen et al., 2014). In addition, both scenarios used edge removal, a warp factor of 200, and a boundary length penalty equal to 0. For more information on these settings or the Zonation software see its GitHub repo .
Once the scenarios were executed, the ranked layers were symbolized in ArcMap (ESRI, 2019). Scenario 2 was also subtracted from Scenario 1 to show the representativeness of existing protected areas.
Results:
Scenario 1 (the 'start-over' scenario):
Scenario 1 ranks show that conservation priority tends to increase as you move towards the coasts (Figure 2.). It seems that the majority of the West coast is a high conservation priority for both countries, along with the South-East coast of the United States. The low-ranked areas tend to coincide with plains, grasslands, and the tundra; whereas the high-ranked areas tend to coincide with more diverse range of cecosystems such as deserts, temperate forests, and the arctic.
Scenario 2 (the 'locked-in protected areas' scenario):
The main trends of the Scenario 2 ranks are relatively the same as those of Scenario 1 (Figure 2.). One noticeable difference at this scale is that very high ranks (0.8 - 1.0) become less dominating near the coasts. The low-ranked areas don't seem to be affected by this scenario. It is also apparent that many of the locked-in protected areas accounted for in this scenario are in low priority regions and as you move South, the size of the existing protected areas diminishes considerably.
Figure 2. Zonation scenario priority ranks for Canada and the United States using the Core Area Zonation (CAZ) removal rule. The left map shows scenario 1, only considering a land mask. The right map shows scenario 2, which considers existing protected areas for hierarchial removal.
Representativeness of Established Protected Areas:
The results show that the North, especially the Arctic of Canada, is generally over-represented by existing protected areas (Figure 3.). In addition, the South-West of the United States has some regions that are under-represented by existing protected areas.
Figure 3. Representativeness of existing protected areas compared to priority values in Zonation calculated by subtracting scenario 2 from scenario 1.
Choosing the 30% of land to protect:
Scenario 1 is not very realistic because it assumes that there are existing protected areas in Canada and the United States. However, it can be useful for visualizing general trends in priority ranks and for assessing the representativeness of the existing protected areas (as seen under the representativeness subheader).
On the other hand, Scenario 2 is more realistic because it has locked-in existing protected areas and ranks the remaining land according to their complementarity to these protected areas. For this reason, Scenario 2 was used for the final selection of protected areas to meet the 30x30 target.
To meet the 30x30 target, the top 30% of ranked cells from Scenario 2 were selected (Figure 4.) which accounts for the land already protected within both countries. The additional protected areas that were selected focus on the West coast of Canada and the United States; a highly-ranked region for both executed scenarios. There is also a focus on the South-East coast of the United States. Over-represented areas, such as the Arctic, do not have many additional protected areas selected. The more under-represented areas in the South-West region have more additional protected areas selected. These new protected areas complement the already established ones due to Zonation's algorithm.
Figure 4. Symbolized high-priority protected areas (top 30% of values) using the second Zonation scenario. The cumulative top 30% also includes existing protected areas, however they are symbolized in different colours to differentiate between the two categories.
Conclusion:
The commitment of Canada and the United States of protecting at least 30% of their lands and oceans by 2030 brings great challenges, but also opportunities for cooperative planning of new protected areas (Simmons et al., 2021; Skene, 2021). Zonation is a software that optimally ranks regions based on biodiversity features while considering complementarity to existing protected areas and connectivity (Moilanen, 2014). Both fine and coarse filter biodiversity features were used to prioritize a 5km 2 grid of Canada and the United States using the Zonation software. The results show that the North is over-represented by existing protected areas and that you move towards the coasts, priority ranks increase. As a result, the selected regions to optimally meet the 30x30 target focus on the West coast of both countries as well as the South-East coast of the United States. Future work could consider the connectivity of the protected areas in greater depth and could account for different conservation proxies via different biodiversity feature Zonation inputs.