Prioritizing Areas for Land Conservation and Forest Management Planning for the Threatened Canada Warbler (Cardellina canadensis) in the Atlantic Northern Forest of Canada
Round 1
Reviewer 1 Report
A brief summary (one short paragraph) outlining the aim of the paper and its main contributions.This paper aims to identify priority areas in Canada’s Atlantic Northern Forest to be protected or sustainably managed for threatened Canada Warbler populations in this region. The authors compare the efficacy of several conservation and land management scenarios, which take into account factors such as population density, connectivity, future climate suitability, anthropogenic disturbance, dispersal, and recent Canada Warbler observations. Based on the resulting areas prioritized by each scenario, the authors provide recommendations for land managers and timber companies to protect Canada Warbler populations and/or mitigate negative impacts of forest management. This paper provides insights into which factors are most important to consider when prioritizing areas for Canada Warblers in Bird Conservation Region 14, which include future climate suitability, and estimates of dispersal distance. In addition, this work provides guidelines for users to determine the conservation value of specific areas in their region, and provides the tools for using Zonation analyses at finer spatial scales for local conservation planning.
Broad comments highlighting areas of strength and weakness. These comments should be specific enough for authors to be able to respond.Strengths
The authors used a strong foundation of existing data sources, including raw data, predictive models, and current ecological research to build their models based on sound knowledge of this species’ life history and habitat needs in this region. Because Canada Warbler habitat associations and responses to forestry vary regionally, I appreciated the scale of the study being limited to a single BCR, within which climatic and environmental conditions should be fairly consistent. The incorporation of spatial distribution, population density, and connectivity of occupied areas, is an elegant way to account for the patchy distribution and conspecific attraction behaviors noted in this species. In addition, I appreciated that the authors included comparisons of scenarios of both current and future climatic conditions that could potentially alter the Canada Warbler distribution, and therefore alter which areas need to be prioritized. Lastly, the consultation with land conservation agencies and forest managers, and the included technical report ensure that this research will provide usable tools for local planning.
Weaknesses
In general, I would like to see a more detailed rationale of: 1) the scale of study; and 2) the land conservation and forest management approaches.
1) Scale
I would like more information on how the scale of the study was chosen. If Canada warbler density in an area is driven by local vegetation structure and complexity, as well as social factors, then is it reasonable to make forestry decisions based on the course scale of landcover covariates? If the scale was chosen based on the Bird Conservation Region, a description of what BCRs are, and their relevance to single-species conservation planning would be valuable. Would the authors suggest a similar method to apply Zonation to prioritize areas for Canada Warblers in other BCRs? What scale do forest managers or provincial conservation organizations make decisions? Are the results of this study (at the BCR level and provincial level) adequate to inform management?
2) Land conservation and forest management approaches.
One of the strengths of this paper is the consultation with stakeholders to develop scenarios for prioritization. It would be useful for the reader to know what feedback and priorities were provided by consulted groups, which led to the development of the seven prioritization scenarios. What exactly are the “land conservation” and “forest management” approaches? What types of forest management would be consistent with promoting colonization of harvested areas? Clear-cutting? Selective harvesting? Understory retention? It would be useful to know what types of harvesting occur in this region, and if they all would be presumed to result in similar outcomes for vegetation regeneration and Canada Warbler colonization. Studies show a varied response of Canada Warbler occupancy, use, and breeding success in response to forestry across their range, likely due to underlying regional characteristics which might alter regeneration rates and overall system productivity of forests (e.g. climate, vegetation, soils), and differences in disturbance intensity. For example, harvest methods can range from removing a relatively small percentage of trees (selective harvest) which generally has minimal effects on forest species loss, to all or most trees (e.g. clearcut harvest) which can result in drastic changes in species composition. Becker et al. (2012) found that Canada Warblers had higher relative abundance in forests that were managed using selection harvest vs. clearcut harvest.
Specific comments referring to line numbers, tables or figures. Reviewers need not comment on formatting issues that do not obscure the meaning of the paper, as these will be addressed by editors.
Line 93-95: The fact that this study area is at the interface of two biomes, makes me wonder about differences in Canada Warbler habitat selection due to potential climatic and vegetation differences, as this species is known to exhibit differential habitat selection across their range. In your opinion, is the BCR14 region consistent enough in these characteristics to be considered as a suitable conservation unit for this species?
Line 130-135: How likely are Canada Warbler to actually shift their distribution with changing climatic? Not only are there time-lags in biological responses to climate change, but some species or populations may not have the plasticity to shift distributions. Do you think that there is evidence that this species, or similar species have shifted, or might shift in the future in response to shifting environmental conditions? Should this be considered when determining which areas to protect or to harvest, if current conditions are still suitable?
Line 398-403: I would like more information on why particular factors were included (i.e. dispersal) whereas others that might be important for determining priority areas were not (e.g., productivity, finer-scale habitat/shrubs). It is wise to offer management solutions such as logging to provide suitable habitat when measures of fitness are not known? Any thoughts about non-ideal habitat use or ecological traps being formed in these areas?
Line 308-309: I would like to know more about why the authors believe that a single-species model for prioritizing conservation areas is more effective than multi-species models.
Line 349-352: Because priority area determination seemed particularly sensitive to variation in dispersal distance, it would be interesting to have more information on true dispersal distances for this species. Is it safe to assume they are similar to warbler species with similar morphometrics? Or should a range of warbler species with similar life histories (migration strategy and distance) be considered. For example, black-throated blue warblers have a more restricted and southerly breeding range than Canada Warblers and also do not migrate as far south in the winter, and both breeding latitude and migration strategy/distance are known to affect dispersal distance. Furthermore, Girvan et al (2007) found that natal dispersal was shorter distances than adult dispersal and suggested perhaps conspecific attraction or post-fledgling prospecting could be more important in determining where first year breeders settle. They also found that long distance dispersers were more common at the periphery of their ranges (trailing edges), which could mean that these populations could act as sources, and that variation in dispersal distance across the breeding range should be taken into account.
Author Response
Thank you for your thoughtful and detailed review. We reply to each of your points below, with your points in square brackets. We note that in our response, we refer to line numbers from the manuscript version with ‘track changes’ enabled, so that the reviewer can see what was changed when referring to the manuscript.
[Broad comments highlighting areas of strength and weakness. These comments should be specific enough for authors to be able to respond.
Strengths
The authors used a strong foundation of existing data sources, including raw data, predictive models, and current ecological research to build their models based on sound knowledge of this species’ life history and habitat needs in this region. Because Canada Warbler habitat associations and responses to forestry vary regionally, I appreciated the scale of the study being limited to a single BCR, within which climatic and environmental conditions should be fairly consistent. The incorporation of spatial distribution, population density, and connectivity of occupied areas, is an elegant way to account for the patchy distribution and conspecific attraction behaviors noted in this species. In addition, I appreciated that the authors included comparisons of scenarios of both current and future climatic conditions that could potentially alter the Canada Warbler distribution, and therefore alter which areas need to be prioritized. Lastly, the consultation with land conservation agencies and forest managers, and the included technical report ensure that this research will provide usable tools for local planning.]
Thank you for your kind words! We, too, are hopeful that we have been able to do work that can actually have a positive influence on conservation outcomes for the Canada Warbler in this region.
[Weaknesses
In general, I would like to see a more detailed rationale of: 1) the scale of study; and 2) the land conservation and forest management approaches.]
We address each of these points individually as below.
1) Scale
I would like more information on how the scale of the study was chosen. If Canada warbler density in an area is driven by local vegetation structure and complexity, as well as social factors, then is it reasonable to make forestry decisions based on the course scale of landcover covariates? If the scale was chosen based on the Bird Conservation Region, a description of what BCRs are, and their relevance to single-species conservation planning would be valuable. Would the authors suggest a similar method to apply Zonation to prioritize areas for Canada Warblers in other BCRs? What scale do forest managers or provincial conservation organizations make decisions? Are the results of this study (at the BCR level and provincial level) adequate to inform management?
The scale of the study was chosen based on the BCR. We have added a new paragraph to section 2.1 Study Area (lines 103-111), to explain what BCRs are and why we’ve chosen the Canadian portion as the study area. It reads:
“The North American Bird Conservation Initiative defines BCRS as ecologically distinct regions that share similar bird communities, habitats, and resource management issues [34]. We selected the scale of the study as the Canadian portion of the Atlantic Northern Forest (BCR 14) for three reasons. First, BCRs are used by Canadian government agencies as planning and management units for other bird species at risk (e.g. [24,35]). Secondly, the Canada Warbler shows evidence of differential habitat selection across BCRs [36], and we wished to limit the modelling to a population with shared habitat requirements. Finally, limiting the scale of the study to the Canadian portion of BCR14 allowed us to use interoperable data and take advantage of existing Canadian bird conservation and forestry networks comprised of government agencies, industry, and NGOs.”
We appreciate the suggestion regarding our method being applied to other BCRs. We did have a similar line at the end of the conclusion but it wasn’t explicit enough, so we have rewritten it to read (lines 476-480):
“We suggest that this approach, designed to support management objectives for a single species, be applied to other species. In particular, we encourage managers to apply this prioritization approach to Canada Warbler populations in other BCRs in Canada and the U.S. to support persistence of the entire breeding population.”
As to the reviewer’s question, “Are the results of this study (at the BCR level and provincial level) adequate to inform management?”, we are not sure that the paper is the appropriate place to answer this question. We have provided the results of the study to government agencies, NGOs, and industrial operators. We have received feedback from some partners that the results have been useful for their purposes, but we have not formally studied the implementation of our results and their effectiveness (would be a good avenue for future research!).
[2) Land conservation and forest management approaches.
One of the strengths of this paper is the consultation with stakeholders to develop scenarios for prioritization. It would be useful for the reader to know what feedback and priorities were provided by consulted groups, which led to the development of the seven prioritization scenarios. What exactly are the “land conservation” and “forest management” approaches? What types of forest management would be consistent with promoting colonization of harvested areas? Clear-cutting? Selective harvesting? Understory retention? It would be useful to know what types of harvesting occur in this region, and if they all would be presumed to result in similar outcomes for vegetation regeneration and Canada Warbler colonization. Studies show a varied response of Canada Warbler occupancy, use, and breeding success in response to forestry across their range, likely due to underlying regional characteristics which might alter regeneration rates and overall system productivity of forests (e.g. climate, vegetation, soils), and differences in disturbance intensity. For example, harvest methods can range from removing a relatively small percentage of trees (selective harvest) which generally has minimal effects on forest species loss, to all or most trees (e.g. clearcut harvest) which can result in drastic changes in species composition. Becker et al. (2012) found that Canada Warblers had higher relative abundance in forests that were managed using selection harvest vs. clearcut harvest.]
In a number of places in the manuscript we refer to two companion reports which detail: a) suggested forest management practices to maintain persistence of Canada Warbler based on a literature review, which includes Becker et al. 2012 (reference #13); and b) the consultation approach used with partners during the development of the prioritization scenarios (reference #37). If the reviewers and editors like, we would be happy to include the habitat guidelines as appendices or supporting materials. We have added new lines summarizing the guidelines. The pre-existing text (lines 46-53) read:
“Management guidelines recently produced for BCR 14 [13] describe two different approaches to promote recovery of this species: permanent land conservation (hereafter ‘land conservation’, or ‘LC’) and responsible forest management (hereafter ‘forest management’, or ‘FM’). Each of these approaches were developed in tandem with, and designed for, two different user communities with distinct needs: 1) land conservancies and government agencies with a mandate to protect species and habitat in-situ; and 2) forest industry staff and government agencies with a mandate to engage in sustainable development of forest resources while minimizing impacts to migratory birds and species at risk.”
And we have appended new lines (lines 53-61) which read: “LC scenarios summarized in Westwood et al. [13] are intended to support conservation activities such as in-situ permanent protection of forested areas supporting Canada Warbler populations, forest blocks with low edge-to-interior ratios, and suitable habitat patches connected by forested corridors to other potential breeding sites to facilitate dispersal. FM scenarios are intended to support responsible forest management activities such as providing a continuous supply of breeding habitat on the landscape, avoiding harvesting and road building in population centres and forested wetlands, and locating areas to implement silvicultural systems most likely to produce desired conditions for breeding habitat 12-20 years after harvesting, among other strategies [13].”
We have also updated line 134 in the document, which originally read: “(Table 1; see [13,37] for a description of stakeholder engagement).” With “Based on solicited input from land trust representatives, agency and industrial forest managers, and scientists familiar with Canada Warbler ecology, we developed four scenarios for land conservation and three for forest management (Table 1; see [36] for a description of stakeholder engagement and [13] for specific details on recommended responsible forest management practices for maintaining Canada Warbler populations on the landscape). “
The reports are:
Westwood A, Reitsma LR, Lambert D. 2017. Prioritizing Areas for Canada Warbler Conservation and Management in the Atlantic Northern Forest of Canada. High Branch Conservation Services. Hartland, VT.
Westwood AR, Harding C, Reitsma L, Lambert D. 2017. Guidelines for Managing Canada Warbler Habitat in the Atlantic Northern Forest of Canada. High Branch Conservation Services, Hartland, VT.
[Specific comments referring to line numbers, tables or figures. Reviewers need not comment on formatting issues that do not obscure the meaning of the paper, as these will be addressed by editors.
Line 93-95: The fact that this study area is at the interface of two biomes, makes me wonder about differences in Canada Warbler habitat selection due to potential climatic and vegetation differences, as this species is known to exhibit differential habitat selection across their range. In your opinion, is the BCR14 region consistent enough in these characteristics to be considered as a suitable conservation unit for this species?]
Between the four coauthors, we have observed Canada Warbler habitats in Nova Scotia, New Brunswick, Gaspesie (southwestern Quebec), Maine, New Hampshire, as well as Manitoba, northwestern Ontario, and Alberta. In our observations, the habitats in BCR14 occurred in moist forests with predominant ferns and mosses as groundcover. This is quite different from the somewhat drier habitats in western Ontario and the prairie provinces, which have extensive shrub cover, often on slopes. In our view, habitats across the Canadian portion of BCR14 are similar enough to analyse together, a perspective which has been reinforced by the strong contrast between eastern and central/western habitats.
[Line 130-135: How likely are Canada Warbler to actually shift their distribution with changing climatic? Not only are there time-lags in biological responses to climate change, but some species or populations may not have the plasticity to shift distributions. Do you think that there is evidence that this species, or similar species have shifted, or might shift in the future in response to shifting environmental conditions? Should this be considered when determining which areas to protect or to harvest, if current conditions are still suitable?]
We certainly acknowledge the potential for time lags in biological (especially vegetation) responses to climate change, as well as alternative types of responses to distributional shifts, such as genetic or behavioral adaptation. Bioclimatic models represent a simplification in this regard, but for the purposes of conservation prioritization, they provide the best available information on future habitat suitability for the species. We did not use them to predict specific scenarios of landscape change, and future projections were down-weighted to partially account for future uncertainty. We have modified our methods description to better reflect these considerations (lines 162-171):
“To account for projected climate-induced shifts in abundance, we used 4-km population density predictions from models based on climate, land-use, and topography covariates as compared to the baseline population density (detailed description in [47]). CAWA climate baseline included predicted population density from 1961-1990. Future projections were for the 2041-2070 time period (CAWA climate 2050) based upon a high-end, business-as-usual emissions scenario (A2), averaged over an ensemble of four global climate models from the Coupled Model Intercomparison Project (CMIP3) dataset [48]. Although these future projections do not incorporate anticipated lags in vegetation responses to climate change [17], we considered them a reasonable representation of long-term future habitat suitability for this species, which in this region is projected to experience relatively moderate shifts in density, as opposed to wholesale range shifts [47].”
[Line 398-403: I would like more information on why particular factors were included (i.e. dispersal) whereas others that might be important for determining priority areas were not (e.g., productivity, finer-scale habitat/shrubs). It is wise to offer management solutions such as logging to provide suitable habitat when measures of fitness are not known? Any thoughts about non-ideal habitat use or ecological traps being formed in these areas?]
We appreciate the reviewer’s suggestion. There simply were no data available for the study area on productivity, finer-scale shrub habitat (Lidar is only available for a small portion of the study area), or ecological traps. If these data become available for the study area in future, it would be interesting and useful to update the model to include them. We have updated both the methods and discussion to reflect the lack of available data on these (important) elements:
Lines 159-160: “We note that no data were available on productivity or occupancy for the Canada Warbler in this region, nor fine-scale data to describe habitat characteristics (e.g. Lidar)”
Lines 465-470: “Future prioritization efforts should include results of assessments of reproductive success and population viability wherever possible in order to most accurately target areas to maintain population density on the landscape. Such data would be particularly valuable if comparing harvested and unharvested areas, to test the hypothesis of whether areas undergoing forest management represent ecological traps for this species [67].”
[Line 308-309: I would like to know more about why the authors believe that a single-species model for prioritizing conservation areas is more effective than multi-species models.]
The line the reviewer is referring to is “Given this species’ high conservation concern, our single-species exercise has potential to aid the rapid implementation of conservation and recovery action.” We will note that we do NOT believe that single species model is more effective than multi-species models, and we do not state this anywhere in the text. Generally, multi-species approaches to conservation are recommended in the literature. However, we were requested by partner agencies from various sectors to do single-species work because of the particular conservation concerns around Canada Warblers. Our paper can be construed as a proof-of-concept for the utility of systematic single-species conservation prioritization when considering management objectives.
[Line 349-352: Because priority area determination seemed particularly sensitive to variation in dispersal distance, it would be interesting to have more information on true dispersal distances for this species. Is it safe to assume they are similar to warbler species with similar morphometrics? Or should a range of warbler species with similar life histories (migration strategy and distance) be considered. For example, black-throated blue warblers have a more restricted and southerly breeding range than Canada Warblers and also do not migrate as far south in the winter, and both breeding latitude and migration strategy/distance are known to affect dispersal distance. Furthermore, Girvan et al (2007) found that natal dispersal was shorter distances than adult dispersal and suggested perhaps conspecific attraction or post-fledgling prospecting could be more important in determining where first year breeders settle. They also found that long distance dispersers were more common at the periphery of their ranges (trailing edges), which could mean that these populations could act as sources, and that variation in dispersal distance across the breeding range should be taken into account.]
We address dispersal distances in detail in several portions of the manuscript, which we reproduce below. The brief answer is that there is no published information on ‘true’ dispersal distances for this species, and as we state in the manuscript, we have estimated dispersal distances based on other species with similar morphometrics. Relevant portions of the manuscript:
Lines 191-203:
“Due to scarce species-specific dispersal data, Carroll et al. [45] completed a prioritization analysis using an estimated dispersal distance (not specific to natal or breeding) of 10 km for landbird species. For Canada Warbler, there are no published natal dispersal estimates and only one known direct observation (10 km; L. Reitsma, unpublished data). Since natal dispersal distances of landbirds are correlated with wing length and body mass [46,47], the Canada Warbler’s average size (wing span = 20-22 cm, mass = 9.5-12.5g; Reitsma et al. 2010) would suggest a median dispersal distance of 50 km [46]. However, estimates for similar-sized species vary widely, from 0.5 km to 40 km [46–49]. Betts et al. [50] measured maximum breeding dispersal distances of 1-3 km for two species of forest-dwelling warblers whose ranges overlap that of Canada Warbler (Black-throated Blue Warbler, Setophaga caerulescens, and Blackburnian Warbler, Setophaga fusca). To account for the uncertainty regarding dispersal estimates for this species we evaluated three different natal dispersal estimates for each scenario: low dispersal distance (LDD, 5 km), medium dispersal distance (MDD, 10 km), and high dispersal distance (HDD, 50 km).”
Lines 399-403:
“Given that dispersal is a key parameter in spatial conservation prioritization [45], our findings highlight the importance of considering a range of dispersal assumptions. Future studies may benefit from multiple scenarios designed to isolate the different ways in which dispersal estimates can influence results.”
Reviewer 2 Report
Westwood et al. present the outcome of a series of simulations exercises to help identify priority areas for the conservation of Canada warbler (CAWA) in eastern Canada. The authors obviously put a lot of time and effort in the simulation exercises and they should be commended for it. However, I still have many questions after reading the manuscript and I think that answering some of them would help readers figure out the significance of this work. I have outlined a series of majors and minor comments below.
I whish could luck to the authors in their endeavor.
#Majors comments:
1) How did you account for the fact that you visited only some of the cells for the CAWA presence in 2005-2009 and 2010-2015? The inclusion of these variables seems to create problems in the prioritization exercise (i.e. they occupied cells end up being prioritized over everything else), so if you are unable to correct the bias in the cells visited vs those that have not been visited it seems to me that it would be better to use these variables as a validation tools. For example, you could report the mean score of the occupied cells versus the mean score of the unoccupied cells for each scenario and see which one did the best job.
2) What would the authors envision as responsible forest management? This is an important question for forest managers and I am a bit surprised it was not defined clearly in the manuscript.
3) The manuscript should mention the importance of Atlantic Northern forest of Canada for Canada Warbler. Although the breeding range extends past Ontario, into the western provinces of Canada, historic data indicate that the highest density of CAWA were observed in the east.
4) Using the standard deviation of the population density as a surrogate of uncertainty is suboptimal since it will give you a measure that is correlated with abundance (i.e. larger density will have a larger standard deviation). A better measure of uncertainty would be to use the coefficient of variation (CV; standard deviation/mean). The authors should be mindful, however, that for skewed distribution (such as count that is over-dispersed) the geometric coefficient of variation might be more appropriated.
#Minor comments:
Lines 108 to 109: Table 1 could be improved by simply identifying the data that are used as input in the model. The distinction between Land Conservation and Forest Management could be made in a specific column or by spelling it out in the scenario code column.
Lines 154 to 167: The link between the three selected dispersal distance and the literature cited is not explicit enough. The 5 km seems to be related to the work of Betts et al., the 10 km seems to be based on the only known direct observation, and the 50 km seems to be based on the Reitsma et al equation. If that is the case, you should make the links more obvious.
Lines 169 to 173: What are the advantage of Core Area Zonation over the other algorithm?
Lines 175 to 211: I had difficulties reconciling the information of LC and FM scenarios. Why where these models run separately? Presenting a table with what variables are being included in the prioritization exercise would definitely help the reader understand the difference between all the different scenarios.
Lines 179 to 181: What are the alpha values in the parenthesis?
Lines 179 to 181: Why are you subtracting the uncertainty in the prioritization exercise and not use a multiplicative process? I would see the inclusion of the error in the prioritization exercise as a multiplicative process rather than an additive one. Is this a limitation of the software?
In addition, the standard deviation of the model output will be related to the mean so you will simply apply a higher penalty to the area with higher densities, which is probably undesirable for your exercise. Given that the Coefficient of Variation (CV = sd/mean) is a ratio it would make more sense to weight the cells by it inverse [i.e., penalty =1- CV/max(CV)].
Lines 188 to 190: Are the CAWA climate baseline based on current climatic values? If so, should more weight be put on what we know now than what we think the future will be (i.e., CAWA climate 2050)?
Lines 198 to 199: Why did you not include the Administrative Units from the start in the LC scenarios? It seems like including them from the start would be more realistic given that there is nearly no conservation planning taking place at the federal level (i.e., multi-provincial) in eastern Canada.
Lines 291 to 294: Given the little amount of variability between the responses, curve presented in Figure 4. Does dispersal distance really matter? These results seems at odds with the results of Figure 3 and the discussion.
Lines 284-286: Looking at Figure 3 we can already see that the ranking of most cells will be unaffected by the dispersal scenarios. Therefore, I wonder if the mean cell-level differences in priority ranking and standard deviation of cell-level in ranking difference presented in Table 3 is a good way to compare across scenarios.
Lines 304 and 305: Is this surprising? Are there high densities of CAWA in this region? Or is this region known for being devoid of CAWA?
Lines 314-316: I do not agree that this is a critical input in the prioritization exercise since the output ranked highly the areas where CAWA was detected. Your prioritization exercise is therefore dependent on the amount of 1 km square cells that you visit during the surveys. In a context where we can't possibly survey every cell it might not be a good idea to rank highly only what we know.
Author Response
We thank the reviewer for their comments. We note that in our response, we refer to line numbers from the manuscript version with ‘track changes’ enabled, so that the reviewer can see what was changed when referring to the manuscript. We reply to each of your points below, with your points in square brackets.
[Westwood et al. present the outcome of a series of simulations exercises to help identify priority areas for the conservation of Canada warbler (CAWA) in eastern Canada. The authors obviously put a lot of time and effort in the simulation exercises and they should be commended for it. However, I still have many questions after reading the manuscript and I think that answering some of them would help readers figure out the significance of this work. I have outlined a series of majors and minor comments below.
I whish could luck to the authors in their endeavor.]
Thank you for your thorough and thoughtful review, and your comments have helped us strengthen the manuscript. Our responses are below, with your comments in square brackets.
[#Majors comments:
How did you account for the fact that you visited only some of the cells for the CAWA presence in 2005-2009 and 2010-2015? The inclusion of these variables seems to create problems in the prioritization exercise (i.e. they occupied cells end up being prioritized over everything else), so if you are unable to correct the bias in the cells visited vs those that have not been visited it seems to me that it would be better to use these variables as a validation tools. For example, you could report the mean score of the occupied cells versus the mean score of the unoccupied cells for each scenario and see which one did the best job.]It was in fact our intent to ensure that cells with known Canada Warbler observations were prioritized over everything else so as to a) prevent foresters using our results from harvesting in areas with known individuals, which is illegal according to the Species At Risk Act and b) indicate areas with likely existing Canada Warblers to prioritize for conservation. But the observations do indeed provide a biased view of high-quality habitat, and that is why we also included the density model in the prioritization exercise. It is the combination of these and other inputs that demonstrates the power of systematic reserve design software (i.e., Zonation) to identify priorities given multiple considerations and trade-offs.
[2) What would the authors envision as responsible forest management? This is an important question for forest managers and I am a bit surprised it was not defined clearly in the manuscript.
In a number of places in the manuscript we refer to two companion reports which detail: a) suggested forest management practices to maintain persistence of Canada Warbler based on a literature review, which includes Becker et al. 2012 (reference #13); and b) the consultation approach used with partners during the development of the prioritization scenarios (reference #37). If the reviewers and editors like, we would be happy to include the habitat guidelines as appendices or supporting materials. We have added new lines summarizing the guidelines. The pre-existing text (lines 46-53) read:
“Management guidelines recently produced for BCR 14 [13] describe two different approaches to promote recovery of this species: permanent land conservation (hereafter ‘land conservation’, or ‘LC’) and responsible forest management (hereafter ‘forest management’, or ‘FM’). Each of these approaches were developed in tandem with, and designed for, two different user communities with distinct needs: 1) land conservancies and government agencies with a mandate to protect species and habitat in-situ; and 2) forest industry staff and government agencies with a mandate to engage in sustainable development of forest resources while minimizing impacts to migratory birds and species at risk.”
And we have appended new lines (lines 53-61) which read: “LC scenarios summarized in Westwood et al. [13] are intended to support conservation activities such as in-situ permanent protection of forested areas supporting Canada Warbler populations, forest blocks with low edge-to-interior ratios, and suitable habitat patches connected by forested corridors to other potential breeding sites to facilitate dispersal. FM scenarios are intended to support responsible forest management activities such as providing a continuous supply of breeding habitat on the landscape, avoiding harvesting and road building in population centres and forested wetlands, and locating areas to implement silvicultural systems most likely to produce desired conditions for breeding habitat 12-20 years after harvesting, among other strategies [13].”
We have also updated line 134 in the document, which originally read: “(Table 1; see [13,37] for a description of stakeholder engagement).” With “Based on solicited input from land trust representatives, agency and industrial forest managers, and scientists familiar with Canada Warbler ecology, we developed four scenarios for land conservation and three for forest management (Table 1; see [36] for a description of stakeholder engagement and [13] for specific details on recommended responsible forest management practices for maintaining Canada Warbler populations on the landscape).“
The reports are:
Westwood A, Reitsma LR, Lambert D. 2017. Prioritizing Areas for Canada Warbler Conservation and Management in the Atlantic Northern Forest of Canada. High Branch Conservation Services. Hartland, VT.
Westwood AR, Harding C, Reitsma L, Lambert D. 2017. Guidelines for Managing Canada Warbler Habitat in the Atlantic Northern Forest of Canada. High Branch Conservation Services, Hartland, VT.
[3) The manuscript should mention the importance of Atlantic Northern forest of Canada for Canada Warbler. Although the breeding range extends past Ontario, into the western provinces of Canada, historic data indicate that the highest density of CAWA were observed in the east.]
We have updated the introduction to includes a BBS reference about declines being steeper in BCR14 than other BCRs (lines 42-44): “In the Atlantic Northern Forest (Bird Conservation Region [BCR] 14), Canada Warbler population declines from 1970-2017 have been much steeper than other those observed in other BCRs from the Breeding Bird Survey in Canada [6].” The reference has been added in the reference section. In terms of historical densities, if the reviewer has a reference they wish for us to include, we ask they include it in the next round of review so we can add it. We are not sure which reference they are referring to in this comment.
[4) Using the standard deviation of the population density as a surrogate of uncertainty is suboptimal since it will give you a measure that is correlated with abundance (i.e. larger density will have a larger standard deviation). A better measure of uncertainty would be to use the coefficient of variation (CV; standard deviation/mean). The authors should be mindful, however, that for skewed distribution (such as count that is over-dispersed) the geometric coefficient of variation might be more appropriated.]
We considered using Coefficient of Variation instead of Standard Deviation, but we thought it more appropriate to discount density by a value that was in the same units as the prediction (i.e., SD), given Zonation’s additive approach (see response below). It is true that SD is not always a great measure of uncertainty because it necessarily increases with density. But for the purpose of discounting the density value, it is appropriate. Alternatively, one could use CV in a multiplicative fashion, but we did not see a need to do so.
[#Minor comments:
Lines 108 to 109: Table 1 could be improved by simply identifying the data that are used as input in the model. The distinction between Land Conservation and Forest Management could be made in a specific column or by spelling it out in the scenario code column.]
We feel it is important to include the description of the scenarios and their intended application to clarify for the reader what each scenario is. To the reviewer’s first point, we have added a column with included data layers for each scenario.
We do see the reviewer’s second point as to making the distinction between Land Conservation and Forest Management to economize words and avoid repetition. We have added a first column entitled ‘Scenario group’ and indicated ‘Land conservation’ or ‘Forest Management’ and removed those words from the scenario descriptions, and changed the title of the scenario description column to ‘Description of areas prioritized’.
[Lines 154 to 167: The link between the three selected dispersal distance and the literature cited is not explicit enough. The 5 km seems to be related to the work of Betts et al., the 10 km seems to be based on the only known direct observation, and the 50 km seems to be based on the Reitsma et al equation. If that is the case, you should make the links more obvious.]
We indicate in the relevant section that Carroll et al. used 10 km, Reitsma estimated 10 km, and that estimates for similar-sized species rang efrom 0.5-40 km. We have updated the final sentence of that section to read: “To account for the uncertainty regarding dispersal estimates for this species and capture the variation in dispersal estimates for similar species, we evaluated three different natal dispersal estimates for each scenario: low dispersal distance (LDD, 5 km), medium dispersal distance (MDD, 10 km), and high dispersal distance (HDD, 50 km).”
[Lines 169 to 173: What are the advantage of Core Area Zonation over the other algorithm?]
We have updated the relevant section of the methods to describe this in text (lines 205-210):
“Zonation uses three primary algorithms to prioritize raster cells for selection: core area zonation (CAZ), additive benefit function, and targets [18]. In this case we did not have an a priori conservation target, which is one reason we chose Zonation over the similar software Marxan [19], which requires proportional or area targets as inputs. We chose CAZ because ensures that the most valuable areas for each feature (“core areas”) are prioritized, rather than allowing trade-offs between features, as is the case with additive benefit functions.”
[Lines 175 to 211: I had difficulties reconciling the information of LC and FM scenarios. Why where these models run separately? Presenting a table with what variables are being included in the prioritization exercise would definitely help the reader understand the difference between all the different scenarios.]
We have added a column to Table 1 to detail the data layers used in each scenario. The models were run separately because each model includes a different combination of input data layers and/or different functions applied to these data layers (e.g. FM1 more highly prioritizes cells close to protected areas, whereas FM2 penalizes cells close to protected areas). The descriptions of how the functions are applied to each scenario are given in section 2.3.2.
[Lines 179 to 181: What are the alpha values in the parenthesis?]
Alpha is the kernel specified in the Zonation manual for uncertainty as associated with the distribution smoothing parameter, just as beta is the value the Zonation manual specifies to quantify uncertainty using the ecological interactions function. We have made this more explicit in the text and updated lines 219-222 and 233-237 to include a reference to the Zonation manual. They now read:
“We then applied the ‘Distribution Smoothing’ function to aggregate areas with cells of high population density connected by dispersal ability of Canada Warbler (dispersal kernel specified by α [18]; LDD = 5 km, α = 4 x 10-4; MDD = 10 km, α = 2 x 10-4 HDD = 50 km, α = 4 x 10-5)”
“We thus prioritized areas where high densities of Canada Warbler in both CAWA climate baseline (weight 0.75) and CAWA climate 2050 (weight 0.75) were connected based on dispersal distance over 36 years (dispersal ability kernel is represented by β [18]; LDD = 5 km dispersal/year, 180 km total, β = 1.1 × 10-5; MDD = 10 km dispersal/year, 360 km total, β = 5.56 × 10-6; HDD = 50 km dispersal/year, 1800 km total, β = 1.1 × 10-6).”
[Lines 179 to 181: Why are you subtracting the uncertainty in the prioritization exercise and not use a multiplicative process? I would see the inclusion of the error in the prioritization exercise as a multiplicative process rather than an additive one. Is this a limitation of the software?
Zonation uses an additive approach to uncertainty discounting that was first published in 2006 and has been widely in multiple peer-reviewed papers.
Moilanen A., Wintle B.A., Elith J., & Burgman M. (2006) Uncertainty analysis for regional-scale reserve selection. Conserv Biol 20, 1688-1697.
This is not necessarily a limitation of the software, because it is also possible to manually discount input features by a multiplicative factor, but we did not see a good reason to do so (see response to the next question).
[In addition, the standard deviation of the model output will be related to the mean so you will simply apply a higher penalty to the area with higher densities, which is probably undesirable for your exercise. Given that the Coefficient of Variation (CV = sd/mean) is a ratio it would make more sense to weight the cells by it inverse [i.e., penalty =1- CV/max(CV)].]
We considered using Coefficient of Variation instead of Standard Deviation, but we thought it more appropriate to discount density by a value that was in the same units as the prediction (i.e., SD), given Zonation’s additive approach. It is true that SD is not always a great measure of uncertainty because it necessarily increases with density. But for the purpose of discounting the density value, it is appropriate. Alternatively, one could use CV in a multiplicative fashion, but we did see a need to do so.
[Lines 188 to 190: Are the CAWA climate baseline based on current climatic values? If so, should more weight be put on what we know now than what we think the future will be (i.e., CAWA climate 2050)?]
The reviewer is referring to the fact that both CAWA climate baseline and CAWA climate 2050s received weights of 0.75 as opposed to CAWA population density which received a weight of 1.0. We chose to assign 0.75 to CAWA climate baseline because this layer was based on climate variables alone and we wished Population density, a finer-scale model based on more robust variables, to be prioritized in stead of a more coarse model.
[Lines 198 to 199: Why did you not include the Administrative Units from the start in the LC scenarios? It seems like including them from the start would be more realistic given that there is nearly no conservation planning taking place at the federal level (i.e., multi-provincial) in eastern Canada.]
Several of our partners in scenario development represent nature conservancies looking to acquire lands to serve as refuges Canada Warbler. Other partners were federal and provincial governments looking to partner with other organizations to protect and manage lands for Canada Warbler. Thus, it did not make sense to stratify lands by ownership from the beginning because these partners were looking to find priority areas to focus their efforts, irrespective of ownership. However, for our forester partners, it made sense to use Administrative Units because they would only be applying interventions on lands tenured for forestry purposes.
Our authors would disagree as well that no conservation planning is happening at the federal level that applies to eastern Canada, and would refer the reviewer to the Aichi Target 1 process and the Key Biodiversity Areas program run by ECCC and WCS.
[Lines 291 to 294: Given the little amount of variability between the responses, curve presented in Figure 4. Does dispersal distance really matter? These results seems at odds with the results of Figure 3 and the discussion.]
We believe that the reviewer is referring to Figure 5 (the efficiency curves) rather than Figure 4. As described in the text (289-291): “For all scenarios, efficiency (proportion of current predicted Canada Warbler population retained per unit area) declined with higher dispersal distance estimates. Including climate change reduced land conservation scenario efficiency with respect to the current population (Figure 5).”
It is true that the efficiency curves do not show a marked difference. However, the actual locations prioritized on the landscape are VERY different under different dispersal scenarios as shown by Figure 3. Although the efficiency may be similar between dispersal estimates, basing one’s conservation or management decisions on the different dispersal estimate scenarios would lead to completely different locations being chosen on the landscape. We do still believe that dispersal distance is a very important factor for this reason, though its impact on model efficiency overall was low.
[Lines 284-286: Looking at Figure 3 we can already see that the ranking of most cells will be unaffected by the dispersal scenarios. Therefore, I wonder if the mean cell-level differences in priority ranking and standard deviation of cell-level in ranking difference presented in Table 3 is a good way to compare across scenarios.]
It is not true that the rankings of most cells were unaffected by the dispersal scenarios. Indeed there were large spatial differences among the scenarios, and those differences are reflected in Table 3. We have clarified that the Zonation rankings range from 0 to 1, so a mean value of 0.12 (the difference between high and medium dispersal distances) means that rankings differed by 12/100 on average. Individual cell differences could be substantially greater.
[Lines 304 and 305: Is this surprising? Are there high densities of CAWA in this region? Or is this region known for being devoid of CAWA?]
The reviewer is referring to lines (now) 350-352: “Within forest management tenures, the northern Gaspé Peninsula of Québec was consistently identified as the area where forest harvesting activities may be most practical while avoiding impacts to current Canada Warbler populations.” We have added a follow-up line (352-354): “This is consistent with Solymos et al. [38], who predicted lower average Canada Warbler population density in the Gaspé Peninsula as compared to the rest of the study area.”
[Lines 314-316: I do not agree that this is a critical input in the prioritization exercise since the output ranked highly the areas where CAWA was detected. Your prioritization exercise is therefore dependent on the amount of 1 km square cells that you visit during the surveys. In a context where we can't possibly survey every cell it might not be a good idea to rank highly only what we know.]
The observations do indeed provide a biased view of high-quality habitat, and that is why we also included the density model in the prioritization exercise. It is the combination of these and other inputs that demonstrates the power of systematic reserve design software (i.e., Zonation) to identify priorities given multiple considerations and trade-offs.
Reviewer 3 Report
This manuscript uses conservation planning software to assess differences in landscape scale conservation strategies for Canada Warbler breeding habitat. The unique angle of this manuscript is that the authors explicitly incorporated a range of natal dispersal estimates to evaluate differences in scenario outcomes. The authors also account for uncertainties in SDM estimates associated with under-sampled areas. The study provides a novel approach to assess and develop long-term, species-specific, landscape level conservation plans. The study is a natural progression of recent research that ties together two important themes in bird conservation: incorporating aspects of the annual cycle and using large-scale data to assess “dynamic conservation” approaches.
Introduction.
L46. “Responsible forest management” needs to be described. What does this actually mean? Are there established BMPs for forest management in BCR 14 to create Canada Warbler habitat?
Overall, the introduction would benefit by adding details to emphasize details of natal dispersal in birds for the following reasons: 1.) This journal is not a bird-specific journal and the concept of natal dispersal for migrating birds is not intuitive for broad audiences. 2.) The ecological / population dynamic consequences of dispersal (or colonization into new areas, etc.) are particularly important for a species that breeds in dynamic habitats, the impacts of forest fragmentation can be emphasized here, 3.) A brief statement summarizing why we don’t have accurate estimates for dispersal for small migrating birds would also bolster the importance of the approach used in this study. I think this can be done easily by adding a few sentences in the fourth paragraph.
The authors specifically addressed uncertainty associated with SDMs in under-sampled areas in the methods and discussion, however this important aspect of the study is not mentioned in the introduction but should be added in.
L76. What are the dispersal distance estimates based on, in general?
Materials and Methods.
This section is well written, easy to follow. The scenarios are comprehensive and relevant. I am not familiar with the Zonation software so can’t comment on the specific analytical methods but the authors did a nice job explaining the process.
L109: What is “high” density? How was it defined? Was the temporal dynamics of CAWA habitat selection / CAWA moving "neighborhoods" over time addressed in the modeling?
L149 / L225: The githib is a great resource and adds to paper.
L158: Change “Since natal dispersal..” to “Because natal dispersal…” (since implies times)
Results
The figures and tables contribute to the manuscript.
Discussion
L320-328. I appreciate the focus on the relevance of scale in the discussion of results, providing guidance for local management is a valuable addition.
L330. This paragraph doesn’t fit here or it needs to be expanded. What are the implications of your “attempt”? What are the take-aways?
L337-352. This is an important result.
Author Response
We thank the reviewer for their comments. We note that in our response, we refer to line numbers from the manuscript version with ‘track changes’ enabled, so that the reviewer can see what was changed when referring to the manuscript. We reply to each of your points below, with your points in square brackets.
[This manuscript uses conservation planning software to assess differences in landscape scale conservation strategies for Canada Warbler breeding habitat. The unique angle of this manuscript is that the authors explicitly incorporated a range of natal dispersal estimates to evaluate differences in scenario outcomes. The authors also account for uncertainties in SDM estimates associated with under-sampled areas. The study provides a novel approach to assess and develop long-term, species-specific, landscape level conservation plans. The study is a natural progression of recent research that ties together two important themes in bird conservation: incorporating aspects of the annual cycle and using large-scale data to assess “dynamic conservation” approaches.]
Thank you for your generous interpretation and we are appreciative for the value you found in our work. We also in particular appreciate your thoughtful, constructive review which made clear and valuable suggestions.
[Introduction.
L46. “Responsible forest management” needs to be described. What does this actually mean? Are there established BMPs for forest management in BCR 14 to create Canada Warbler habitat?]
The established BMPs are in the accompanying guidelines for habitat management which we would be happy to include as appendices or supporting material and which we reference throughout the manuscript. We have added new lines summarizing the guidelines. The pre-existing text (lines 46-53) read:
“Management guidelines recently produced for BCR 14 [13] describe two different approaches to promote recovery of this species: permanent land conservation (hereafter ‘land conservation’, or ‘LC’) and responsible forest management (hereafter ‘forest management’, or ‘FM’). Each of these approaches were developed in tandem with, and designed for, two different user communities with distinct needs: 1) land conservancies and government agencies with a mandate to protect species and habitat in-situ; and 2) forest industry staff and government agencies with a mandate to engage in sustainable development of forest resources while minimizing impacts to migratory birds and species at risk.”
And we have appended new lines (lines 53-61) which reads: “Summarized in Westwood et al. [13], LC scenarios are intended to support conservation activities such as in-situ permanent protection of forested areas supporting Canada Warbler populations, forest blocks with low edge-to-interior ratios, and suitable habitat patches connected by forested corridors to other potential breeding sites to facilitate dispersal. FM scenarios are intended to support responsible forest management activities such as providing a continuous current supply of breeding habitat on the landscape, avoiding harvesting and road building in population centres and forested wetlands, and locating areas to implement silvicultural systems most likely to produce desired conditions for breeding habitat 12-20 years after harvesting, among others [13].”
[Overall, the introduction would benefit by adding details to emphasize details of natal dispersal in birds for the following reasons: 1.) This journal is not a bird-specific journal and the concept of natal dispersal for migrating birds is not intuitive for broad audiences. 2.) The ecological / population dynamic consequences of dispersal (or colonization into new areas, etc.) are particularly important for a species that breeds in dynamic habitats, the impacts of forest fragmentation can be emphasized here, 3.) A brief statement summarizing why we don’t have accurate estimates for dispersal for small migrating birds would also bolster the importance of the approach used in this study. I think this can be done easily by adding a few sentences in the fourth paragraph.]
To address these points, we have revised the fourth paragraph (lines 69-81) to read:
“Most analyses using conservation planning software are designed to support the planning or evaluation of protected area networks [e.g. 20–22]. Other objectives include locating zones of interest for further study [24], evaluating trade-offs between land uses [25,26], or estimating the value of ecosystem services [27]. Conservation planning algorithms can be informed by estimates of dispersal, which is an important factor in wildlife habitat selection and response to climate change (i.e., the ability to move from current to future suitable habitats). For the Canada Warbler, published estimates of natal dispersal (distance from an individual’s birth site to their breeding site) are not available. Estimates of natal dispersal for other passerines are frequently based on small sample sizes with a great deal of uncertainty, and estimates for species of similar sizes to the Canada Warbler range widely [28–31]. Conservation planning exercises are typically based on single dispersal estimates, and the implications of uncertainty in these estimates are unknown.”
[The authors specifically addressed uncertainty associated with SDMs in under-sampled areas in the methods and discussion, however this important aspect of the study is not mentioned in the introduction but should be added in.]
We have added further material to the revised paragraph 4 (lines 81-84): “Furthermore, conservation planning algorithms often rely on species distribution models which predict habitat suitability, species occurrence, or population density across a landscape. However, these models are inherently more uncertain in under-sampled areas, and this uncertainty is also not always accounted for during conservation prioritization exercises.”
[L76. What are the dispersal distance estimates based on, in general?]
The rationale behind the 3 dispersal distance estimates is described in detail in the methods section (lines 190-203) and we would prefer to leave the rationale in that section to avoid repetition throughout the manuscript. The relevant paragraph is reproduced below for the reviewer’s reference:
“Connectivity functions in Zonation rely on estimates of dispersal distance to determine whether features or populations are ‘connected.’ Due to scarce species-specific dispersal data, Carroll et al. [52] completed a prioritization analysis using an estimated dispersal distance (not specific to natal or breeding) of 10 km for landbird species. For Canada Warbler, there are no published natal dispersal estimates and only one known direct observation (10 km; L. Reitsma, unpublished data). Since natal dispersal distances of landbirds are correlated with wing length and body mass [29,30], the Canada Warbler’s average size (wing span = 20-22 cm, mass = 9.5-12.5g; Reitsma et al. 2010) would suggest a median dispersal distance of 50 km [30]. However, estimates for similar-sized species vary widely, from 0.5 km to 40 km [28–31]. Betts et al. [53] measured maximum breeding dispersal distances of 1-3 km for two species of forest-dwelling warblers whose ranges overlap that of Canada Warbler (Black-throated Blue Warbler, Setophaga caerulescens, and Blackburnian Warbler, Setophaga fusca). To account for the uncertainty regarding dispersal estimates for this species and capture the variation in dispersal estimates for similar species, we evaluated three different natal dispersal estimates for each scenario: low dispersal distance (LDD, 5 km), medium dispersal distance (MDD, 10 km), and high dispersal distance (HDD, 50 km).”
[Materials and Methods.
This section is well written, easy to follow. The scenarios are comprehensive and relevant. I am not familiar with the Zonation software so can’t comment on the specific analytical methods but the authors did a nice job explaining the process.
L109: What is “high” density? How was it defined? Was the temporal dynamics of CAWA habitat selection / CAWA moving "neighborhoods" over time addressed in the modeling?]
We were unable to determine specifically which section the reviewer is referring to. In our version of the original manuscript, L109 corresponds to Table 1. Table 2 indicates the units of the Population density layer (males/ha), thus, ‘high density’ areas have more predicted males/ha than low density areas. The temporal dynamics of CAWA habitat selection were not addressed in the models constructed by Solymos et al.
[L149 / L225: The githib is a great resource and adds to paper.
L158: Change “Since natal dispersal..” to “Because natal dispersal…” (since implies times)]
Change made.
[Results
The figures and tables contribute to the manuscript.
Discussion
L320-328. I appreciate the focus on the relevance of scale in the discussion of results, providing guidance for local management is a valuable addition.
L330. This paragraph doesn’t fit here or it needs to be expanded. What are the implications of your “attempt”? What are the take-aways?]
Now that we have added a line in the introduction about uncertainty and undersampled areas in SDMs (at your prompting, thank you), this paragraph is better contextualized. However, we agree that in its original form it was not as well-connected as it should have been. Below we paste the original paragraph and the rewrite (
Original: “Species distribution models, such as those used as conservation objectives in our scenarios, include many assumptions regarding the drivers of a realized niche, and also assume an equilibrium between a species and its environment [56]. We attempted to account for uncertainties related to under-sampled areas [57,58] by discounting densities by the standard deviation of mean population density. By accounting for uncertainty in this way, we focused priorities on areas of lower scenario uncertainty [51,59].”
Rewrite (lines 378-385): “The predictions made by species distribution models, frequently used as input features in conservation prioritization exercises, are inherently more uncertain in under-sampled areas [60,61]. We accounted for this uncertainty by discounting densities by the standard deviation of mean population density, and thus were able to focus priorities on areas of lower scenario uncertainty [54,62].”
Reviewer 4 Report
This manuscript entitled "Prioritizing areas for land conservation and forest management planning for the threatened Canada Warbler (Cardellina canadensis) in the Atlantic Northern Forest of Canada" is in my believe an important and interesting manuscript. It is well written and structured, and gives interesting results. I only have one remark which I think should be clarified in the method section: in section 2.3.2 (line 146), in the different scenarios you refer to certain variables and gives their weight for in the model, see for example line 175. These weight need to be more specified!
Author Response
We thank the reviewer for their comments. We note that in our response, we refer to line numbers from the manuscript version with ‘track changes’ enabled, so that the reviewer can see what was changed when referring to the manuscript. We reply to each of your points below, with your points in square brackets.
This manuscript entitled "Prioritizing areas for land conservation and forest management planning for the threatened Canada Warbler (Cardellina canadensis) in the Atlantic Northern Forest of Canada" is in my believe an important and interesting manuscript. It is well written and structured, and gives interesting results. I only have one remark which I think should be clarified in the method section: in section 2.3.2 (line 146), in the different scenarios you refer to certain variables and gives their weight for in the model, see for example line 175. These weight need to be more specified!
We have added a sentence in section 2.3.2 prioritization analysis to explain how weights work in Zonation (lines 217-219): “In Zonation, features are generally given a standard weight of 1.0 unless they are to be discounted due to uncertainty (such as future climate projections) or increased in value due to particular management objectives [18].”
Round 2
Reviewer 2 Report
I have reviewed the manuscript “Prioritizing areas for land conservation and forest management planning for the threatened Canada Warbler (Cardellina canadensis) in the Atlantic Northern Forest of Canada” by Westwood et al. The revised manuscript addresses many points raised by the reviewers and is significantly improved. I thank the authors for all their efforts.
I have a few additional comments that I am sure that the authors will be able to address.
Minor comments:
In their rebuttal the authors mention that the one of their objectives of their prioritization exercise was to ensure that cells with known Canada Warbler observations were prioritized over everything else so as to prevent foresters using our results from harvesting in areas with known individuals, which is illegal according to the Species At Risk Act. I think that should touch on this subject as It provide a solid rationale for their inclusion in the prioritization exercise.
Figure numbering: Make sure that the Figure number in the text and in the figure caption match. With the current edit you seem to have two Figure 1.
Lines 208-210: There seems to be a word missing in the sentence “We chose CAZ because ensures that the most valuable […]”
Lines 330-332: Table 3: The author sticked to their guns and insisted to present the mean cell-level difference in priority ranking but, as they pointed out in their rebuttal, the distribution of their ranking in bounded between 0 and 1 which make the interpretation of the mean difficult for the average reader. I think it would be clearer and more straightforward to present the change would make more sense to present the change in ranking in term of percentage like they did in their rebuttal.
Author Response
Thank you for your continued attention to our paper. A note as we respond to your comments. Firstly, the manuscript version that we downloaded from the system still contained the ‘track changes’ from the previous round of revision, so we accepted them all so as to indicate new revisions with track changes. This has caused substantial changes in the numbering system. To be clear, we are including our ‘before’ and ‘after’ text in our response, and when we refer to line numbers, it is in the new version with track changes on.
[I have reviewed the manuscript “Prioritizing areas for land conservation and forest management planning for the threatened Canada Warbler (Cardellina canadensis) in the Atlantic Northern Forest of Canada” by Westwood et al. The revised manuscript addresses many points raised by the reviewers and is significantly improved. I thank the authors for all their efforts.]
Thank you for your efforts! Genuinely! You’ve clearly put a lot of time and consideration into our paper and its clarity is greatly improved as a result.
[I have a few additional comments that I am sure that the authors will be able to address.
Minor comments:
In their rebuttal the authors mention that the one of their objectives of their prioritization exercise was to ensure that cells with known Canada Warbler observations were prioritized over everything else so as to prevent foresters using our results from harvesting in areas with known individuals, which is illegal according to the Species At Risk Act. I think that should touch on this subject as It provide a solid rationale for their inclusion in the prioritization exercise.]
Good suggestion. We have updated section 2.3.1, data layers. The previous relevant line read “To determine suspected breeding locations, we divided point occurrences of Canada Warblers into two groups (CAWA presence 2005-2009 and CAWA presence 2010-2015) to capture areas showing persistent observations of Canada Warblers over time.”
We have updated this to read (lines 155-158): “Including recent presence information allows conservation planners to locate areas of high likelihood of extant Canada Warblers to consider for protection, and forest managers to avoid or operations in areas where they may harm, kill, or harass Canada Warbler or their nests or eggs illegal activities under Canada’s Species At Risk Act, S.C. 2002).”
[Figure numbering: Make sure that the Figure number in the text and in the figure caption match. With the current edit you seem to have two Figure 1.]
Oops! Figure numbering is fixed.
[Lines 208-210: There seems to be a word missing in the sentence “We chose CAZ because ensures that the most valuable […]”]
We added the missing ‘it’. Good catch!
[Lines 330-332: Table 3: The author sticked to their guns and insisted to present the mean cell-level difference in priority ranking but, as they pointed out in their rebuttal, the distribution of their ranking in bounded between 0 and 1 which make the interpretation of the mean difficult for the average reader. I think it would be clearer and more straightforward to present the change would make more sense to present the change in ranking in term of percentage like they did in their rebuttal.]
Okay, we have changed the table to express mean percent change and clarified the table title (below, and lines 331-334).
Table 3: Percent change in mean cell-level rankings across Zonation prioritization scenarios to support land conservation for the Canada Warbler when comparing between different estimated dispersal distances: 5 km (LDD), 10 km (MDDs), and 50 km (HDD). Scenario descriptions given in Table 1.
|
Scenario |
Dispersal estimates compared |
Mean percent change in priority ranking of a given cell between scenarios |
Standard deviation of percent change in priority ranking |
|
LC1 |
MDD-LDD |
3 |
3 |
|
LC1 |
HDD-MDD |
9 |
7 |
|
LC2 |
MDD-LDD |
4 |
4 |
|
LC2 |
HDD-MDD |
12 |
9 |
|
LC3 |
MDD-LDD |
3 |
4 |
|
LC3 |
HDD-MDD |
7 |
7 |
|
LC4 |
MDD-LDD |
4 |
5 |
|
LC4 |
HDD-MDD |
12 |
11 |
