- Email: [email protected]
Critical habitat designation under the US Endangered Species Act: How are biological criteria used?
B I O L O G I C A L C O N S E RVAT I O N
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/biocon
Short communication
Critical habitat designation under the US Endangered Species Act: How are biological criteria used? Karen E. Hodges*, Jason Elder Biology and Physical Geography, University of British Columbia Okanagan, 3333 University Way, Kelowna, BC, Canada V1V 1V7
A R T I C L E I N F O
A B S T R A C T
Article history:
Critical habitat designation under the US Endangered Species Act (ESA) is not working as
Received 4 March 2008
Congress intended. Issues include the use of science during designation, the costly and liti-
Received in revised form
gious delays in designation that have led to repeated lawsuits, and the potential overlap
1 July 2008
with other ESA protections. In this paper, we address a neglected aspect of critical habitat
Accepted 14 July 2008
designation: how the biologically-based designation criteria of the US Fish and Wildlife Ser-
Available online 28 August 2008
vice are used during the designation process. We primarily examine whether taxon (within terrestrial animals) or legal status (whether critical habitat was designated after a court
Keywords:
challenge) affect the use of the criteria. Court-ordered cases used more criteria than
Terrestrial animals
non-court-ordered cases. There were also differences in use of criteria with respect to
Lawsuits
taxon and region, and a weak relationship with the year of designation. Criteria that
Habitat
focused on discrete elements, such as nest sites or locations where required food species
United States
occurred, were used more often than criteria that addressed broader ecological needs such
Policy
as space for normal behaviour or representation of historic range conditions. Revising the critical habitat designation criteria and enforcing their consistent use during designation would be helpful for conservation of imperiled species in the United States. 2008 Elsevier Ltd. All rights reserved.
1.
Introduction
Many species at risk of extinction in the United States are declining because of habitat loss and degradation (Wilcove et al., 1998; Lawler et al., 2002). The US Endangered Species Act (ESA), one of the oldest such acts in the world, requires designation of critical habitat for listed species in recognition that protecting habitats is necessary to support recovery of imperiled species (Patlis, 2001). Similar provisions exist in Canada’s 2002 Species at Risk Act and Australia’s 1992 Endangered Species Protection Act, as well as in several state or provincial laws.
In the US, listed species with critical habitat appear to perform better than species without (Taylor et al., 2005; Suckling and Taylor, 2005). The history of the implementation of ESA has, however, been dominated by a low rate of designation of critical habitat (Hoekstra et al., 2002), despite the statutory requirement for it; two possible findings, ‘not prudent’ and ‘not determinable’ were intended by Congress to be rarely used but instead have been heavily used (Hagen and Hodges, 2006). Many lawsuits have been filed to counter this lack of designation; the courts have uniformly found that designation is non-discretionary and have imposed deadlines for
* Corresponding author: Tel.: +1 250 807 8763; fax: +1 250 807 8005. E-mail addresses: [email protected] (K.E. Hodges), [email protected] (J. Elder). 0006-3207/$ - see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2008.07.013
B I O L O G I C A L C O N S E RVAT I O N
designation. Since 1997, all critical habitat designation by the US Fish and Wildlife Service (FWS) has been as a result of these court-ordered decisions (Manson, 2003, 2004). FWS, the agency that is responsible for designation of critical habitat for terrestrial species, has repeatedly expressed a desire to develop a streamlined approach to critical habitat designation to reduce some of the current problems (USFWS, 1999; GAO, 2003). Previous analyses of the designation problems have focused primarily on process, timing, the required economic analyses, and cost (e.g. Patlis, 2001; Hoekstra et al., 2002). To date, there has been little analysis about how the biological criteria created by the FWS are used during designation. The FWS currently uses guidelines for designating critical habitat that were formulated in 1984, then updated in 2001 to apply to the National Marine Fisheries Service as well (Table 1). The 1984 guidelines replaced a set of 1975 guidelines that covered the same broad biological issues, albeit with different wording (Table 1). Our objective in this paper is to characterize how these biological designation criteria are used in practice by the FWS. We focus on four main factors that may impact use of the criteria: court-ordered status, taxon, region, and date of designation. Criteria use might differ with court-ordered status if these high-profile cases engender more attention to the designations. Taxa vary in their biology (e.g. habitat and dietary specificity, mobility, range size), as well as in how well known they are scientifically, so criteria use might vary with taxon. Finally, region and year of designation might affect use of criteria due to experience with designation or different FWS cultures in different places and times. ESA has also been amended several times, as well as subjected to substantial case law for its interpretation, so a temporal signature could reflect variation in interpretation through time. Regions also vary with respect to the number of listed species and the dominant threats facing listed species, which could lead to
2663
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
variation in use of criteria by region. We therefore focus our analysis on four descriptive questions: (1) Which of the FWS criteria are the major ones used for designation of critical habitat? (2) Are there differences in which criteria are emphasized for species that received critical habitat without court intervention and species for which critical habitat was court-ordered? (3) Are different criteria emphasized for different taxa? (4) Does the lead region or the year of designation affect how many criteria are used as critical habitat is designated?
2.
Methods
Our analysis is based on the terrestrial animals (mammals, birds, reptiles, amphibians, snails, insects, and arachnids) that had critical habitat as of December 2006. At that time, the FWS Threatened and Endangered Species System (TESS)(USFWS, 2006) webpage showed that 476 species had critical habitat out of 1311 listed domestic species (as of February 2008 that had climbed to 508 of 1351 species). Agency guidelines (USFWS, 1984, 2001) preclude designating critical habitat for foreign species. We considered distinct population segments as distinct records for analysis. We excluded aquatic species (marine mammals, sea turtles, fish, clams, and crustaceans) because aquatic and terrestrial habitat requirements are often quite different. We excluded plants because plants are not mobile, so locations of populations may be more easily identifiable than for mobile species, and because 85% of at-risk plants with critical habitat (245 species) are the result of a single court-ordered case in Hawaii (Conservation Council for Hawaii v. Babbitt, 1998), so analysis within this taxon
Table 1 – The US Fish and Wildlife Service criteria for designation of critical habitat under the Endangered Species Act 1984, 2001 Criteriaa 1 – Space for individual and population growth, and for normal behavior 2 – Food, water, air, light, minerals, or other nutritional or physiological requirements 3 – Cover or shelter 4 – Sites for breeding, reproduction, rearing of offspring, germination, or seed dispersal 5 – Habitats that are protected from disturbance or are representative of the historic geographical and ecological distributions of a species
1975 Criteriab (1) Space for normal growth, movements, or territorial behavior (2) Nutritional requirements, such as food, water, minerals (4) Cover or shelter (3) Sites for breeding, reproduction, or rearing of offspring (5) Other biological, physical, or behavioral requirements
Subcriteriac
Name used hereafter
No sub-category
Space
2a. Food 2b. Water 2c. Other physiological needs No sub-category 4a. Reproductive sites 4b. Protection from predators 4c. Special needs for juvenile habitat 4d. Special needs for adult habitat 5a. Protection from disturbance 5b. Representativeness of historical range
Resources
Cover Reproduction
Distribution
a Taken verbatim from the 1984 USFWS criteria (updated in 2001 to include NMFS). b Taken verbatim from the 1975 USFWS criteria; note that the numbering was slightly different than in 1984 and 2001. c We subdivided three of the criteria for more detailed analyses. Hereafter, we use ‘subcriteria’ to include the undivided criteria 1 and 3 and the subcriteria specified for the other criteria.
3.
Results
Very few cases used all five of the available designation criteria. On average, non-court-ordered cases used 2.4 ± 0.3 (mean ± 1 SE) criteria, while court-ordered cases used 4.2 ± 0.1 criteria (t = 6.23, p < 0.01). For any given criterion, on average 44% of the cases that were not court-ordered did not address the criterion during designation (Fig. 1A). For
(A) Major criteria 100
Not Court-ordered
**
**
Court-ordered
**
80
*
60 40 20 0 Space
100
Resources
Cover
Reproduction
(B) Subcriteria **
Distribution
Not Court-Ordered Court-Ordered
*
80
** **
60 **
**
40
** **
20
Sa fe ty Ju v. Ha bi ta Ad t ul tH ab it a t Di st ur ba Re nc e pr es en ta t io n
Ph ys io Re l pr o. Si te s
W at er
0 Fo od
would primarily represent this non-independent set of species. Our data set therefore consisted of 89 records: 47 cases for which critical habitat was court-ordered, and 42 where designation occurred without litigation. In this sample of species, 55 are Endangered, 31 are Threatened, and 3 bird species were delisted after critical habitat had been designated (one extinct, one recovered, and one judged not to be a distinct population). This distribution among the Threatened and Endangered risk categories is representative of all ESA-listed species within these taxa (v2 = 0.00, p = 0.98). Furthermore, similar proportions occurred in the court-ordered vs. not court-ordered groups (i.e. 36% and 33% of species were Threatened, respectively). We examined the Final Rule of each critical habitat designation to assess whether each criterion was identified as a driver of the final designation. Critical habitat designation is often a lengthy process and proposed critical habitat designations are published for public commentary; final rules include responses to salient comments. We did not examine the original documents listing a species as Threatened or Endangered (unless critical habitat was designated at the same time), nor the proposed critical habitat designations, because we expected the final designation documents would identify the criteria used for each designation. The FWS designation guidelines from 1975, 1984, and 2001 (Table 1) are similar enough in content that we considered all cases together rather than breaking the analysis by date (the 2001 guidelines were modified to include the National Marine Fisheries Service but are otherwise unchanged from 1984). We applied the 1984/2001 criteria retroactively to the 32 cases that had critical habitat assigned prior to 1984. Some of the five designation criteria could include distinct types of biological information, so we also subdivided three of the criteria into component parts (Table 1) so that we could analyze whether particular types of biological information were more heavily used than others. From the Final Rules, we also recorded whether human-induced habitat threats were identified, whether other species were included, and whether criteria that were not used were identified as a knowledge gap for the species. We used G-tests for 2 · 2 contingency tables, log-linear analysis for larger contingency tables, analysis of variance (ANOVA) for numbers of criteria or subcriteria used, and regression to test the effects of designation date. The five criteria or 11 subcriteria are non-independent response variables that were considered for each designation case, thus preventing contingency table analyses directly comparing use of the different criteria. We conducted statistical analyses with Statistica (StatSoft, 2006) and PopTools (Hood, 2005).
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
percentage using criterion
B I O L O G I C A L C O N S E RVAT I O N
percentage using criterion
2664
Fig. 1 – Frequency with which each criterion was used during critical habitat designation, with respect to whether designation was court-ordered. (A) The use of the five criteria in the FWS guidelines. (B) The use of subcriteria identified for criteria 2 (food, water, other physiological needs), 4 (reproductive sites, safety from predators, juvenile habitat, adult habitat), and 5 (disturbance, representation). Differences that were significant at the 0.05 level are marked with *, and at the 0.01 level with **. non-court-ordered cases, Reproduction was most likely to be used (79% of cases), and the Space criterion was least likely to be used (45%). For court-ordered cases, the Resources criterion was used the most (94%), and Distribution the least (72%). Reproduction was the only criterion that was used similarly between court-ordered and not court-ordered cases (83% and 79% of cases, respectively; G = 0.27, p = 0.60). The other four criteria were 1.4–1.9 times as likely to be used during designation for court-ordered cases than for non-court-ordered cases. Court-ordered cases were also significantly more likely to use each subcriterion than were non-court-ordered cases (Fig. 1B). The one exception to this pattern was the subcriterion of Safety from Predators, but even it had a trend towards greater use in court-ordered cases. Subcriteria were on average more than four times as likely (range 1.3–14.3 times) to be used during designation for court-ordered cases than for non-court-ordered cases. For both court-ordered and not court-ordered cases, Food and Reproductive Sites were the most-used subcriteria, and Other Physiological Requirements and Special Requirements for Adult Habitat were least likely to be used. The majority of designation documents identified humancaused habitat degradation or loss as a threat to the species (62% not court-ordered, 60% court-ordered, G = 0.049, p = 0.82). In terms of the designation documents themselves,
B I O L O G I C A L C O N S E RVAT I O N
Number of criteria Used
5
100
Not Court-Ordered Court-Ordered
2665
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
(A) Space
80
4
60
3 40
2
20
5
1 12
10
11
14
14
7
0
16
100
0
(B) Nutrients Birds
Reptiles & Amphibians
Invertebrates
Fig. 2 – Number of criteria used during critical habitat designation with respect to taxonomic group and whether designation was court-ordered. For each species, we counted how many of the five criteria were used during critical habitat designation. Means and standard errors are shown. There was no independent effect of taxon (F3,81 = 0.88, p = 0.45), but there was a significant impact of whether designation was court-ordered (F1,81 = 47.2, p < 0.01) and a significant interaction between taxon and legal status (F3,81 = 2.76, p = 0.047). Sample sizes are given in each bar.
just over a third of the species had designation documents that contained other species (40% not court-ordered, 36% court-ordered, G = 0.172, p = 0.68). Court-ordered cases were significantly less likely than non-court-ordered cases to have one or more of the criteria identified as a knowledge gap for the species (12% versus 32%, G = 5.18, p = 0.02). The number of criteria used during designation varied with taxon and whether designation was court-ordered (Fig. 2). Court-ordered cases considered 3.8–4.7 criteria per taxonomic group versus 2.0–3.3 for non-court-ordered cases. The significant interaction term between legal status and taxon arose because the disparity in number of criteria used within taxon between court-ordered and non-court-ordered cases was more severe for invertebrates (a 4 criteria difference) and reptiles plus amphibians (2 criteria difference) than for birds or mammals (1 criteria difference). We obtained similar results when we considered the eleven subcriteria: species that had critical habitat designated without court-order had 3.3 ± 0.3 subcriteria used during designation, whereas court-ordered designation cases considered 6.5 ± 0.3 subcriteria (taxon F3,81 = 0.61, p = 0.61, legal status F1,81 = 63.0, p < 0.01, interaction F3,81 = 3.10, p = 0.031). There were also some taxonomic differences with respect to which criteria were used during designation of critical habitat (Fig. 3). The Nutrients criterion varied among taxa especially for non-court-ordered cases, as the criterion was used in 100% of invertebrate cases, 73% of bird cases, 67% of mammal cases, and only 21% of reptile plus amphibian cases. Both Reproduction and Distribution showed trends towards a taxonomic pattern (Reproduction p = 0.084, Distribution p = 0.093). Reproduction was used slightly more often for mammals in non-court-ordered cases and less often for court-ordered invertebrates. The use of the Distribution criterion was highly variable for invertebrates, with no use during
80 60 40 20 0 100
(C) Cover 80
Percentage of Cases
Mammals
60 40 20 0
100
(D) Reproduction
80 60 40 20 0 100
(E) Distribution
80 60 40 20 0 Not court-ordered Mammals
Birds
Court-ordered Reptiles& Amphibians
Invertebrates
Fig. 3 – Use of the five primary criteria in relation to taxon and whether critical habitat was court-ordered. Panels A–E represent the 5 major criteria and each bar shows the percentage of cases that used each criterion during designation. The sample sizes within each panel are the same as in Fig. 2. We used log-linear analysis to test the impact of taxon (d.f. = 3) and legal status (d.f. = 1) on use of each criterion. (A) Space: taxon v2 = 1.68, p = 0.64, legal status v2 = 16.59, p < 0.01. (B) Nutrients: taxon v2 = 10.38, p = 0.016, legal status v2 = 10.21, p < 0.01. (C) Cover: taxon v2 = 3.64, p = 0.30, legal status v2 = 12.55, p < 0.01. (D) Reproduction: taxon v2 = 6.64, p = 0.084, legal status v2 = 1.16, p = 0.28. (E) Distribution: taxon v2 = 6.41, p = 0.093, legal status v2 = 4.62, p = 0.032.
2666
B I O L O G I C A L C O N S E RVAT I O N
Number of Criteria Used
5 4 3 2 1
Not court-ordered 0 1975
1980
1985
1990
1995
Court-ordered 2000
2005
Fig. 4 – The number of criteria used with respect to date and whether critical habitat designation was court-ordered. The smallest circles represent one case each; thereafter, the larger the circle, the more cases represented with increments of one case per size change. The only exception is that the very largest circle represents eight cases, and the second largest circle six cases. The multiple regression of date and legal status has r2 = 0.33; legal status t = 0.076, p = 0.94, date t = 2.18, p = 0.03. non-court-ordered cases and 94% for court-ordered cases. For Space and Cover, taxa were treated similarly. The number of criteria used increased slightly through time (Fig. 4), but the multiple regression of time and court-ordered status explained only 33% of the variance. Because date and court-ordered status strongly co-vary, we also ran separate regressions of year for court-ordered and non-court-ordered cases. For cases that were not court-ordered, there was a small but significant increase in the number of criteria used through time, but the r2 was a low 0.102. For the court-ordered cases, there was no significant trend through time (r2 = 0.002). We observed similar patterns when we examined the number of subcriteria used through time (data not shown). The seven FWS regions handled substantially different numbers of designations (Fig. 5). Region 1, which includes Cal-
Number of Criteria Used
5
Not Court-Ordered Court-Ordered
4 3 2 1 15
29
5
11
16
1
6
6
0 Region 1
Region 2
Region 4
Regions3,5,6,7
Fig. 5 – Number of criteria used by region with respect to whether critical habitat was court-ordered. Values are means ± 1 SE. Region 1 is the Pacific Region, including California and Hawaii. Region 2 is the Southwest, including Texas. Region Four is the Southeast, including Florida and Alabama. These five states contain by far the majority of all listed species, so these regions are shown separately. The final category includes the Great Lakes, Northeast, Mountain-Prairie, and Alaska regions. Sample sizes are given in each bar. Region was significant (F3,81 = 5.90, p = 0.001), as was court-ordered status (F1,81 = 16.2, p < 0.01); they did not interact significantly (F3,81 = 0.79, p = 0.50).
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
ifornia and Hawaii, had about half of the species (44 of 89). The southwest (Region 2) and southeast (Region 4) had another 33 of the species, leaving only 12 in the northeast, Alaska, the Rockies, and the prairies. Fewer criteria were used in the regions that handled the fewest species. For the 12 species handled by Regions 3, 5, 6, or 7, the average number of criteria used was 1.8 ± 0.2 for non-court-ordered and 3.0 ± 0.5 for court-ordered species; for regions 1, 2, and 4 combined, these values were 3.0 ± 0.2 and 4.4 ± 0.1, respectively. For subcriteria, results were similar: the conglomerate of Regions 3, 5, 6, and 7 used fewer subcriteria (2.0 ± 0.3 for court-ordered, 4.7 ± 0.7 not court-ordered) than the other regions (3.5 ± 0.3 and 6.8 ± 0.3, respectively; status F1,81 = 41.3, p < 0.01, region F3,81 = 4.95, p = 0.003, interaction F3,81 = 1.96, p = 0.13).
4.
Discussion
The biological criteria that have been in place since 1975 to guide critical habitat designation (USFWS, 1975, 1984, 2001) have been used unevenly for terrestrial animals. For the majority of species, some criteria were neither used for designation of critical habitat nor identified as knowledge gaps. Of the factors we examined, whether a species had court-ordered critical habitat had the most impact on the number of criteria used. Taxon affected both the number and identity of the criteria used. Region and date of designation had minor impacts on the number of criteria used. Easily identified aspects of habitat, such as Food and Reproductive sites, were much more likely to be used than were aspects such as Disturbance regime or Space for Normal Behaviour.
4.1.
Factors affecting use of designation criteria
Although the date of designation and whether it was court-ordered co-vary strongly, with all FWS designations since 1997 being court-ordered (Manson, 2003, 2004), it is likely that the variability in criteria use derives mainly from court-ordered status rather than date. In the litigious climate that has arisen around critical habitat designation (and other aspects of the ESA, e.g. listing and failure to consult as required), the FWS is sometimes being sued for the way in which critical habitat was designated (e.g. New Mexico Cattle Growers Association v. US Fish and Wildlife Service, 2004), the clarity with which it was designated (Home Builders Association of Northern California v. USFWS, 2003), as well as whether it was designated at all. These cases probably prompt more thorough and descriptive designation than would arise in the absence of such lawsuits. Similarly, the emphasis in the recent political climate on the use of defensible peer-reviewed science in policy activities (GAO, 2003) has probably reinforced increased attention to the designation guidelines during designation. Furthermore, in Home Builders Association of Northern California v. USFWS (2003), the court found that FWS needed to be much more explicit in its identification of what critical habitat consists of (for the Alameda whipsnake, Masticophis lateralis euryxanthus) and how that determination was reached, a finding that may have set a precedent for more explicit critical habitat documents.
B I O L O G I C A L C O N S E RVAT I O N
Taxon affected both how many and which criteria were used. Mammals and birds differed by only one criterion between court-ordered and non-court-ordered cases, but this disparity was higher for reptiles and amphibians (two criteria) and invertebrates (four criteria). This pattern may reflect knowledge about species, with better known groups showing less impact of court-ordered status. The two criteria with the biggest increase in use with court-ordered status are Space and Cover, followed by Distribution. All three of these criteria address habitat structure, spatial patterns, or behaviour, all of which may be poorly known relative to Nutrient Requirements and Reproduction. It thus appears that court-ordered status is having more impact on the use of the more challenging criteria. We suspect that the disparities in criteria use among taxa at least partially reflect genuine differences in biology. Critical habitat designation might work better for all species if criteria were tailored for specific taxa (or given more emphasis). For example, the ‘‘Space for normal behaviour’’ criterion is not particularly relevant for plants or for cave-dwelling invertebrates, but could be useful for wide-ranging carnivores like Canada lynx (Lynx canadensis) and wolves (Canis lupus). Despite a genuine lack of knowledge for some species, fewer court-ordered cases identified criteria for which data were inadequate than did non-court-ordered cases. We are not certain why this pattern has arisen. Fewer criteria were used during the very early cases and in regions that handle few species. Through the years, document complexity and length has noticeably increased, e.g. from the 3 page document (USFWS, 1976) that designated critical habitat for the American crocodile (Crocodylus acutus), California condor (Gymnogyps californianus), Indiana bat (Myotis sodalis), and Florida manatee (Trichechus manatus) to the 43 page designation (USFWS, 2004) for tiger salamanders (Ambystoma californiense) in one county in California. Although multi-species documents have occurred throughout the time span for both court-ordered and non-court-ordered species, the recent cases have been more likely to focus on species that are similar in their ecology, e.g. seven cave-dwelling invertebrates in Texas (USFWS, 2003) or four subspecies of fox on the California Channel Islands (USFWS, 2005). Through time, there has been a concomitant rise in the number of comments from the public and from the expert reviews requested by FWS as part of the designation process. Later documents are also much more likely to refer to the designation guidelines clearly than are earlier documents. These observations suggest that increasing experience has contributed to the increased use of criteria through time.
5.
Implications for species
We do not know to what extent the number of criteria used or the specific criteria used affect the size or location of the final designated critical habitat. Nor is it easy to relate the final area of designated critical habitat to criteria use, as taxon, historic range, current range, and the required economic analysis are also likely to affect the area designated. We do not necessarily expect an increase in area with increased use of criteria: for example, consideration of the historic Distribution could well subsume all four other criteria within it. In contrast, adding
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
2667
information from Reproduction and Space for Normal Behaviour could increase the designated area relative to consideration of either criterion in isolation. Consideration of more criteria may also affect the choice of specific locations contained within the designated critical habitat, e.g. if reproduction has been documented in some locations but not others. The Reproduction and Resources criteria were used more often than the other criteria. Our breakdown of these criteria into subcriteria shows that nest sites are the dominant aspect of reproductive sites considered (in contrast to facets such as specific habitat needed by juveniles or by adults, e.g. aquatic vs. terrestrial areas for juvenile and adult salamanders, or lek sites). The Resources criterion was often used to identify things like host plants or primary prey species. Because these two criteria frequently identify localized resources (specific food patches or substrates suitable for nests), rather than the larger areas required for traveling to locate mates, suitable protection of adults from predation, or habitats that are representative, emphasizing these two criteria may lead to delineation of smaller areas of critical habitat than are truly required for species’ recovery. It is troubling that many designation documents do not specify knowledge gaps for species, because identification of knowledge gaps could be useful for suggesting future work or guiding appropriate actions in the face of uncertainty. The existing documents varied in whether un-used criteria were simply not mentioned or were identified as knowledge gaps. This inconsistency may damage species’ recovery, especially since the un-used criteria tended to be those that addressed larger-scale spatial patterns (Space, Cover, Distribution).
6.
Recommendations and conclusions
Over-all, the existing criteria for critical habitat designation are inconsistently applied. Although the biggest separation in usage arose between court-ordered and non-court-ordered cases, we cannot dismiss a possible temporal signature affecting this result. Regardless, the high variability in how designation is conducted suggests that the existing criteria are not functioning as intended, at least for terrestrial animals. This conclusion echoes findings from the GAO (2003), the National Research Council (1995), and the FWS itself (1999) that the process for designating critical habitat should be redesigned. As a minimal first step at improving the process, we think the FWS should require a more consistent document format that makes it transparently clear for each criterion whether it was used, and, if so, how. Requiring a clearer delineation with respect to each criterion would make it easier to assess how designation was done and would also point out key areas for additional research. Even the court-ordered cases are not doing particularly well at consistently using – or at least considering and dismissing – all criteria during designation. It is possible that criteria are considered and dismissed during the process without documentation in the Final Rule for each case. We have no basis for assessing how prevalent such a practice might be; if it is common, then the lack of documentation in the final designations is problematic because it is not pub-
2668
B I O L O G I C A L C O N S E RVAT I O N
licly transparent how decisions are reached. Alternatively, if most Final Rules do summarize what was evaluated during the designation process, then it appears that most designations use only some of the available criteria. We think that the patterns we have observed here indicate a deeper malady, namely that it is difficult to use the existing criteria to arrive at delineated critical habitat. Although the criteria cover a range of biological considerations, they still are at a high level of generality that may make it hard in practice to know how to use a set of data to determine critical habitat. We encourage FWS and others to revisit whether these criteria are truly functional tools in the process of designating critical habitat for imperiled species.
Acknowledgements S. Pascuzzo helped to populate the database used in this paper. Funding was provided by an internal research grant from the University of British Columbia Okanagan to K.E.H.
R E F E R E N C E S
Conservation Council for Hawaii v. Babbitt. 1998. 2 F. Supp. 2d 1280, Number 97-00098 (US District Court). GAO (US General Accounting Office), 2003. Endangered Species: Fish and Wildlife Service uses best available science to make listing decisions, but additional guidance needed for critical habitat designations. GAO-03-803. US Government Printing Office, Washington, D.C., USA. Hagen, A.N., Hodges, K.E., 2006. Resolving critical habitat designation failures: reconciling law, policy, and biology. Conservation Biology 20, 399–407. Hoekstra, J.M., Fagan, W.F., Bradley, J.E., 2002. A critical role for critical habitat in the recovery planning process? Not yet. Ecological Applications 12, 701–707. Hood, G.M., 2005. PopTools version 2.7.1. Available at. Home Builders Association of Northern California v. USFWS, 2003. CV F01-5722 (US District Court). Lawler, J.J., Campbell, S.P., Guerry, A.D., Kolozsvary, M.B., O’Connor, R.J., Seward, L.C.N., 2002. The scope and treatment of threats in endangered species recovery plans. Ecological Applications 12, 663–667. Manson, C., 2003. Testimony before the subcommittee on Fisheries, Wildlife, and Water of the Senate Committee on Environment and Public Works. 108 Senate hearings. Endangered Species Act: critical habitat issues (10 April 2003). Manson, C., 2004. Testimony before the House Resources Committee. 108 House Hearings. Regarding H.R. 2933, the critical habitat reform bill act of 2003 (28 April 2004).
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
New Mexico Cattle Growers Association v. US Fish and Wildlife Service, 2004. CIV 02-0199 (US District Court). NRC (National Research Council), 1995. Science and the Endangered Species Act. National Academy Press, Washington, D.C. Patlis, J.M., 2001. Paying tribute to Joseph Heller with the Endangered Species Act: when critical habitat isn’t. Stanford Environmental Law Journal 20, 133–217. StatSoft, 2006. Statistica, version 7.1. Tulsa, Oklahoma, USA. Suckling, K., Taylor, M., 2005. Critical habitat and recovery. In: Goble, D.D., Scott, J.M., Davis, F.W. (Eds.), The Endangered Species Act at 30: Renewing the Conservation Promise. Island Press, Washington, D.C, pp. 75–89. Taylor, M., Suckling, K., Rachlinski, J., 2005. The effectiveness of the Endangered Species Act: a quantitative analysis. BioScience 55, 360–368. USFWS (US Fish and Wildlife Service), 1975. Endangered and threatened species: notice on critical habitat areas. Federal Register 40: 17764–17765. USFWS (US Fish and Wildlife Service), 1976. Determination of Critical Habitat for American Crocodile, California Condor, Indiana Bat, and Florida Manatee. Federal Register 41: 41914– 41916. USFWS (US Fish and Wildlife Service), 1984. Listing endangered and threatened species and designating critical habitat. Federal Register 49:38908–38912. USFWS (US Fish and Wildlife Service), 1999. Notice of intent to clarify the role of habitat in endangered species conservation. Federal Register 64:31871–31874. USFWS (US Fish and Wildlife Service), 2001. Joint regulations (United States Fish and Wildlife Service, Department of the Interior and National Marine Fisheries Service, National Oceanic and Atmospheric Administration, Department of Commerce); Endangered species committee regulations, listing endangered and threatened species and designating critical habitat. 50 Code of Federal Regulations 424.12: 405– 406. USFWS (US Fish and Wildlife Service), 2003. Endangered and Threatened Wildlife and Plants; Designation of Critical Habitat for Seven Bexar County, TX, Invertebrate Species; Final Rule. Federal Register 68: 17155–17231. USFWS (US Fish and Wildlife Service), 2004. Endangered and threatened wildlife and plants; designation of critical habitat for the California tiger salamander (Ambystoma californiense) in Santa Barbara County, final rule. Federal Register 69: 68568– 68609. USFWS (US Fish and Wildlife Service), 2005. Final Determination Concerning Critical Habitat for the San Miguel Island Fox, Santa Rosa Island Fox, Santa Cruz Island Fox, and Santa Catalina Island Fox. Federal Register 70: 67924–67929. USFWS (US Fish and Wildlife Service), 2006. Threatened and endangered species system (TESS): species with critical habitat.. Wilcove, D.S., Rothstein, D., Dubow, J., Phillips, A., Losos, E., 1998. Quantifying threats to imperiled species in the United States. BioScience 48, 607–615.
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/biocon
Short communication
Critical habitat designation under the US Endangered Species Act: How are biological criteria used? Karen E. Hodges*, Jason Elder Biology and Physical Geography, University of British Columbia Okanagan, 3333 University Way, Kelowna, BC, Canada V1V 1V7
A R T I C L E I N F O
A B S T R A C T
Article history:
Critical habitat designation under the US Endangered Species Act (ESA) is not working as
Received 4 March 2008
Congress intended. Issues include the use of science during designation, the costly and liti-
Received in revised form
gious delays in designation that have led to repeated lawsuits, and the potential overlap
1 July 2008
with other ESA protections. In this paper, we address a neglected aspect of critical habitat
Accepted 14 July 2008
designation: how the biologically-based designation criteria of the US Fish and Wildlife Ser-
Available online 28 August 2008
vice are used during the designation process. We primarily examine whether taxon (within terrestrial animals) or legal status (whether critical habitat was designated after a court
Keywords:
challenge) affect the use of the criteria. Court-ordered cases used more criteria than
Terrestrial animals
non-court-ordered cases. There were also differences in use of criteria with respect to
Lawsuits
taxon and region, and a weak relationship with the year of designation. Criteria that
Habitat
focused on discrete elements, such as nest sites or locations where required food species
United States
occurred, were used more often than criteria that addressed broader ecological needs such
Policy
as space for normal behaviour or representation of historic range conditions. Revising the critical habitat designation criteria and enforcing their consistent use during designation would be helpful for conservation of imperiled species in the United States. 2008 Elsevier Ltd. All rights reserved.
1.
Introduction
Many species at risk of extinction in the United States are declining because of habitat loss and degradation (Wilcove et al., 1998; Lawler et al., 2002). The US Endangered Species Act (ESA), one of the oldest such acts in the world, requires designation of critical habitat for listed species in recognition that protecting habitats is necessary to support recovery of imperiled species (Patlis, 2001). Similar provisions exist in Canada’s 2002 Species at Risk Act and Australia’s 1992 Endangered Species Protection Act, as well as in several state or provincial laws.
In the US, listed species with critical habitat appear to perform better than species without (Taylor et al., 2005; Suckling and Taylor, 2005). The history of the implementation of ESA has, however, been dominated by a low rate of designation of critical habitat (Hoekstra et al., 2002), despite the statutory requirement for it; two possible findings, ‘not prudent’ and ‘not determinable’ were intended by Congress to be rarely used but instead have been heavily used (Hagen and Hodges, 2006). Many lawsuits have been filed to counter this lack of designation; the courts have uniformly found that designation is non-discretionary and have imposed deadlines for
* Corresponding author: Tel.: +1 250 807 8763; fax: +1 250 807 8005. E-mail addresses: [email protected] (K.E. Hodges), [email protected] (J. Elder). 0006-3207/$ - see front matter 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.biocon.2008.07.013
B I O L O G I C A L C O N S E RVAT I O N
designation. Since 1997, all critical habitat designation by the US Fish and Wildlife Service (FWS) has been as a result of these court-ordered decisions (Manson, 2003, 2004). FWS, the agency that is responsible for designation of critical habitat for terrestrial species, has repeatedly expressed a desire to develop a streamlined approach to critical habitat designation to reduce some of the current problems (USFWS, 1999; GAO, 2003). Previous analyses of the designation problems have focused primarily on process, timing, the required economic analyses, and cost (e.g. Patlis, 2001; Hoekstra et al., 2002). To date, there has been little analysis about how the biological criteria created by the FWS are used during designation. The FWS currently uses guidelines for designating critical habitat that were formulated in 1984, then updated in 2001 to apply to the National Marine Fisheries Service as well (Table 1). The 1984 guidelines replaced a set of 1975 guidelines that covered the same broad biological issues, albeit with different wording (Table 1). Our objective in this paper is to characterize how these biological designation criteria are used in practice by the FWS. We focus on four main factors that may impact use of the criteria: court-ordered status, taxon, region, and date of designation. Criteria use might differ with court-ordered status if these high-profile cases engender more attention to the designations. Taxa vary in their biology (e.g. habitat and dietary specificity, mobility, range size), as well as in how well known they are scientifically, so criteria use might vary with taxon. Finally, region and year of designation might affect use of criteria due to experience with designation or different FWS cultures in different places and times. ESA has also been amended several times, as well as subjected to substantial case law for its interpretation, so a temporal signature could reflect variation in interpretation through time. Regions also vary with respect to the number of listed species and the dominant threats facing listed species, which could lead to
2663
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
variation in use of criteria by region. We therefore focus our analysis on four descriptive questions: (1) Which of the FWS criteria are the major ones used for designation of critical habitat? (2) Are there differences in which criteria are emphasized for species that received critical habitat without court intervention and species for which critical habitat was court-ordered? (3) Are different criteria emphasized for different taxa? (4) Does the lead region or the year of designation affect how many criteria are used as critical habitat is designated?
2.
Methods
Our analysis is based on the terrestrial animals (mammals, birds, reptiles, amphibians, snails, insects, and arachnids) that had critical habitat as of December 2006. At that time, the FWS Threatened and Endangered Species System (TESS)(USFWS, 2006) webpage showed that 476 species had critical habitat out of 1311 listed domestic species (as of February 2008 that had climbed to 508 of 1351 species). Agency guidelines (USFWS, 1984, 2001) preclude designating critical habitat for foreign species. We considered distinct population segments as distinct records for analysis. We excluded aquatic species (marine mammals, sea turtles, fish, clams, and crustaceans) because aquatic and terrestrial habitat requirements are often quite different. We excluded plants because plants are not mobile, so locations of populations may be more easily identifiable than for mobile species, and because 85% of at-risk plants with critical habitat (245 species) are the result of a single court-ordered case in Hawaii (Conservation Council for Hawaii v. Babbitt, 1998), so analysis within this taxon
Table 1 – The US Fish and Wildlife Service criteria for designation of critical habitat under the Endangered Species Act 1984, 2001 Criteriaa 1 – Space for individual and population growth, and for normal behavior 2 – Food, water, air, light, minerals, or other nutritional or physiological requirements 3 – Cover or shelter 4 – Sites for breeding, reproduction, rearing of offspring, germination, or seed dispersal 5 – Habitats that are protected from disturbance or are representative of the historic geographical and ecological distributions of a species
1975 Criteriab (1) Space for normal growth, movements, or territorial behavior (2) Nutritional requirements, such as food, water, minerals (4) Cover or shelter (3) Sites for breeding, reproduction, or rearing of offspring (5) Other biological, physical, or behavioral requirements
Subcriteriac
Name used hereafter
No sub-category
Space
2a. Food 2b. Water 2c. Other physiological needs No sub-category 4a. Reproductive sites 4b. Protection from predators 4c. Special needs for juvenile habitat 4d. Special needs for adult habitat 5a. Protection from disturbance 5b. Representativeness of historical range
Resources
Cover Reproduction
Distribution
a Taken verbatim from the 1984 USFWS criteria (updated in 2001 to include NMFS). b Taken verbatim from the 1975 USFWS criteria; note that the numbering was slightly different than in 1984 and 2001. c We subdivided three of the criteria for more detailed analyses. Hereafter, we use ‘subcriteria’ to include the undivided criteria 1 and 3 and the subcriteria specified for the other criteria.
3.
Results
Very few cases used all five of the available designation criteria. On average, non-court-ordered cases used 2.4 ± 0.3 (mean ± 1 SE) criteria, while court-ordered cases used 4.2 ± 0.1 criteria (t = 6.23, p < 0.01). For any given criterion, on average 44% of the cases that were not court-ordered did not address the criterion during designation (Fig. 1A). For
(A) Major criteria 100
Not Court-ordered
**
**
Court-ordered
**
80
*
60 40 20 0 Space
100
Resources
Cover
Reproduction
(B) Subcriteria **
Distribution
Not Court-Ordered Court-Ordered
*
80
** **
60 **
**
40
** **
20
Sa fe ty Ju v. Ha bi ta Ad t ul tH ab it a t Di st ur ba Re nc e pr es en ta t io n
Ph ys io Re l pr o. Si te s
W at er
0 Fo od
would primarily represent this non-independent set of species. Our data set therefore consisted of 89 records: 47 cases for which critical habitat was court-ordered, and 42 where designation occurred without litigation. In this sample of species, 55 are Endangered, 31 are Threatened, and 3 bird species were delisted after critical habitat had been designated (one extinct, one recovered, and one judged not to be a distinct population). This distribution among the Threatened and Endangered risk categories is representative of all ESA-listed species within these taxa (v2 = 0.00, p = 0.98). Furthermore, similar proportions occurred in the court-ordered vs. not court-ordered groups (i.e. 36% and 33% of species were Threatened, respectively). We examined the Final Rule of each critical habitat designation to assess whether each criterion was identified as a driver of the final designation. Critical habitat designation is often a lengthy process and proposed critical habitat designations are published for public commentary; final rules include responses to salient comments. We did not examine the original documents listing a species as Threatened or Endangered (unless critical habitat was designated at the same time), nor the proposed critical habitat designations, because we expected the final designation documents would identify the criteria used for each designation. The FWS designation guidelines from 1975, 1984, and 2001 (Table 1) are similar enough in content that we considered all cases together rather than breaking the analysis by date (the 2001 guidelines were modified to include the National Marine Fisheries Service but are otherwise unchanged from 1984). We applied the 1984/2001 criteria retroactively to the 32 cases that had critical habitat assigned prior to 1984. Some of the five designation criteria could include distinct types of biological information, so we also subdivided three of the criteria into component parts (Table 1) so that we could analyze whether particular types of biological information were more heavily used than others. From the Final Rules, we also recorded whether human-induced habitat threats were identified, whether other species were included, and whether criteria that were not used were identified as a knowledge gap for the species. We used G-tests for 2 · 2 contingency tables, log-linear analysis for larger contingency tables, analysis of variance (ANOVA) for numbers of criteria or subcriteria used, and regression to test the effects of designation date. The five criteria or 11 subcriteria are non-independent response variables that were considered for each designation case, thus preventing contingency table analyses directly comparing use of the different criteria. We conducted statistical analyses with Statistica (StatSoft, 2006) and PopTools (Hood, 2005).
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
percentage using criterion
B I O L O G I C A L C O N S E RVAT I O N
percentage using criterion
2664
Fig. 1 – Frequency with which each criterion was used during critical habitat designation, with respect to whether designation was court-ordered. (A) The use of the five criteria in the FWS guidelines. (B) The use of subcriteria identified for criteria 2 (food, water, other physiological needs), 4 (reproductive sites, safety from predators, juvenile habitat, adult habitat), and 5 (disturbance, representation). Differences that were significant at the 0.05 level are marked with *, and at the 0.01 level with **. non-court-ordered cases, Reproduction was most likely to be used (79% of cases), and the Space criterion was least likely to be used (45%). For court-ordered cases, the Resources criterion was used the most (94%), and Distribution the least (72%). Reproduction was the only criterion that was used similarly between court-ordered and not court-ordered cases (83% and 79% of cases, respectively; G = 0.27, p = 0.60). The other four criteria were 1.4–1.9 times as likely to be used during designation for court-ordered cases than for non-court-ordered cases. Court-ordered cases were also significantly more likely to use each subcriterion than were non-court-ordered cases (Fig. 1B). The one exception to this pattern was the subcriterion of Safety from Predators, but even it had a trend towards greater use in court-ordered cases. Subcriteria were on average more than four times as likely (range 1.3–14.3 times) to be used during designation for court-ordered cases than for non-court-ordered cases. For both court-ordered and not court-ordered cases, Food and Reproductive Sites were the most-used subcriteria, and Other Physiological Requirements and Special Requirements for Adult Habitat were least likely to be used. The majority of designation documents identified humancaused habitat degradation or loss as a threat to the species (62% not court-ordered, 60% court-ordered, G = 0.049, p = 0.82). In terms of the designation documents themselves,
B I O L O G I C A L C O N S E RVAT I O N
Number of criteria Used
5
100
Not Court-Ordered Court-Ordered
2665
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
(A) Space
80
4
60
3 40
2
20
5
1 12
10
11
14
14
7
0
16
100
0
(B) Nutrients Birds
Reptiles & Amphibians
Invertebrates
Fig. 2 – Number of criteria used during critical habitat designation with respect to taxonomic group and whether designation was court-ordered. For each species, we counted how many of the five criteria were used during critical habitat designation. Means and standard errors are shown. There was no independent effect of taxon (F3,81 = 0.88, p = 0.45), but there was a significant impact of whether designation was court-ordered (F1,81 = 47.2, p < 0.01) and a significant interaction between taxon and legal status (F3,81 = 2.76, p = 0.047). Sample sizes are given in each bar.
just over a third of the species had designation documents that contained other species (40% not court-ordered, 36% court-ordered, G = 0.172, p = 0.68). Court-ordered cases were significantly less likely than non-court-ordered cases to have one or more of the criteria identified as a knowledge gap for the species (12% versus 32%, G = 5.18, p = 0.02). The number of criteria used during designation varied with taxon and whether designation was court-ordered (Fig. 2). Court-ordered cases considered 3.8–4.7 criteria per taxonomic group versus 2.0–3.3 for non-court-ordered cases. The significant interaction term between legal status and taxon arose because the disparity in number of criteria used within taxon between court-ordered and non-court-ordered cases was more severe for invertebrates (a 4 criteria difference) and reptiles plus amphibians (2 criteria difference) than for birds or mammals (1 criteria difference). We obtained similar results when we considered the eleven subcriteria: species that had critical habitat designated without court-order had 3.3 ± 0.3 subcriteria used during designation, whereas court-ordered designation cases considered 6.5 ± 0.3 subcriteria (taxon F3,81 = 0.61, p = 0.61, legal status F1,81 = 63.0, p < 0.01, interaction F3,81 = 3.10, p = 0.031). There were also some taxonomic differences with respect to which criteria were used during designation of critical habitat (Fig. 3). The Nutrients criterion varied among taxa especially for non-court-ordered cases, as the criterion was used in 100% of invertebrate cases, 73% of bird cases, 67% of mammal cases, and only 21% of reptile plus amphibian cases. Both Reproduction and Distribution showed trends towards a taxonomic pattern (Reproduction p = 0.084, Distribution p = 0.093). Reproduction was used slightly more often for mammals in non-court-ordered cases and less often for court-ordered invertebrates. The use of the Distribution criterion was highly variable for invertebrates, with no use during
80 60 40 20 0 100
(C) Cover 80
Percentage of Cases
Mammals
60 40 20 0
100
(D) Reproduction
80 60 40 20 0 100
(E) Distribution
80 60 40 20 0 Not court-ordered Mammals
Birds
Court-ordered Reptiles& Amphibians
Invertebrates
Fig. 3 – Use of the five primary criteria in relation to taxon and whether critical habitat was court-ordered. Panels A–E represent the 5 major criteria and each bar shows the percentage of cases that used each criterion during designation. The sample sizes within each panel are the same as in Fig. 2. We used log-linear analysis to test the impact of taxon (d.f. = 3) and legal status (d.f. = 1) on use of each criterion. (A) Space: taxon v2 = 1.68, p = 0.64, legal status v2 = 16.59, p < 0.01. (B) Nutrients: taxon v2 = 10.38, p = 0.016, legal status v2 = 10.21, p < 0.01. (C) Cover: taxon v2 = 3.64, p = 0.30, legal status v2 = 12.55, p < 0.01. (D) Reproduction: taxon v2 = 6.64, p = 0.084, legal status v2 = 1.16, p = 0.28. (E) Distribution: taxon v2 = 6.41, p = 0.093, legal status v2 = 4.62, p = 0.032.
2666
B I O L O G I C A L C O N S E RVAT I O N
Number of Criteria Used
5 4 3 2 1
Not court-ordered 0 1975
1980
1985
1990
1995
Court-ordered 2000
2005
Fig. 4 – The number of criteria used with respect to date and whether critical habitat designation was court-ordered. The smallest circles represent one case each; thereafter, the larger the circle, the more cases represented with increments of one case per size change. The only exception is that the very largest circle represents eight cases, and the second largest circle six cases. The multiple regression of date and legal status has r2 = 0.33; legal status t = 0.076, p = 0.94, date t = 2.18, p = 0.03. non-court-ordered cases and 94% for court-ordered cases. For Space and Cover, taxa were treated similarly. The number of criteria used increased slightly through time (Fig. 4), but the multiple regression of time and court-ordered status explained only 33% of the variance. Because date and court-ordered status strongly co-vary, we also ran separate regressions of year for court-ordered and non-court-ordered cases. For cases that were not court-ordered, there was a small but significant increase in the number of criteria used through time, but the r2 was a low 0.102. For the court-ordered cases, there was no significant trend through time (r2 = 0.002). We observed similar patterns when we examined the number of subcriteria used through time (data not shown). The seven FWS regions handled substantially different numbers of designations (Fig. 5). Region 1, which includes Cal-
Number of Criteria Used
5
Not Court-Ordered Court-Ordered
4 3 2 1 15
29
5
11
16
1
6
6
0 Region 1
Region 2
Region 4
Regions3,5,6,7
Fig. 5 – Number of criteria used by region with respect to whether critical habitat was court-ordered. Values are means ± 1 SE. Region 1 is the Pacific Region, including California and Hawaii. Region 2 is the Southwest, including Texas. Region Four is the Southeast, including Florida and Alabama. These five states contain by far the majority of all listed species, so these regions are shown separately. The final category includes the Great Lakes, Northeast, Mountain-Prairie, and Alaska regions. Sample sizes are given in each bar. Region was significant (F3,81 = 5.90, p = 0.001), as was court-ordered status (F1,81 = 16.2, p < 0.01); they did not interact significantly (F3,81 = 0.79, p = 0.50).
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
ifornia and Hawaii, had about half of the species (44 of 89). The southwest (Region 2) and southeast (Region 4) had another 33 of the species, leaving only 12 in the northeast, Alaska, the Rockies, and the prairies. Fewer criteria were used in the regions that handled the fewest species. For the 12 species handled by Regions 3, 5, 6, or 7, the average number of criteria used was 1.8 ± 0.2 for non-court-ordered and 3.0 ± 0.5 for court-ordered species; for regions 1, 2, and 4 combined, these values were 3.0 ± 0.2 and 4.4 ± 0.1, respectively. For subcriteria, results were similar: the conglomerate of Regions 3, 5, 6, and 7 used fewer subcriteria (2.0 ± 0.3 for court-ordered, 4.7 ± 0.7 not court-ordered) than the other regions (3.5 ± 0.3 and 6.8 ± 0.3, respectively; status F1,81 = 41.3, p < 0.01, region F3,81 = 4.95, p = 0.003, interaction F3,81 = 1.96, p = 0.13).
4.
Discussion
The biological criteria that have been in place since 1975 to guide critical habitat designation (USFWS, 1975, 1984, 2001) have been used unevenly for terrestrial animals. For the majority of species, some criteria were neither used for designation of critical habitat nor identified as knowledge gaps. Of the factors we examined, whether a species had court-ordered critical habitat had the most impact on the number of criteria used. Taxon affected both the number and identity of the criteria used. Region and date of designation had minor impacts on the number of criteria used. Easily identified aspects of habitat, such as Food and Reproductive sites, were much more likely to be used than were aspects such as Disturbance regime or Space for Normal Behaviour.
4.1.
Factors affecting use of designation criteria
Although the date of designation and whether it was court-ordered co-vary strongly, with all FWS designations since 1997 being court-ordered (Manson, 2003, 2004), it is likely that the variability in criteria use derives mainly from court-ordered status rather than date. In the litigious climate that has arisen around critical habitat designation (and other aspects of the ESA, e.g. listing and failure to consult as required), the FWS is sometimes being sued for the way in which critical habitat was designated (e.g. New Mexico Cattle Growers Association v. US Fish and Wildlife Service, 2004), the clarity with which it was designated (Home Builders Association of Northern California v. USFWS, 2003), as well as whether it was designated at all. These cases probably prompt more thorough and descriptive designation than would arise in the absence of such lawsuits. Similarly, the emphasis in the recent political climate on the use of defensible peer-reviewed science in policy activities (GAO, 2003) has probably reinforced increased attention to the designation guidelines during designation. Furthermore, in Home Builders Association of Northern California v. USFWS (2003), the court found that FWS needed to be much more explicit in its identification of what critical habitat consists of (for the Alameda whipsnake, Masticophis lateralis euryxanthus) and how that determination was reached, a finding that may have set a precedent for more explicit critical habitat documents.
B I O L O G I C A L C O N S E RVAT I O N
Taxon affected both how many and which criteria were used. Mammals and birds differed by only one criterion between court-ordered and non-court-ordered cases, but this disparity was higher for reptiles and amphibians (two criteria) and invertebrates (four criteria). This pattern may reflect knowledge about species, with better known groups showing less impact of court-ordered status. The two criteria with the biggest increase in use with court-ordered status are Space and Cover, followed by Distribution. All three of these criteria address habitat structure, spatial patterns, or behaviour, all of which may be poorly known relative to Nutrient Requirements and Reproduction. It thus appears that court-ordered status is having more impact on the use of the more challenging criteria. We suspect that the disparities in criteria use among taxa at least partially reflect genuine differences in biology. Critical habitat designation might work better for all species if criteria were tailored for specific taxa (or given more emphasis). For example, the ‘‘Space for normal behaviour’’ criterion is not particularly relevant for plants or for cave-dwelling invertebrates, but could be useful for wide-ranging carnivores like Canada lynx (Lynx canadensis) and wolves (Canis lupus). Despite a genuine lack of knowledge for some species, fewer court-ordered cases identified criteria for which data were inadequate than did non-court-ordered cases. We are not certain why this pattern has arisen. Fewer criteria were used during the very early cases and in regions that handle few species. Through the years, document complexity and length has noticeably increased, e.g. from the 3 page document (USFWS, 1976) that designated critical habitat for the American crocodile (Crocodylus acutus), California condor (Gymnogyps californianus), Indiana bat (Myotis sodalis), and Florida manatee (Trichechus manatus) to the 43 page designation (USFWS, 2004) for tiger salamanders (Ambystoma californiense) in one county in California. Although multi-species documents have occurred throughout the time span for both court-ordered and non-court-ordered species, the recent cases have been more likely to focus on species that are similar in their ecology, e.g. seven cave-dwelling invertebrates in Texas (USFWS, 2003) or four subspecies of fox on the California Channel Islands (USFWS, 2005). Through time, there has been a concomitant rise in the number of comments from the public and from the expert reviews requested by FWS as part of the designation process. Later documents are also much more likely to refer to the designation guidelines clearly than are earlier documents. These observations suggest that increasing experience has contributed to the increased use of criteria through time.
5.
Implications for species
We do not know to what extent the number of criteria used or the specific criteria used affect the size or location of the final designated critical habitat. Nor is it easy to relate the final area of designated critical habitat to criteria use, as taxon, historic range, current range, and the required economic analysis are also likely to affect the area designated. We do not necessarily expect an increase in area with increased use of criteria: for example, consideration of the historic Distribution could well subsume all four other criteria within it. In contrast, adding
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
2667
information from Reproduction and Space for Normal Behaviour could increase the designated area relative to consideration of either criterion in isolation. Consideration of more criteria may also affect the choice of specific locations contained within the designated critical habitat, e.g. if reproduction has been documented in some locations but not others. The Reproduction and Resources criteria were used more often than the other criteria. Our breakdown of these criteria into subcriteria shows that nest sites are the dominant aspect of reproductive sites considered (in contrast to facets such as specific habitat needed by juveniles or by adults, e.g. aquatic vs. terrestrial areas for juvenile and adult salamanders, or lek sites). The Resources criterion was often used to identify things like host plants or primary prey species. Because these two criteria frequently identify localized resources (specific food patches or substrates suitable for nests), rather than the larger areas required for traveling to locate mates, suitable protection of adults from predation, or habitats that are representative, emphasizing these two criteria may lead to delineation of smaller areas of critical habitat than are truly required for species’ recovery. It is troubling that many designation documents do not specify knowledge gaps for species, because identification of knowledge gaps could be useful for suggesting future work or guiding appropriate actions in the face of uncertainty. The existing documents varied in whether un-used criteria were simply not mentioned or were identified as knowledge gaps. This inconsistency may damage species’ recovery, especially since the un-used criteria tended to be those that addressed larger-scale spatial patterns (Space, Cover, Distribution).
6.
Recommendations and conclusions
Over-all, the existing criteria for critical habitat designation are inconsistently applied. Although the biggest separation in usage arose between court-ordered and non-court-ordered cases, we cannot dismiss a possible temporal signature affecting this result. Regardless, the high variability in how designation is conducted suggests that the existing criteria are not functioning as intended, at least for terrestrial animals. This conclusion echoes findings from the GAO (2003), the National Research Council (1995), and the FWS itself (1999) that the process for designating critical habitat should be redesigned. As a minimal first step at improving the process, we think the FWS should require a more consistent document format that makes it transparently clear for each criterion whether it was used, and, if so, how. Requiring a clearer delineation with respect to each criterion would make it easier to assess how designation was done and would also point out key areas for additional research. Even the court-ordered cases are not doing particularly well at consistently using – or at least considering and dismissing – all criteria during designation. It is possible that criteria are considered and dismissed during the process without documentation in the Final Rule for each case. We have no basis for assessing how prevalent such a practice might be; if it is common, then the lack of documentation in the final designations is problematic because it is not pub-
2668
B I O L O G I C A L C O N S E RVAT I O N
licly transparent how decisions are reached. Alternatively, if most Final Rules do summarize what was evaluated during the designation process, then it appears that most designations use only some of the available criteria. We think that the patterns we have observed here indicate a deeper malady, namely that it is difficult to use the existing criteria to arrive at delineated critical habitat. Although the criteria cover a range of biological considerations, they still are at a high level of generality that may make it hard in practice to know how to use a set of data to determine critical habitat. We encourage FWS and others to revisit whether these criteria are truly functional tools in the process of designating critical habitat for imperiled species.
Acknowledgements S. Pascuzzo helped to populate the database used in this paper. Funding was provided by an internal research grant from the University of British Columbia Okanagan to K.E.H.
R E F E R E N C E S
Conservation Council for Hawaii v. Babbitt. 1998. 2 F. Supp. 2d 1280, Number 97-00098 (US District Court). GAO (US General Accounting Office), 2003. Endangered Species: Fish and Wildlife Service uses best available science to make listing decisions, but additional guidance needed for critical habitat designations. GAO-03-803. US Government Printing Office, Washington, D.C., USA. Hagen, A.N., Hodges, K.E., 2006. Resolving critical habitat designation failures: reconciling law, policy, and biology. Conservation Biology 20, 399–407. Hoekstra, J.M., Fagan, W.F., Bradley, J.E., 2002. A critical role for critical habitat in the recovery planning process? Not yet. Ecological Applications 12, 701–707. Hood, G.M., 2005. PopTools version 2.7.1. Available at
1 4 1 ( 2 0 0 8 ) 2 6 6 2 –2 6 6 8
New Mexico Cattle Growers Association v. US Fish and Wildlife Service, 2004. CIV 02-0199 (US District Court). NRC (National Research Council), 1995. Science and the Endangered Species Act. National Academy Press, Washington, D.C. Patlis, J.M., 2001. Paying tribute to Joseph Heller with the Endangered Species Act: when critical habitat isn’t. Stanford Environmental Law Journal 20, 133–217. StatSoft, 2006. Statistica, version 7.1. Tulsa, Oklahoma, USA. Suckling, K., Taylor, M., 2005. Critical habitat and recovery. In: Goble, D.D., Scott, J.M., Davis, F.W. (Eds.), The Endangered Species Act at 30: Renewing the Conservation Promise. Island Press, Washington, D.C, pp. 75–89. Taylor, M., Suckling, K., Rachlinski, J., 2005. The effectiveness of the Endangered Species Act: a quantitative analysis. BioScience 55, 360–368. USFWS (US Fish and Wildlife Service), 1975. Endangered and threatened species: notice on critical habitat areas. Federal Register 40: 17764–17765. USFWS (US Fish and Wildlife Service), 1976. Determination of Critical Habitat for American Crocodile, California Condor, Indiana Bat, and Florida Manatee. Federal Register 41: 41914– 41916. USFWS (US Fish and Wildlife Service), 1984. Listing endangered and threatened species and designating critical habitat. Federal Register 49:38908–38912. USFWS (US Fish and Wildlife Service), 1999. Notice of intent to clarify the role of habitat in endangered species conservation. Federal Register 64:31871–31874. USFWS (US Fish and Wildlife Service), 2001. Joint regulations (United States Fish and Wildlife Service, Department of the Interior and National Marine Fisheries Service, National Oceanic and Atmospheric Administration, Department of Commerce); Endangered species committee regulations, listing endangered and threatened species and designating critical habitat. 50 Code of Federal Regulations 424.12: 405– 406. USFWS (US Fish and Wildlife Service), 2003. Endangered and Threatened Wildlife and Plants; Designation of Critical Habitat for Seven Bexar County, TX, Invertebrate Species; Final Rule. Federal Register 68: 17155–17231. USFWS (US Fish and Wildlife Service), 2004. Endangered and threatened wildlife and plants; designation of critical habitat for the California tiger salamander (Ambystoma californiense) in Santa Barbara County, final rule. Federal Register 69: 68568– 68609. USFWS (US Fish and Wildlife Service), 2005. Final Determination Concerning Critical Habitat for the San Miguel Island Fox, Santa Rosa Island Fox, Santa Cruz Island Fox, and Santa Catalina Island Fox. Federal Register 70: 67924–67929. USFWS (US Fish and Wildlife Service), 2006. Threatened and endangered species system (TESS): species with critical habitat.