In an attempt to evaluate the efficacy of the U.S. Endangered Species Act (ESA)--arguably one of North America's most crucial conservation tools--a pair of researchers from the University of Ottawa has sifted through decades of congressional reports outlining the recovery progress of threatened species. Their final assessment? Either the ESA is not nearly as effective as scientists and legislators once thought, or we simply don't have enough species recovery data in order to make a thorough and accurate evaluation of the bill's impacts.


(A Delmarva Peninsula Fox Squirrel, Sciurus niger cinereus, one of the first species to be placed on the endangered species list.)


Passed in 1973, the ESA outlines several major tools for reducing extinction of species and degradation of their habitats, including: protection from take (including harassment, harm, and killing), section 7 consultation (whereby federal agencies consult with the Fish and Wildlife Service to ensure that proposed activities will not threaten listed species), provision of funding (for research, enforcement, and the purchase of threatened areas), design and implementation of recovery plans, and recognition of critical areas (needed for both current habitat requirements and those that may arise in the future). Although these are the most commonly implemented elements of the ESA, not all species recovery plans involve each of these measures, and there are also other tools that are used on a case-by-case basis. The authors of the current report reasoned that variations in tool use might relate to variations in recovery, such that the most effective techniques would yield the biggest improvements in species numbers.

They are not the first researchers to look for this sort of relationship between ESA activities and conservation success. Several previous studies have investigated the effects of various tools--especially funding, recovery plan development, and critical habitat designation--on species recovery. However, findings of that research have been contradictory. Perhaps more importantly, the authors of those studies were looking for the presence/absence of an effect rather than attempting to measure the effect size. For instance, while researchers have reported that amount of funding is significantly related to species status, they did not focus on the strength of that relationship (e.g., the quantity of variation in status explained by variation in funding), or compare it to those found for other conservation tools--important analyses to perform if you are interested in identifying which of several influential tools is likely to have the greatest impact on conservation outcomes.


(Whooping crane, Grus americana, and chick. This is another of the first species to be placed on the endangered list. One dubious benefit of this "honor" is that this species was able to avoid the average 11-year wait associated with the listing process.)

In the current study, researchers evaluated the recovery status for all species listed prior to 2003. Recovery was measured in two ways: number of recovery objectives achieved (as evaluated on a scale of 1-4), and population trends (e.g., declining, stable, or increasing). In order to determine which factors had the greatest impact on recovery, the researchers also noted the length of time each species had been officially listed, the length of time a recovery plan had been in place, the length of time critical habitat had been designated, and the amount of federal funding provided for the recovery effort. Data on all of these factors were extracted from biennial species recovery status reports submitted to Congress between 1988 and 2006. The scientists also hypothesized that another important predictor of recovery might be the amount of published literature available on each species, since knowledge can improve recovery efforts and also since funding facilitates research. Information on this variable was obtained by searching an academic database for papers on each species.

Ultimately, the scientists were able to include 1179 species in their study--61% plants, 14% invertebrates, 9% fish, 6% birds, 5% mammals, 3% reptiles, and 2% amphibians. Cumulatively, conservation progress trends did not look very promising; over the study period, populations tended to remain stable or trend towards a slight decline, and, on average, less than a quarter of recovery objectives were achieved for most species. 


(A Texas blind salamander, Typhlomolge rathbuni, another endangered species list pioneer.)


The proportion of recovery objectives achieved was significantly related to proportion of requested funding provided, amount of peer-reviewed literature published, and both length of time listed and length of time with a recovery plan. However, despite the clear statistical significance of these factors, cumulatively they only explained 13% of the variation in recovery patterns observed across all species--suggesting that no one conservation tool was having a particularly strong effect on population trends. Similar results were obtained from an analysis on change in population status; in this case, number of years listed and proportion of requested funding provided were both associated with population improvements, but together these variables explained less than 10% of variation in recovery.

Interestingly, patterns varied according to taxonomic group, with birds, mammals, and fish displaying higher recovery rates than plants, amphibians, and invertebrates. In fact, when the researchers repeated their analysis within each group, rather than across all groups, as much as ~40% of variation in their data was explained. In birds, for instance, number of years listed explained ~21% of variation in population status, while proportional funding and critical habitat designation together explained 39.9% of variation in mammals.

Birds and mammals are two of the most intensively studied groups of organisms, which suggests that it is not a coincidence that these are the two taxa with the strongest patterns. Species status was only provided an average of 68% of the time across all the congressional reports, and data were most common for birds and fish, but least common for plants. Further, the likelihood that species status was known was strongly related to the amount of peer-reviewed literature available on that species. Taken together, these results suggest that there are many species for whom there are not sufficient data on which to base an estimate of population status, and potentially many more species in which the given status may not be representative of conditions on the ground.


(The shortnose sturgeon, Acipenser brevirostrum, is an early-listed species whose stocks were depleted in some areas of the US by the mid-20th century.)


Thus, the authors feel that the seeming low efficacy of the ESA tools examined here--listing, recovery plan development, critical habitat designation, and funding--may not actually result from a flaw in the methods so much as a flaw in our ability to gauge how well those methods are working. They take some consolation from the fact that the "best among the weak predictors of recovery" is the number of years a species has been listed, implying that protection from take and section 7 consultation do provide benefits, even if they are not huge. 

Two of the major problems associated with current species recovery reports are the use of qualitative data--with species recovery patterns often described relative to previous reports--and a lack of peer review, such that some reports merely reflect staff opinions rather than statistically supported results. An emphasis on quantitative data and rigorous analysis should help improve accuracy and make the reports more scientifically useful in the future. Unfortunately, this may require more time and money, or, at the very least, a redistribution of current resources. However, the authors point out that efforts to address these problems should benefit ESA efforts as a whole, as well as potentially helping the species protected by the bill.

---

If you have suggestions about future Anthrophysis posts, or want to share photos/links/comments, please visit the Anthrophysis Facebook page!


Gibbs, K.E., and Currie, D.J. 2012. Protecting endangered species: Do the main legislative tools work? PLoS ONE 7(5):e35730.

Thanks to the following websites for providing the images used in this post:

http://www.wilddelmarva.com/endangered-survivor/

http://www.learner.org/jnorth/crane/index.html

http://en.wikipedia.org/wiki/Texas_blind_salamander

http://cnre.vt.edu/efish/families/acipenseridae.html