Access
This article is part of Nature's premium content.
Published online 10 December 2008 | Nature 456, 682-683 (2008) | doi:10.1038/456682a
News
Feathers fly over Hawaiian bird
Dispute could stymie efforts to save rare honeycreeper.
High on a Hawaiian mountain, scientists and wildlife managers are clashing over how best to save an endangered bird. The dispute is halting research, blocking student projects and potentially undermining conservation plans for a rainforest that cloaks part of the Big Island of Hawaii.
To read this story in full you will need to login or make a payment (see right).
Comments
Reader comments are usually moderated after posting. If you find something offensive or inappropriate, you can speed this process by clicking 'Report this comment' (or, if that doesn't work for you, email webadmin@nature.com). For more controversial topics, we reserve the right to moderate before comments are published.
It is inexcusable that study of the akepa is being prevented by Hakalau Refuge. That being said, your article appears very one-sided to me. What data does the refuge have to support their claims? They?ve been managing this species for decades. Surely the refuge?s data are worthy? Perhaps there is a similar article in one of the management journals that vilify Freed? Regardless of who is right or wrong, it is likely that the management plans of the refuge would benefit from a peer review process involving both managers and scientists. A similar review process has been implemented for rare species recovery plans on the Army?s Pohakuloa Training Area ? directly adjacent to the refuge. Perhaps one of the other managing bodies (the Nature Conservancy, US Army, US Fish and Wildlife Service) could act as a mediator to resolve the conflict and help revise the refuge?s planning process, if needed? Sean Gleason Macquarie University NSW, Australia
Lenny Freed and his team have done admirably detailed work on the population biology and contemporary evolution of the endangered Hawaiian birds whose major refuge is the Hakalau reserve at Pua Akala. I am not acquainted with the background of the disagreements with Fish & Wildlife, but I do know that disagreements among humans over science and values are among the principle threats to endangered species - we have seen them in other places in Hawaii (the Hawaiian Crow, the Alala, and the Hawaiian land sails), in Texas (the Barton Springs Salamander), and in the Serengeti (the African Hunting Dog). Whenever such disagreements occur, they result in lose-lose situations. The endangered species lose, and the humans working on them lose - all the humans, on all sides of the disagreements. These Hawaiian birds are among the crown jewels of life on this planet and part of the inheritance of all mankind. They deserve the best we can give them. I strongly advocate the intervention of an objective authority who can bring all parties to the table to review the situation and work out a constructive path forward. The MacArthur Foundation, the Nature Conservancy, and the Governer of the State of Hawaii would all appear to be logical candidates to play that role.
Being familiar with Hawaiian forest bird conservation issues, we were surprised by the tone and contents of Mr Dalton?s article concerning the akepa in Hawaii (10 December 2008). Regrettably, it does not appear to approach the standards one might normally associate with Nature. For example: other than John Jeffrey of the U.S. Fish and Wildlife Service, did Mr. Dalton interview anyone in the course of his investigation who did not agree with Dr. Freed? Did Mr. Dalton review the results of a recent analysis of 20 years of avian census data from the Hakalau refuge that concluded the akepa population is stable? The information was made available to him before this article was published and would appear rather germane to his conclusions. Was Mr. Dalton aware of the minutes of the 8-9 October 2008 meeting of more than 30 biologists and managers, including Dr. Freed, during which the status of akepa at Hakalau was considered in great detail over two full days? If so, Mr. Dalton would have learned that the final setting of research and management priorities that resulted from the meeting failed to concur with Dr. Freed?s findings, his proposals to expand his research or his management recommendations. Did Mr. Dalton ask FWS officials to explain why Dr. Freed's recent permit applications were not approved? Was Dr. Freed?s Endangered Species permit in fact ?suspended? as he claims in his recent Evolutionary Ecology Research paper, or did it in fact merely expire? If, as shown in FWS records, it expired and FWS declined to renew it, did Mr. Dalton somehow miss this in his investigation? If Mr. Dalton thought FWS had refused to renew it, would it not have been standard journalistic practice to ask for an explanation? Or, did he simply unquestioningly accept whatever Dr. Freed told him? We realize there are at least two sides of every story and that it can be hard to make sense of what is happening to a very small, rare bird that lives in dense, isolated forest on a Hawaiian volcano. Nevertheless if we are to have any hope of saving the akepa and other Hawaiian species, we need to insist on the best possible science and science reporting, uncomfortable as the process may be. We hope that Nature will be part of this process. Sheila Conant and David Duffy
As Rex Dalton pointed out, John Jeffrey, the senior Hakalau Forest National Wildlife Refuge biologist, did not read our published paper entitled ?Incipient extinction of a major population of the Hawaii akepa owing to introduced species?. Rex told us that Jeffrey felt the U.S. Fish and Wildlife Service (USFWS) could ignore the paper because it was not peer reviewed. However, the Michael Rosenzweig quote in the Dalton article showed how totally wrong Jeffrey was. Jeffrey was joined by Sheila Conant and David Duffy, who in their comment also felt that long-term trend analysis showed no problems, which was the justification for the USFWS and their selected peer reviewers to disregard our work. So, we need to emphasize that we acknowledged the trend analysis of the long-term census data in the incipient extinction paper, and formally dealt with it. We suspect that Conant and Duffy did not read the entire paper because they did criticize our treatment of the census data and analyses. This comment may clarify the issues. The United States Geological Survey, Biological Resources Division (USGS-BRD), performed the long-term trend analyses from 1987-2008. They also performed additional analyses that were ignored by Jeffrey, the rest of the USFWS, and Conant and Duffy. One analysis was long-term trend in the high-density akepa site from 1987-2005. USGS-BRD documented a decrease in akepa and an increase in introduced Japanese white-eye, exactly as we did, even before the akepa crashed in 2006. However, when they analyzed trends over the refuge as a whole, they did not find a decrease in akepa or an increase in white-eye. They did not separately analyze the remaining portion of the refuge, although it is difficult to imagine that the akepa would have increased and the white-eye decreased in that remaining portion of the refuge to exactly compensate for the opposite changes in the high-density site. The USGS-BRD performed a second set of analyses of census data from 1999-2008 in the middle elevation stratum that included our long-term study site. These 10-year trend analyses showed a decline in all eight native species but no decline for the white-eye. These patterns are also consistent with our demographic analysis for the akepa, the increase in the white-eye, and declines in other species of native birds we documented in the incipient extinction paper. The short-term analysis also incorporates stations distant from our study area, so the problems we document may be more widespread than we could extrapolate with our data. Important statistical issues were neglected in the long-term trend analyses. A parametric analysis requires assessment of the fit of the model to the data. Autocorrelation is one of several tests for serial data. The errors also need to be random. The long-term trend analysis almost certainly violates this assumption. If the last 10 years of a 22-year series are showing a decline, then the residuals of a regression showing no trend over the 22 years will have a run of points below the no-trend line, and these negative residuals will also be increasing in magnitude for most of the 10 years. The statistician Frederick Mosteller even asserted that the longest run of points above or below the median could be used as a test for trend in a series. The runs invalidate the model of simple trend that was used, because a model is only as good as its assumptions. The USFWS is thus relying on a model that is inappropriate. Diverse tests of residuals should be conducted and a new model tested that better meets the assumptions. We suggested in the incipient extinction paper that trend analysis needs to incorporate a break point to reflect different slopes on each side of the point, just as we did to show that juvenile survival of the akepa was lower after 1999 than before 2000. We provided evidence that the environmental change justifies the break point at year 2000 because this is when the Japanese white-eye increased and started competing with the akepa and other native birds. There is another approach to analysis that may make even better use of the census data. We pointed out in the incipient extinction paper that the longer the data set the less likely that trend analysis can detect a change unless the change persists for a long time. This makes it impossible for a trend analysis to identify sudden changes. If people have confidence in the long-term trend analysis, then they should have confidence in using data from years 1987-1999, when we documented that the akepa population was viable, to calculate a single estimate of density for that entire time period. This estimate could then be the fixed y-intercept for analyses using later years like 2000-2008 as separate points. We suspect that this approach would accurately show a decline in all species of native birds. This approach could also have the advantage of better assessing the response of the birds to white-eye control or other management. The response would be indicated by the return of the density to the level of the y-intercept. Without loss of information, it would be easier to detect change with this approach because the effective data set would be shorter. It would use the information from 1987-1999 as the baseline for comparing the future, rather than trying to re-estimate the entire trend with data from each new year. We also emphasize that just seeing akepa cannot falsify our results. We documented incipient extinction, not extinction itself. We have shown in previous work that flock size in sites with low-density akepa is larger than would be predicted by density alone. To see flocks in sites where we showed that color-banded, resident akepa had crashed is not evidence that they have not crashed. These flocks could aggregate birds from over a larger area than before. In addition, seeing birds during the mating season could be misleading if the observer happened upon male group activity. In the incipient extinction paper, we used day-long inventories of the study area, repeated several times per month, and for several months to document the crash. Sheila Conant and David Duffy, in their comment, also point out that our endangered species permit could have simply expired and that this disrupted our work, implying no malice on the part of federal biologists. Nothing could be further from the truth. It was suspended because we had not completed diagnosis of malaria in blood samples for the regional office of the USFWS. Why? This was because John Jeffrey had commanded us not to analyze the blood samples until the Service approved the methodology, although we published in peer-reviewed literature why our critics were wrong. Moreover, the USFWS ignored our 2006 paper entitled ?DNA quality and accuracy of avian malaria PCR diagnostics: a review? (Condor 108:459-473), that should have resulted in instant approval of our methodology. But the Service required approval by the USGS-BRD and it was those same scientists who were putting their blood samples in the wrong type of buffer for accurate malarial diagnostics, and who were erroneously blaming primers for false positives, that spearheaded the charge that we were not giving best available science to the USFWS. There are additional items that should be credited to John Jeffrey. One is that he required us to remove all artificial cavities for the endangered akepa, including ones used by at least 8 females, who nested more successfully than females using natural cavities. Second, Jeffrey told Dalton that they could disregard our paper entitled ?Explosive increase in ectoparasites in Hawaiian forest birds? because we did not set out to study ectoparasites, despite showing that 7 out of 8 species of native birds with chewing lice were less likely to be recaptured than those without lice, and every species of native bird with chewing lice had higher prevalence of major fault bars in wing and tail feathers, representing nutritional stress while moulting. Jeffrey?s statement implies that any new threat to Hawaiian birds we discover could be ignored at their discretion. In fact, he specifically forbade us from collecting the birds? lice to determine if they were native or introduced, even though national wildlife refuges are supposed to know what species are present. Third, he took away the $2 million field station that we built with MacArthur Foundation and National Science Foundation support, telling Rex Dalton that it was because there were not enough graduate students using it. Dalton points out in his article that Jeffrey blocked graduate student research if we were the academic supervisors, leading in part to the loss of students! Every graduate student from every university has been welcomed at the field station, but only the refuge can issue special use permits for the research. If the refuge did not issue the permits, it is illogical to blame us. Changed environments can dismantle adaptation, which conservation biologists with the stature of Conant and Duffy should recognize as a problem that can only be addressed using appropriate management. We documented in the incipient extinction paper the dismantling of two life history adaptations (nestling overgrowth and seasonal variation in sex allocation) by food limitation. With the limitation, juvenile survival has become lower and the sex ratio of young birds has dropped from 53% to 13% females. The akepa population will go extinct because young females are vanishing rare, even if males are unaffected and adult survival is normal, which it is not. This problem occurred even before the crash, so in principle young females may be rare throughout middle elevation forests. Adult females in the population are becoming increasingly geriatric, a phenomenon that cannot be detected by census data. Documenting the presence of young females can only be accomplished by close inspection of young birds in the hand, and this requires permits for mist-netting, not just more modeling using existing and future census data. Leonard Freed and Rebecca Cann.
I did some avian research at the Pohakuloa site collecting house finches and I have to say working with the military's staff ecologists was very pleasant. No problems. It's hard to understand what exactly the problem is here? Could it be as ridiculous as the USFW folks being squeamish about the solution to the problem, i.e. eliminating white-eyes from the area? I know as a museum scientist working with birds there are some rather odd, and often strongly held, opinions regarding culling birds, even introduced birds. I've worked with other conservation agencies who adopt a similar attitude. In my opinion it shouldn't take much evidence to justify removing an introduced species from an ecosystem as delicate as those in Hawaii. Erring on the side of removal in my opinion would be the most prudent approach if the feds are really interested in preserving native Hawaiian ecosystems.