The proposed Windmar RE windfarm near Guayanilla occupies habitat of Caprimulgus noctitherus, the Puerto Rican nightjar or Guabairo. As a result of this, the Environmental Impact Statement addresses the potential impact of the project on this species. A consultant’s report* was prepared on behalf of Windmar RE by Paul Kerlinger of Kerlinger & Curry LLC. documenting surveys carried out in 2003, and a follow-up memo* from Kerlinger details additional surveys carried out in 2004. These results are also discussed (in more rosy terms) in the Habitat Conservation Plan**
Kerlinger and associates surveyed the site in 2003 and again in 2004, and estimated that there were 33 nightjar territories on the site in 2003 and 46 territories in 2004. Kerlinger attributes this increase to one of three causes
- That the observers had become better at finding birds;
- That there was an overall increase in the nightjar population; or
- That “the access roads that have been cut through the forest at the WindMar site provided better foraging habitat such that more territories could occupy the site”
Kerlinger favours the third explanation – that the construction of roads opened up the canopy, making foraging easier for the birds. He also says that “the trails at the WindMar site may now provide better foraging habitat that actually attracts nightjars from other areas or permits them to live on smaller territories by making foraging better.” He supports the assertion with a few anecdotes, including:
On one night at the Punta Ventana property, where more access roads were established, the data collection team reported an adult with two recent fledgling nightjars foraging along one of the access roads, and another older fledgling foraging along another.
In all of these, Kerlinger fails to consider a variation of (1) – that construction and widening of roads makes it easier for researchers to move through the forest and may change their perceptions of the direction and distance of bird calls. Dry forests are difficult to move through in the night. While narrow trails had been cut for the 2003 survey, wider trails make for easier access, and allow the observers to travel more quietly. It is rather curious that Kerlinger neglected to address this possibility. In addition, his anecdotes about observing adults with fledglings is really pretty meaningless – basically he says that, once the roads were cut, he was able to see fledglings along the roads. Is it really worth mentioning that the presence of roads makes it easier to see animals along roads?
In discounting the idea of population increase, Kerlinger suggests that “[i]t is important to note that the density of nightjars on a per hectare basis was higher on the Wind Mar project site than has been reported in most parts of Puerto Rico. Given the high density, one would expect birds to disperse to other locations rather than become even denser at the Wind Mar site.” In the Habitat Conservation Plan Guarnaccia expands on this by noting “Besides, singing territory size did not decrease as one would expect in more densely packed habitat (Weeden 1965). Instead, it appeared to increase slightly.” Guarnaccia uses this to argue in favour of the idea that increased road density has improved the habitat for nightjars.
I find this rationale puzzling. As Guarnaccia indicates, as density increases, territory size usually shrinks. Territory size tends to be driven by two things – access to food, and competition with conspecifics. Competition with other animals consumes resources, so it’s unprofitable to fight for more territory than you absolutely need. But if you don’t have enough territory to raise your offspring, then there is a strong impetus to compete for additional territory. If you can’t secure enough territory at a site, then you are likely to look elsewhere.
If the construction of the roads increased the resource base and led to an increase in population size, then you would expect territory size to shrink. An increase in territory size (assuming it is real) is more likely to be driven by a decline in habitat quality. It is the opposite of what you would expect given an improvement in habitat quality. Therein lies the problem – it is unreasonable to use the same explanation for conflicting observations. While increased population density is consistent with the explanation favoured by Guarnaccia and Kerlinger, increase territory size is inconsistent with that explanation. While there isn’t enough data to distinguish between an observer effect or a real population increase, I think it’s reasonable to reject Kerlinger and Guarnaccia’s explanation, based on the data provided.
It’s possible to attribute the observations to either an observer effect or a real population increase. An increase in the number of territories is consistent with a road effect: better access made for better surveys. On the other hand, an increase in population size (or degradation of habitat elsewhere) would result in additional territories being occupied. Increased territory sized (if the increase is real and not just a statistical artefact) would be consistent with habitat degradation (perhaps as a consequence of road construction).
What’s more important in all this is that they have only presented two years of data. Two data points will always appear to show a trend, whether one exists or not. But the reality is that it takes a minimum of three data points to be able to say anything about your estimate. Kerlinger emphasises that their sampling intensity was greater than that of Vilella and Zwank. While an increase in nightjar population over the last two decades is reasonable (based on sightings in the last few years which suggest range extensions), it is also possible to attribute these to increased awareness of the species leading to an increased observer effect.
It turns out that I am not the only one unimpressed with Curry and Kerlinger’s work (scroll down to the story after the one about this project).
*Kerlinger, Paul. 2003. A Preconstruction Study of Abundance and Distribution of the Federally Endangered Puerto Rican Nightjar at the WindMar Re Project, Guayanilla, Puerto Rico;
Kerlinger, Paul. 2004. 2004 Territorial Boundaries of Puerto Rican Nightjars at the WindMar Project Site in Guayanilla, Puerto Rico.
**Guarnaccia, John. 2005. Final Draft: Habitat Conservation Plan. WindMar RE Project Guayanilla, Puerto Rico
Other posts on the subject:
- Guayanilla Windfarm – general thoughts on the topic.
- Species-area curves – when they get the get the most basic biology so badly wrong, you tend to lose confidence in what they have to say rather rapidly.