A critique of the paper “Potential causes for amphibian declines in Puerto Rico” by Burrowes, Joglar and Green

One of the great things about blogs is that it allows scientists to express their ideas in a more open and direct way. The old way to criticize a paper was to write a letter or a paper and try to get it published. Now, anyone can provide arguments and discuss a paper for the world to see.

In 2004 I tried to publish a rebuttal of the paper “Potential causes for amphibian declines in Puerto Rico” by Burrowes, Joglar and Green, published in Herpetologica. Unfortunately, the editors decided that the wording in their paper was vague enough that it made it hard for them to justify my comments as a rebuttal. I recently stumbled upon the original emails, so I decided to make my original comments of that paper available here. As a disclaimer, I worked with the first two authors for several years and I was one of the students that collected the data used in that paper during that time.

The full citation of the paper is:

Burrowes, Patricia A., Rafael L. Joglar, and David E. Green. 2004. Potential causes for amphibian declines in Puerto Rico. Herpetologica 60: 141-154.

Some of the problems of the paper are:

  • There was no mention of the intensive deforestation of the island (>85% deforested by the 1940’s) during the first half of the 20th century.
  • “…El Yunque, one of the best protected forests on the island.” (Discussion, p. 148) is misleading since El Yunque receives thousands of visitors every year, hundreds of researchers work there, big areas are secondary forest which were used for agriculture less than a century ago, and it is isolated from other forests in the island by urban developments. Is misleading because this makes it seem like a protected untouched virgin forest.
  • The trends at Palo Colorado Forest were from an unreplicated 200 m2 (100m x 2m) transect.
  • The trends at the Elfin Forest were from an unreplicated 450 m2 (3 adjacent plots of 25m x 6m) transect (next to one of the paved trails) which was reduced around 1998 to 300 m2 (just two of the plots were surveyed).
  • The Palo Colorado transect was also used as a trail. I recently went to the end of the Palo Colorado transect to record a population of E. hedricki and was amazed by the lack of vegetation and width of the trail in the transect when I saw it from a fresh point of view. The declines could be artifacts of  the transects due to vegetation trampling and compactation of the soil by footsteps. The transect looked like any other mud trail in the forest.
  • The climate trends were analyzed only from the data of the Pico del Este station when this station is located far away from the transects and can not represent what is happening in the complete forest. It is near the summit of the mountain (1051m elevation), next to a road and next to almost no vegetation in an area with the lowest temperatures and highest wind velocities in the forest. Furthermore, they state that “Analysis […] revealed a few generalizations for El Yunque” (Results – Climate Change, p.147) using the data only from Pico del Este, which is located at the Elfin (or Dwarf) Forest. The Elfin Forest occupied 1.4% of the cover area of the forest in 1989.
  • Data from other climate stations in the forest were ignored, including the data collected monthly at the UPR station next to the Palo Colorado transect by Joglar’s team.
  • The climate data used for the analysis were monthly means, while the population analysis used yearly means.
  • They used mean number of adults per year in figure 1, when they have monthly data. This is an oversimplification that is not explained why it was chosen.
  • There was no mention in the paper methods about normality, or lack of and transformation of the data.
  • Concluding that E. coqui is declining makes no biological sense, being the most common species in the island and a very prolific one. This is the famous recent invader of Hawaii.
  • In fig. 1 both E. portoricensis and E. coqui in the Elfin Forest had a similar U-shaped curve, this suggests a problem in the data collection because I doubt they have a similar response, particularly when they had different responses after Hurricane Hugo. These two species were the only ones surveyed both visually and acoustically in that transect, the other species, E. unicolor and E. gryllus, were only surveyed acoustically.
  • They report that “We found no trend in abundances of E. coqui, E. cochranae, and Leptodactylus albilabris in the lowlands (San Juan, approximately 0m).” These results were from small transects inside the University of Puerto Rico campus and from opportunistic observations, hardly a way to determine trends in amphibians, as has been discussed a lot in the literature.
  • They did not mention that initially the Palo Colorado forest transect was used for a mark-recapture study, which was later dropped. This change in objectives could affect the data collected since the students would be looking for adults initially and not looking for juveniles since they can not be marked, then the trend reverses when the juveniles (living in the understory) are easier to detect than the adults (which usually go to the canopy).
  • The populations monitored for presence/absence (E. richmondi, E. wightmanae, and E. hedricki; Materials and Methods – Status of amphibians, p.143) were always located next to the main roads in the forest (PR-191 and FS-10). No populations inside the forest were monitored.
  • The statement “…the cave-dwelling frog, E. cooki, remains stable in San Lorenzo” (Results – Status of amphibians, p. 145) was based on a single population studied monthly in a private farm. A general survey has never been done because the area is composed of private farms and it was difficult to obtain permission and access to the farms to visit the caves. Also, E. cooki does not inhabits caves exclusively since a recent thesis demonstrated that this species has higher densities in two rivers in the same mountain range when compared with the densities reported in a previous paper by Burrowes and Joglar.

It is very serious to say that some species are declining or not when the data collected was from a single point. This can yield wrong conclusions regarding the declining amphibian population phenomenon, possibly wasting effort and money on working with a “declining” species and ignoring the really endangered ones. And worst of all, since the scientific community is still trying to explain this phenomenon, this paper can create confusion and take interest from possible real hypotheses.

Edited in March 2012 to add superscripts and italics.

This entry was posted in Declining Amphibian Populations, Eleutherodactylus, Puerto Rico. Bookmark the permalink.