Frog population plots (El Verde and Bisley)



Short name: 


Data set ID: 



Population records were compiled for the tropical frog, Eleutherodactylus coqui, by conducting four-night censuses of four 20 X 20 m long-term study plots in the Luquillo Experimental Forest of Puerto Rico between 1987 and 2017.

Date Range: 
1987-01-02 00:00:00 to 2007-12-30 00:00:00

Additional Project roles: 

Name: Eda Melendez-Colom Role: Data Manager


Mark-multiple-recapture during four-night census of four 20 X 20 m plots. Frogs were located visually between 2000-2400 h using headlamps. Adults (greater than or equal to 25 mm) were captured and measured to 0.5 mm with dial calipers. Each was permanently marked by clipping a unique combination of toes. Sex and reproduction condition were noted. Detailed information on methodology can be found in Woolbright 2005 Herpetological Review 36:139-142.

Additional information: 

From a written communication of the owner:

Study plots were established in the El Verde study area in 1987 and in the Bisley study area in 1988. Four nights census of the plots were conducted through 2011 at the Bisley site and through 2017 at the
El Verde site. Basic methods remained constant throughout the study: time constrained one hour searches of each plot, not counting handling time for marking and
measuring frogs. The goal was to restrict all searches to the peak activity hours of 2000 to2400 h. For most of the study that allowed the two plots at a study area to
be censused on the same night. However, with peak frog numbers during the period after Hurricane Hugo, it was necessary to search only one plot per night and sometimes
that took longer than four hours because of handling time. Census was generally done twice a year, but the frequency increased after Hurricane Hugo to detect the nature
of the response, and also during some sabbatical years. Likewise, frequency was lower during periods with other time conflicts.

Information content collected on individual frogs increased during the early years of the study. At first I made no attempt to determine the gender of small frogs
(less than about 32 mm SVL). Those frogs are labelled with an s for subadult until about 1993 at which time I started guessing at their gender (M or F) based on body
type. I also denoted with a question mark frogs that were too big to be labelled with an s but for which I could not detect either vocal sac or eggs. I noted calling
males and gravid females from early on and denoted them with a number 3 in their capture record. By 1993 I had developed a system of recording each capture as a 1, 2,
or 3. Status 1 indicates that I was uncertain of the gender and could be wrong. Status 2 indicated good evidence of gender (visible small eggs in females and
discernable vocal sacs in males). Status 3 continued to indicate active participation in the breeding population (large eggs judged ready to be laid in females and
active calling by males). Thus, a mature male might be a 3 one night and a 2 the next, depending on whether he was observed calling. Likewise, a female with small
eggs might be a 1 on a night when they were not visible through the body wall and a 2 on a night when they were. Data were not retroactively adjusted for frogs that
were captured in subsequent field seasons: if I judged a frog to be a female 1 one season and then saw it calling the next (male 3), I did not go back and change the
prior record.

A note is needed on the comparability of body size measurements. In field seasons that include Woolbright as one of the data collectors, I did all marking and
measuring myself, and those body sizes should be directly comparable for such purposes as calculating growth rates of individuals. Care should be taken not to use
body sizes taken by others for such purposes. Although I did train all field assistants who collected data in my absence, and thus their measurements should give at
least a reasonable approximate of adult size patterns for that season, I have discovered that it is nearly impossible to train people well enough to ensure that they
will get the exact same measurements that I do.

Throughout most of the study, populations were low enough that there were no surviving frogs left with low numbers by the time I had used all the numbers up and
needed to start over. Therefore the same number system (0001 to 4444) was used many times over the 30year study with no concern about confusing the current frog with
the same toe clip from the previous cycle. However, after Hurricane Hugo with an increase in both the frog population density and the frequency of census, that was
not the case. So in those years I added two additional series of numbers in which I clipped a thumb (the medial 5th toe on the rear foot). The frogs with the thumb
clipped on the fourth (ones) foot were designated o (e.g., 2321o) and those with the thumb clipped on the 3rd (tens) foot were designated t (e.g., 4223t). When a
clipping error or natural mark (such as the result of a predation attempt) resulted in two toes being shortened on the same foot, I put those two digits in
parentheses such as 21(23)2. For other natural marks like deformities or missing limbs, I typically just used an abbreviated description of the feature.
Old mark in the comments column indicates that the frog in question was marked in a previous field season. Those frogs are listed in numerical order at the end of
the list of new marks for that season.



Time Period: