Grinnell and Swarth had 20 primary base camps distributed around the San Jacinto region. We will use these as the framework for our effort. The biologists of the 1908 expedition, apportioned between two teams, spent a total of 184 team-days in the field from 1 May to 5 September. They spent from 3 to 25 days at each camp, collecting and observing around it as could be done on foot. They recorded their activities in great detail, in notebooks which the Museum of Vertebrate Zoology has scanned electronically and made available on its website. They took about 40 photographs of the sites and marked their sites and routes on topographic maps, all of which have been scanned as well. These records will enable us to ensure that we revisit the same areas covered 100 years ago as exactly as possible.
Of course, the 1908 expedition was run under constraints from which biologists working a century later have been freed. Grinnell and his team traveled by railroad, horse-drawn wagon, or on foot. As a result, they visited most sites only once. Today, all but three sites are accessible by motor vehicle, enabling us to spread our effort more evenly through the seasons to get a more comprehensive picture of the region’s biology than was possible in 1908. We propose surveying each site repeatedly, adding a visit to each site in winter, and distributing the remaining visits through the spring and summer as is appropriate for the elevation of the site and the animals’ life cycle (more in the spring for low-desert sites, shifting toward summer for higher mountain sites). Furthermore, we propose spreading the effort in the field over three years. Because of the wide swings in rainfall from one year to the next, spreading the study over three years increases the chances that the study will encompass a representative sample of current climate conditions.
The Museum of Vertebrate Zoology has developed a protocol for the resurveys of all its historic expeditions, and our proposed expedition to the San Jacinto region will follow this closely. The protocol is designed to replicate the effort of 1908 as closely as possible while increasing standardization that will enable the effort of 2008–10 to be replicated even more closely in the future. The coordination of the survey protocol also ensures that the results of the San Jacinto resurvey can be compared with those of other similar resurveys being carried out elsewhere by the University of California.
Birds will be sampled by three methods. Around each of the 20 camp sites, we will establish 10 points along a path corresponding as closely as possible to the area covered in 1908 while encompassing the maximum habitat diversity possible along a route that can be covered in one morning on foot. The route will be walked in the same way on each of four visits to the site. All birds along the route will be counted, with each of the 10 points being the center of a variable-distance count for 7 minutes, according to the Museum of Vertebrate Zoology’s protocol. The route will be covered between dawn and 10:00 AM. A subset that can be covered in one hour beginning at dusk will be selected for surveys of nocturnal birds. No recordings will be used on the walk out, but recordings of the relevant species may be played to elicit responses on the walk back. The routes and points will be documented by photographs and coordinates recorded by means of the global positioning system (GPS).
Second, birds will be trapped by mist net, with five nets deployed around the camp sites in locations where the trapping is likely to be successful (location amid vegetation, shaded through the morning, sheltered from wind). The nets will be set up and closed in the evening, then opened before dawn the next morning. We will record the time each net is opened and closed as well as the number and identity of birds caught. The time of closure will have to be determined in the field by sun and wind conditions, but the recording of the time they are left open allows a calculation of capture rate. The nets’ situation and location will be recorded by photography and GPS as for the count points. Because some camp sites are in the famously windy San Gorgonio Pass, and because mist-netting is ineffective in even light wind, we must anticipate that this technique may not always be practical at every site.
Third, all birds observed incidental to other activities will be noted. Time not devoted to more structured sampling will be used to search areas visited by Grinnell and Swarth’s team on a more informal basis. For a few species with very localized distributions, such as the Black Swift (Cypseloides niger) and Whip-poor-will (Caprimulgus vociferus), we will make trips to previously reported sites to assess the species’ current status even if they are not in areas surveyed by Grinnell and Swarth.
Two methods will be used to survey for reptiles and amphibians. First, pitfall traps will be used to trap non-aquatic amphibians and reptiles. We will place 50 32-ounce plastic cups in the ground, arranging them approximately 10 meters apart in two meandering lines. We will sample them daily for 5 days, installing and opening them on day 1, checking them each of the following four consecutive mornings, then retrieving the cups at the end of each 5-day period and refilling the hole with dirt. All trap lines will be georeferenced by hand-held GPS units using the WGS-84 datum.
Second, we will follow time-constrained meandering transects of various lengths and durations (to be determined based on habitat conditions) to sample the general herpetofauna, as well as focused surveys for species such as the Southern Rubber Boa (Charina umbratica) and Mountain Yellow-legged Frog (Rana muscosa; the Fuller’s Mill site is one of the few places in southern California where this species survives). Transect surveys will be conducted in the full variety of habitats around each camp site, including along stream courses. Both types of transects will encompass multiple techniques, including visual searching, nooses tied to fishing poles, and flipping of woody and coarse debris.
The use of multiple trapping and survey methods will ensure that the full range of reptile and amphibian species is sampled. Each animal located will be georeferenced by hand-held GPS units using the WGS-84 datum, identified, weighed, measured (as needed for identification), and photographed (as needed for identification) before being released, unless it is retained as a specimen.
Rodents and Insectivores
Rodents will be sampled by live-trapping. At each site, we will set out a line of 40 Sherman and 10 Tomahawk traps as close as possible to Grinnell and Swarth’s trapping sites as can be determined from their field notes and as needed to sample the habitat diversity around each site. The traps will be opened for four consecutive days and nights, baited, and provided with cotton batting for the animals’ insulation. The coordinates of the ends and center of each trap line will be recorded by the GPS.
Shrews (and some rodents resistant to other trapping methods) will be sampled by pitfall trapping (see Herpetofauna). Each animal trapped, by any technique, will be identified, weighed, measured (as needed for identification), photographed (as needed for identification), and marked before being released, unless it is retained as a specimen.
Some species, such as the flying squirrel (Glaucomys sabrinus), may require more custom-tailored techniques for detection.
At each of the 20 camp sites we will set up three stations for detection of carnivores. The stations will be distributed to sample the habitat diversity around the sites, and the coordinates of each will be recorded by the GPS. The focus of each station will be a 12-inch metal stake wrapped around its top with a pipe cleaner and baited with a scent lure (Carman’s Pro-Choice) suitable for multiple species of carnivores. The scent is applied to the pipe cleaner with a toothbrush. Below the pipe cleaner, double-sided tape is wrapped around the stake. Placing the bait on a removable stake allows us to remove the scent after the survey. Using a toothbrush to apply lure to a pipe cleaner leaves a consistent amount of lure at each station. At each station, a Game-Vu digital camera is placed 1 to 2 meters from the stake and 20 cm off the ground. The camera is enclosed in a metal box, protecting it from weather, animals, and people (York et al. 2001). The camera is activated by a motion sensor. Also, animals often rub against the stake, leaving hairs on both the pipe cleaner and the double-sided tape, and these hairs (collected and compared with hairs on museum specimens) identify the animal, providing independent corroboration of identifications from the photographs (Ernest et al. 2002, Taberlet et al. 1999). After an animal has visited the station, the stake is cleaned and the pipe cleaner and tape are replaced.
Bats will be surveyed by means of electronic and audible detection of their vocalizations and mist-net capture. At each of the 20 camp sites we will set up from one to six Anabat detectors, distributed to sample the habitat diversity around the sites and in places where bats are likely to be active (possible roost sites, sources of drinking water, riparian settings, woodland edges and interiors). The coordinates of each detector will be recorded by the GPS. Each detector is placed in a weatherproof plastic container with an extension cable connecting it to a microphone placed 1 meter above ground. Each detector will be operated from dusk to dawn through our stay at each camp site.
The detector records the bats’ vocalizations in computer memory, and we use a filtering program to sort the recordings by their time/frequency signature. We use the program Analook W to convert the recordings to sonograms, then compare the sonograms visually with a library of bat calls to identify the species.
Mist-netting will be used when appropriate to capture bats for species identification. Suitable mist-netting locations include over drinking water sources, in vegetation flyways, and along forested edges. The nets’ situation and location will be recorded by photography and a hand-held GPS unit. Mist-nets will be erected and opened at approximately sunset and will be monitored continuously for a period of at least 3 hours beyond sunset. They will be closed and taken down after the monitoring period. All captured bats will be identified to the species, weighed and measured, photographed, and then released immediately after processing.
Some mammals, especially ungulates and rabbits, are best recorded visually. At each camp, we will establish a spot-lighting survey route. In this technique, a team of two mammalogists, each carrying a spotlight, will walk together along a pre-established route at night. Each shines the light back and forth through 180° on one side of the route, covering the full field of vision. Animals are located by their reflected eyeshine and identified through binoculars or spotting scope. The color of the eyeshine also helps identify the species. The location, distance, and direction of the animal are recorded, as with the variable-distance point counts of birds.
All mammals observed incidental to other activities will be noted.
Animals will be collected under permit from the California Department of Fish and Game and U.S. Fish and Wildlife Service (for birds), with a goal of up to three specimens per species per site. Specimens are needed because of the long time-scale on which this study is based and to uphold the highest standard of scientific documentation. Just as the concepts and definitions of some species have changed from 1908 to the present, these are likely to continue changing. For example,
Grinnell and Swarth recognized only two species of chipmunk in the San Jacinto Mountains, yet it is now known there are actually three. Without collected specimens the identification of the chipmunks recorded by Grinnell and Swarth would have been lost. Some species, even some with profound differences in biology, differ only subtly in external appearance, making voucher specimens essential to confirming identifications. For birds, migratory individuals often mix with resident populations, and dissection, assessment of physiological condition, and identification of specimens to subspecies are needed in these cases to evaluate a bird’s status accurately. An accurate understanding of what is going on inside an animal is essential to interpreting changes, which can be represented by changes in seasonal activity, breeding, and migration as well as simple occurrence. Furthermore, analysis of DNA may give perspective on a population’s history of dispersal or isolation, so preserving samples of genetic material, with specimens to attest to the samples’ identity, is also critical. It is our goal that a new expedition also serve as a basis for evaluation of future changes. Just as we judge the value of the 1908 expedition, ultimately, on the specimens and data Grinnell and Swarth left for us to evaluate, future generations will judge our effort on the same basis.