Diamondback Terrapins in the Chesapeake Bay – 2002 Beach Survey
<< Back to Summary page | Proceed to Results page>>
The survey was conducted to assess the presence of terrapins in the Chesapeake Bay in Maryland based on evidence related to nesting. On the Eastern shore, the area surveyed extended from Handy’s Point to Crisfield, MD, on the western shore the area extended from Bodkin Point through the mouth of the Potomac River. We also surveyed all the islands, including including Eastern Neck, Kent, James, Barren, Hooper complex, Bloodsworth, Adam, Holland, Spring, South Marsh, and Smith Island.
The survey was conducted over 36 days between 5/30-8/2/2002. Daily field assistants ranged from 2-8 people, mean of 7.7 person/day, working 8 am to 5 pm with a 1 hour break at 12-1 pm. The survey protocol directed each 2-3 person crew to follow the shoreline by boat identifying areas open to the water and potentially available to nesting terrapin females. Tidal creeks were surveyed until waters became too shallow to access by boat. When any sandy areas or open sites were observed, handheld GPS units were activated. A crew member was dropped off at one end of the beach and recorded a start location point (e.g. "waypoint1" or "wpt1") with latitude and longitude in decimal degrees. Crew members walked along the tide mark following the shore in one direction searching for signs of turtle nesting, including turtle tracks, nest, or shells. When the end of the beach was reached, the end location point ("wpt2") was recorded. Extremely long beaches were surveyed by multiple crew members, each responsible for covering a segment of the whole transect.
Locations for each sign of turtle nesting were recorded into the GPS and described on the data sheet. Presence/absence of turtle activity (1/0) was based on turtles, crawls or tracks for adults and hatchlings, nests, egg-shards, empty shells, turtles observed swimming close to the shoreline of the site, as well as reports from local residents that terrapins have been using the site. If a female was observed digging a nest, or a mound was confirmed to contain freshly deposited eggs, we recorded it as "active." All other nests were identified by a hole, and/or shell fragments scattered around it. These were considered to be"predated", unless otherwise noted in the data. However, due to time constraints, unless fresh signs of a predator were obvious, no steps were taken to identify the specific predator involved. Terrapin nests are known to be vulnerable to a wide variety of mammalian and avian predators, including racoon, fox, otter, gulls (Pfau & Roosenburg, 2010).
In cases when packets of eggs shards were found associated to no specific nest, if clustered, we assigned them to that site rather than considering them as having been randomly transported by waves or wind.
Between each beach survey or whenever possible, data locations were transcribed from the handheld GPS units onto the data sheets to minimize potential loss of information.
Boats used for the survey included center console skiffs: 18’ Privateer, 17’ Boston Whaler, an 18’ Sea Ark (Jon boat) and 21’ Parker. Boats traveled from 6 mph (no wake zones) to 15 mph depending on location and potential for nesting habitat, hugging as close to the shoreline as the draft of the boat would permit. In cases when the probability of terrapins nesting was close to none (e.g. cliffs, continuous bulk head walls), boats ran at about 30 mph, while crew checked for small openings with pockets of sandy shore for turtles.
Handheld GPS units used were Magellan Meridian and Magellan Meridian Marine (Magellan Corporation, Santa Clara, Ca), which are accurate to 3 m with activated WAAS.
Data collection fields present in downloadable data.
- beach_m – length (in meters) of exposed shoreline
- measure - either ‘transect’ or ‘shoreline’. The length beach_m was calculated initially by the straight line distance between starting and ending waypoints for each site (transect) or following the curvature of the shoreline (shoreline). This shoreline method involved using digital orthophotographic quarter quads and maps available from Google Earth
- record_id – a unique number we assigned to each beach & survey. There may be gaps in number sequence where we surveyed larger beaches with two observers, then merged data at a later date
- survey_day – date of survey
- survey_tm – time of survey
- lat – latitude in NAD83 geographic coordinates from GPS unit
- lon- longitude in NAD83 geographic coordinates from GPS unit
- density - relative density of signs of beach use ("counts") divided by beach_m
- presence - 1 for true, 0 for false, if any signs of turtle use were present; anything from "counts" plus: tracks, turtles in the water near the shore, reports of nesting by locals
- counts - the basis of density; count of detections of current use, such as eggshells, nests, turtles present (live or dead), hatchlings. Unlike presence, this does not include the tracks, turtles in water, or local reports
- observ - descriptive summary of observations that are related to turtle activity; notes by surveyor
- nest_type - active or predated (see description under Methods Details, above)
- egg_shards - Presence or absence of eggshells. Included in "counts"
- tracks - Crawl trails by adults or hatchlings. Included in "presence"
- evaluation – the surveyor's qualitative assessment of status of terrapin use of the beach
This dataset is from a survey organized and conducted by Paula F. P. Henry, G. Michael Haramis and Dan Day at USGS Patuxent Wildlife Research Center. Mark Wimer created this website and provided the technical assistance to make the dataset readily accessible as maps and through the metadata, from the PWRC website.
We would like to acknowledge the help and support from the US Army Corps of Engineers, US Department of Defense, US Fish and Wildlife Service, Maryland Department of Environment, and Maryland Department of Natural Resources.
We also would also like to thank the many field assistants, volunteers, Refuge and University staff, biologists, veterinarians, and the many Maryland home owners without whom the survey could not have been conducted safely and effectively.
Use Constraints for applying these data
Each survey site was visited once in 2002. For this reason, we cannot estimate the probability that beaches actually used by terrapins for nesting in 2002 were missed (the detection probability). The lack of a detection probability estimate, the nature of a single-observer bias, and the uncertainty of true numbers of terrapin nests that year relative to those observed at a single point in time, are reasons why we refer to the counts / length as "relative density." For inference, be cautioned that these data are from a number of years ago, and should not be taken to represent present-day occurrences. Many factors may affect what constitutes as good nesting habitat and for the presence and detection of terrapins, including:
- sea level change at the beach (through sea-level rise or beach accretion/subsidence processes)
- changes in the beach exposure (more of less cover as a results of increases in the growth of invasive plants or alternatively erosion or storm damage and subsequent loss of vegetation
- observer, time-of-day, and time-of-year biases
- predators and their behavior.
Having noted these qualifications, these data are useful in determining potential nesting range of the Diamondback Terrapin within the Chesapeake Bay, and are a good starting point for modeling terrapin distribution or in designing a new survey.
<< Back to Summary page | Proceed to Results page>>
Last updated: website – April 2012; data collected 2002.
Disclaimer: Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
A terrapin egg up close. Photo: USGS.