Measurement Data of Polar Bears Captured in the Chukchi and Southern Beaufort Seas, 1981-2017

Immobilized bears were weighed to the nearest kg using a load cell, tripod, and chain hoist in most, but not all years that bears were captured. A vestigial premolar was extracted for age determination upon first capture, except for dependent young which were aged based on size and dentition. "Straight-line body length" (SLBL) was measured as the straight line distance from the tip of the nose to either the end of the last tail vertebrae (for bears caught after 2001) or to the base of the tail (for Southern Beaufort Sea bears caught prior to 2002) using a measuring tape extended several centimeters above the bear, avoiding variation with the body contour, in sternal recumbancy (Rode et al., 2010). Tail length was measured from the base of the tail to the last tail vertebrae. An additional measure of length, "total length" was also collected by measuring the distance from a bear’s nose to the tip of the last tail vertebrae along the contour of the bear's body while the bear was lying on its side. Mean tail lengths were used to standardize SLBL measurements by adding mean tail length to SLBL measures collected in the SB prior to 2002 (Rode et al., 2010). Calipers were used to measure the zygomatic width of bear skulls. The girth of a bear was measured as the circumference of the abdomen immediately behind the forelimbs using a rope tightened to matte the fur but not indent the skin. Each bear was assigned a condition score ranging from 1 to 5 per the definitions for each category defined by Stirling et al. (2008). Bioelectrical impedance analysis was conducted on polar bears captured in the Chukchi Sea 2008-2017 using the protocol described in Farley and Robbins (1994). Bears that were rigid (i.e., their muscles were not relaxed), moving, or wet and not movable onto a tarp were excluded from BIA. An adipose tissue sample was collected from Chukchi Sea bears captured 2008-2017 using a 6mm diameter biopsy punch approximately 15 cm lateral from the base of the tail in a fatty region of the rump. Samples were stored in airtight Teflon vials or cryovials at -80°C until analyzed. Lipid was extracted from each polar bear fat biopsy as described in Thiemann et al. (2006). Adipose tissue lipid content was expressed as mass % of sample wet weight. The quantity of ingesta was estimated for Chukchi Sea bears captured 2008-2017. Ingesta was estimated from palpation using the categories of "full", "partially full", or "empty". "Full" bears were identified based on tight, extended bellies and/or observations of feeding on a carcass just prior to capture. "Partially full" bears were identified based on bellies that were extended but not tight and/or defecation while handling. "Empty" bears were those that were not observed feeding prior to capture, had bellies that could be palpated with no apparent sounds or movement of a recent meal, did not defecate during capture, and had no fecal material within their rectum. We also used urea and creatinine (UC) ratio in blood serum to identify bears with empty guts at the time of capture since bears that have not eaten in 7-10 days have reduced UC ratios (Rode et al., 2018). We could not infer whether a bear that had elevated UC ratio had a full belly at the time of capture because the meal that elevated their UC ratio could have occurred at any time during the past 7-10 days (Derocher et al. 1990). Urea and creatinine levels were measured in blood samples that were collected from each captured bear in a red-top, no-additive blood tube. Blood samples were kept unfrozen until the end of the day of collection when they could be centrifuged and the serum removed in a lab. Serum samples were stored frozen in a -80°C freezer until analysis. Each serum sample was thawed and run through an Abaxis VetScan VS2 with a comprehensive rotor to estimate the urea and creatinine levels (Rode et al. 2018). We identified bears that likely had not eaten in the past 7-10 days and would therefore have empty guts as those with an 80% probability of having fasted based on modeled values of UC ratios relative to feeding behavior of captive bears (Derocher et al. 1990) as described in Rode et al. (2016). It should be noted that although an effort was made to remove potential incorrectly measured or incorrectly recorded data, some erroneous values may still occur within this data set and users may want to evaluate whether additional outliers should be removed as potential erroneous values.