GPS Collar Data–based Breeding Status Discrimination Of Giant Panda

Breeding is an important part in the history of animal life, and also the focus of the zoology research. The breeding state is the different period of the animals’ breeding, including delayed implantation, pregnancy, prenatal, postpartum, abortion and so on. The research on the reproductive state of animals has a very important significance in the research of animal ecology theory and the practical application of wild animal protection. Accuratly grasp animals’ breeding statu is conducive to understand the essence of animal behavior. For rare wild animals, understanding and mastering their breeding status, is more conducive to the implementation of scientific protection management plans.The giant panda is an endemic species of our country, and it is also the world’s flagship species. Many researches on the giant panda had been carried out, including the research on the breeding of giant pandas. The research of captive giant pandas’ breeding had entered a deep level in four giant panda breeding research bases(Chengdu, Wolong, Louguantai and Beijing), however, due to the lack of technology and methods, the study on the breeding of wild pandas still remains in the qualitative description stage. In order to enrich the ecological knowledge of giant pandas, this study attempts to find a new method, which can be used to quantitative and further study on the breeding of wild giant pandas.We used two wildness training giant pandas( “Hesheng” and “Qianqian”) weared Global Position System(GPS) collars as the research object. By contrasting the behavior data obtained from observation and motion sensor, we estimated the classification threshold value of digital behavior, and established the method to judge the behavior of giant panda. On this basis, using the reintroducted giant panda(“Luxin”) delivered in 2012 as the next part research object, we classified the digital behavior of “Luxin” in the period of its breeding year and non-breeding year. Then, we combined the average daily activities rate with the activities range of “Luxin” estimated by GPS site to determine directly the breeding status of “Luxin”. The main results are as follows:1. By analyse the activities of two wildness training giant pandas, we estimated the threshold of 32. Based on this value, we calculated the daily average activities rate of these two giant pandas were 0.50±0.02 and 0.52±0.02, respectively. However, by the threshold of 0, the daily average activities rate of the two giant pandas were, respectively, 0.59±0.01 and 0.70±0.01. The result of Mann-Whitney U test showed that the difference between the daily average activities rate calculated based on 32 and 0 was significant(P<0.01). Therefore, using the threshold of 32 could identified effectively the behavior of giant panda. It is a necessary condition for observing indirectly the behavior of wild pandas.2. By the threshold of 32, we clasffied the acitivities of “Luxin” and calculated the daliy average acitivities rate under three breeding status(delayed implantation, pregnancy, postpartum) in both breeding year and non-breeding year. The daily average activities rate in non-breeding year under the breeding status of delayed implantation, pregnancy and postpartum was 0.50±0.02, 0.25±0.05 and 0.16±0.06, respectively. In breeding year, the rate of delayed implantation, pregnancy and postpartum was, respectively, 0.40±0.03, 0.26±0.03 and 0.06±0.02. Besides, the difference between the daliy activities in the non-breeding year(0.24±0.01) and breeding year(0.21±0.01) was highly significant(P<0.01). Concurrent Comparing the daily average activities rate under three breeding status in breeding year and non- breeding, the results showed that highly significant(P<0.01), in the period of delayed implantation and postpartum, between the two years; but in the period of pregnancy, no significant difference was found. Generally, the difference about the daily average activities rate of the wild giant pandas was highly significant, not only between in the two years but also in the delayed implantation and postpartum of the two years. Therefore, using the daily average activities rate calculated by the threshold of 32 could distiguish the breeding status of wild giant panda. In addition, it could also estimate whether the wild giant panda successful mating, parturition and feeding infant or not.3. This study tracked and monitored the reintroduced giant panda “Luxin” up to 30 months by using the locate function of GPS. We recorded a total of 1402 GPS sites(317 days) about the period of delayed implantation, pregnancy and postpartum in breeding year and non-breeding year. The activities range is 2612.11 hm2 in breeding year and 1222.01 hm2 in non-breeding year by Kernel. In breeding year : the lagest activities rang is in delayed implantation period(2070.09 hm2); the smaller activities rang is in pregnancy period(1775.30 hm2); the smallest activities rang is in postpartum period(14.95 hm2). However, the activities rang in non-breeding year is on the contrary:The lagest activities rang is in postpartum period(1018.89 hm2); the smaller activities rang is in pregnancy period(432.50 hm2); the smallest activities rang is in delayed implantation period. It shows that the giant panda has different activities range in breeding year, non-breeding year and different breeding status. This may be due to different foraging strategies that they choose to meet the needs of their own and breeding energy in different periods. Thus, whether it is in the breeding period, or in which breeding status can be clear judged by GPS sites.