Understanding population dynamics, and how it is influenced by exogenous and endogenous factors, is important to the study and conservation of species. Moreover, for migratory species, the phenology and duration of use of a given location can also influence population structure and dynamics. For many species, breeding abundance, survival, and reproductive performance, as well as phenology of nesting, are often the most accessible, and therefore, practical elements of their life history to study. For a population of hawksbill sea turtles (Eretmochelys imbricata), we modeled population change for nesters and total adult females, survival, and breeding probability, from 25 yr of intensive tagging data. We modeled breeding probability as a function of the number of years since last breeding and tested for differences between neophyte and experienced nesters. For each year, we also estimated the number of clutches deposited per female, and phenology of use, for neophytes and experienced nesters. To implement the analysis, we developed a novel generalized multistate open robust design mark–recapture modeling framework, with parameters for survival and transition probabilities, and for each primary period, state structure and arrival, persistence, and detection probabilities. Derived parameters included abundance of observable and unobservable components of the population, residence time, expected arrival and departure periods, and per‐period intensity of study area use. Abundance of nesters increased over most of the time series. Survival probability was 0.935 ± 0.01 (estimate ± SE). Virtually all hawksbills skipped at least one year of nesting. Breeding probability increased by skipping a second year, but then decreased thereafter. Subsequent breeding probability was lower for neophyte nesters than for experienced nesters, but the effect was weaker than the effect of years since breeding. Clutch frequency varied by year, with no discernable pattern of differences between neophytes and experienced nesters. Mean arrival and departure dates also varied, with a slight shift of nesting activity to earlier in the season. The multistate open robust design model developed here provides a flexible framework for modeling the dynamics of structured migratory populations and the phenology and duration of their seasonal use of study areas.

中文翻译：

---

多状态开放健壮设计：根据标签数据，种群动态，生殖力和海龟物候

了解种群动态，以及种群动态如何受到外源性和内源性因素的影响，对于物种的研究和保护非常重要。此外，对于迁徙物种而言，给定位置的物候和使用期限也可能影响种群结构和动态。对于许多物种而言，繁殖丰度，存活率和生殖性能以及巢的物候通常是最容易获得的，因此也是其生活史中最需要研究的实用要素。对于a海龟（Eretmochelys imbricata），我们从25年的密集标签数据中模拟了筑巢者和成年雌性种群，存活率和繁殖概率的种群变化。我们将繁殖概率作为自上次繁殖以来年数的函数进行建模，并测试了新生植物和有经验的筑巢者之间的差异。对于每一年，我们还估计了每只雌性存放新手和有经验的筑巢者的离合器的数量，以及使用的物候。为了进行分析，我们开发了一种新颖的广义多状态开放鲁棒设计标记-捕获建模框架，该模型具有生存和过渡概率以及每个主要周期，状态结构以及到达，持久性和检测概率的参数。派生参数包括人口中可观察到的和不可观察到的分量的丰富程度，停留时间，预期的到达和离开时间，以及研究区域使用的周期性强度。在大多数时间序列中，嵌套器的数量增加了。生存概率为0.935±0.01（估计值±SE）。实际上，所有都至少跳过了一年的筑巢。跳过第二年，繁殖可能性增加，但此后降低。初生巢穴的随后繁殖概率要比有经验的巢穴低，但其影响要比育种以来的年限弱。离合器的频率每年变化，新生植物与有​​经验的巢穴之间没有明显的差异。平均到达和离开的日期也有所不同，筑巢活动略有转移到本季节的早些时候。