Browsing University of Alaska Fairbanks by Subject "Data processing"
Now showing items 1-2 of 2
Quantifying fisher responses to environmental and regulatory dynamics in marine systemsCommercial fisheries are part of an inherently complicated cycle. As fishers have adopted new technologies and larger vessels to compete for resources, fisheries managers have adapted regulatory structures to sustain stocks and to mitigate unintended impacts of fishing (e.g., bycatch). Meanwhile, the ecosystems that are targeted by fishers are affected by a changing climate, which in turn forces fishers to further adapt, and subsequently, will require regulations to be updated. From the management side, one of the great limitations for understanding how changes in fishery environments or regulations impact fishers has been a lack of sufficient data for resolving their behaviors. In some fisheries, observer programs have provided sufficient data for monitoring the dynamics of fishing fleets, but these programs are expensive and often do not cover every trip or vessel. In the last two decades however, vessel monitoring systems (VMS) have begun to provide vessel location data at regular intervals such that fishing effort and behavioral decisions can be resolved across time and space for many fisheries. I demonstrate the utility of such data by examining the responses of two disparate fishing fleets to environmental and regulatory changes. This study was one of "big data" and required the development of nuanced approaches to process and model millions of records from multiple datasets. I thus present the work in three components: (1) How can we extract the information that we need? I present a detailed characterization of the types of data and an algorithm used to derive relevant behavioral aspects of fishing, like the duration and distances traveled during fishing trips; (2) How do fishers' spatial behaviors in the Bering Sea pollock fishery change in response to environmental variability; and (3) How were fisher behaviors and economic performances affected by a series of regulatory changes in the Gulf of Mexico grouper-tilefish longline fishery? I found a high degree of heterogeneity among vessel behaviors within the pollock fishery, underscoring the role that markets and processor-level decisions play in facilitating fisher responses to environmental change. In the Gulf of Mexico, my VMS-based approach estimated unobserved fishing effort with a high degree of accuracy and confirmed that the regulatory shift (e.g., the longline endorsement program and catch share program) yielded the intended impacts of reducing effort and improving both the economic performance and the overall harvest efficiency for the fleet. Overall, this work provides broadly applicable approaches for testing hypotheses regarding the dynamics of spatial behaviors in response to regulatory and environmental changes in a diversity of fisheries around the world.
SunStar: an implementation of the generalized STAR methodSTAR ... is a method of computing species trees from gene trees. Later, STAR was generalized and proven to be statistically consistent given a few conditions (Allman, Degnan, and Rhodes 2013). Using these conditions, it is possible to investigate robustness in the species tree inference process, the lack of which will produce instabilities in the tree resulting from STAR. We have developed a software package that estimates support for inferred trees called SunStar.