Quantifying iridescent coloration in animals: a method for improving repeatability

Quantification of animal colors is important to a variety of fields of research, especially those dealing with visual communication and sexual selection. Most animal colors are easily measured using well-established spectrophotometric techniques. However, the unique characteristics of iridescent colors present particular challenges and opportunities to quantify novel color metrics. Due to the fine-scale angle dependence of iridescent coloration, color metrics, such as hue and brightness, must be measured using methods that allow for repeatable comparison across individuals (e.g., by carefully controlling and measuring viewing geometry). Here, we explain how the optical characteristics of iridescent colors should be considered when developing measurement techniques, describe the pitfalls of some commonly used techniques, and recommend improved methods and metrics (angular degree of color change and breadth of reflectance) for quantifying iridescent color. In particular, most studies of iridescent birds to date have used less than ideal procedures and have not provided repeatability estimates for their methods. For example, we demonstrate here that measuring coloration from overlapping patches of iridescent feathers may be problematic, and we argue against methods that do not carefully control viewing geometry. We recommend measuring iridescence at maximal reflectance angles using an apparatus that allows for sample rotation, and we compare this technique to some other commonly used methods using iridescent gorget and crown feathers from Anna’s hummingbirds (Calypte anna). Our apparatus allows for the quantification of angular color change, and we found that maximal reflectance measurements using single feathers are highly repeatable both within feather samples and among samples within an individual.     

Tracking the response of industrial fishing fleets to large marine protected areas in the Pacific Ocean

Large marine protected areas (MPAs) of unprecedented size have recently been established across the global oceans, yet their ability to meet conservation objectives is debated. Key areas of debate include uncertainty over nations’ abilities to enforce fishing bans across vast, remote regions and the intensity of human impacts before and after MPA implementation. We used a recently developed vessel tracking data set (produced using Automatic Identification System detections) to quantify the response of industrial fishing fleets to 5 of the largest MPAs established in the Pacific Ocean since 2013. After their implementation, all 5 MPAs successfully kept industrial fishing effort exceptionally low. Detected fishing effort was already low in 4 of the 5 large MPAs prior to MPA implementation, particularly relative to nearby regions that did not receive formal protection. Our results suggest that these large MPAs may present major conservation opportunities in relatively intact ecosystems with low immediate impact to industrial fisheries, but the large MPAs we considered often did not significantly reduce fishing effort because baseline fishing was typically low. It is yet to be determined how large MPAs may shape global ocean conservation in the future if the footprint of human influence continues to expand. Continued improvement in understanding of how large MPAs interact with industrial fisheries is a crucial step toward defining their role in global ocean management.     

The viticultural system and climate change: coping with long-term trends in temperature and rainfall in Roussillon,France

Mediterranean viticulture could suffer from hotter and drier growing seasons over the coming decades. The present article focuses on the wine-producing area Côtes-du-Roussillon-Villages near Perpignan, in southern France. We used observational daily data (1925–2010) from Perpignan weather station and daily outputs (2001–2060) of the regional climate model ARPEGE-RETIC-V4 from Météo-France with scenarios A2, A1B and B1, to assess the exposure of the regional wine system to changes in temperature and precipitation, both in the recent past and the coming decades (1925–2060). Temperatures during the growing season and summer temperature extremes have been increasing continuously since the mid-1980s and are projected to increase faster from the mid-2040s. Precipitation is highly variable and very low in summer, and projections suggest greater uncertainty, and more extreme drought events could be expected. The analysis of climate data was complemented by thirty-two in-depth interviews with local actors of the wine industry to assess the impacts of climate change on their activities and potential adaptive options. Producers reported negative impacts of recent changes in climate in conjunction with a difficult economic situation. Analyses of historical, social and economic backgrounds are important to fully conceptualize the nature and extent of climate change risks in the region. This case study provides important insights into the roles of non-climatic factors in the generation of vulnerability for Mediterranean agricultural systems facing rapid climate change.     

Tutor choice and imitation accuracy during song learning in a wild population of the Puget Sound white-crowned sparrow

Songbirds learn to sing by modeling their songs on the songs of other males through a process of social learning. Models of social learning predict that animals should be selective in what and when they learn. In this study, we asked whether young males in a wild population of the Puget Sound white-crowned sparrow, Zonotrichia leucophrys pugetensis, were selective in their choice of tutor models and what factors influenced how accurately they imitated tutors’ songs. We first examined two strategies for tutor choice: whether pupils have a conformity bias and/or a preference for high-quality tutors. In keeping with a conformity bias, tutors that sang song types that were relatively common within a radius of about 500 m of their territory were more likely to be imitated than were tutors that sang rarer song types. Most potential tutors were not imitated by pupils. Aspects of tutor quality, such as age, pairing status, and survival to the next year had no effect on whether a tutor’s song was imitated. Secondly, we tested whether pupil repertoire size, pupil quality, and local abundance of tutor models affected the accuracy of song imitations. We found a trade-off between repertoire size and tutor imitation accuracy with males that sang two or more song types developing significantly poorer imitations than males that sang one type. We discuss possible functions of a conformity learning strategy and factors that could produce a trade-off between imitation accuracy and repertoire size.