National and Kapodistrian University of Athens
Η διαχείριση της ανιχνευσιμότητας στις υποβρύχιες οπτικές δειγματοληψίες: βέλτιστες εκτιμήσεις του εύρους εξάπλωσης και της αφθονίας θαλάσσιων μεγαβενθικών οργανισμών
Abstract
dc:descriptionThe distribution and abundance of marine organisms are some of the main state variables in the case of marine biological monitoring. Ecological assessments of protected species and species in sensitive habitats require the use of visual, non-destructive sampling techniques. However, the nature of the underwater environment and the physiology of the diver-researcher carrying out the underwater survey create conditions for errors and increase the probability of inaccurate assessments. One of the most important such parameter which, if ignored, may lead to significant errors concerns the incomplete detection of the individuals of the species being surveyed. The aim of this PhD dissertation is to investigate and highlight the problem of incomplete detection in underwater visual survey methods and to adapt methodological approaches that account for imperfect detection for their optimal application in the underwater environment.The methods of estimating abundance and/or population density through visual sampling that were considered suitable for application in the underwater environment are as follows: (a) Strip sampling, (b) Distance Sampling (DS), the Lincoln-Petersen mark-recapture or Double Count (DC) method, and (d) Mark Recapture Distance Sampling (MRDS). In order to compare their assumptions regarding detectability and their effectiveness in their assessments, these methods were simultaneously applied to four independent studies targeting the estimation of population size of different megabenthic organisms: the Aplysina aerophoba sponge, the Pinna nobilis bivalve, the cephalopod Octopus vulgaris octopus, and the sympatric seahorses Hippocampus hippocampus and H. guttulatus. In the case of MRDS, this was the first time it was applied by divers in an underwater visual survey scheme. Underwater visual counts were performed for all 4 studies by scuba diving and data were subsequently analysed with the proposed analytical approach of each method. The final population size estimates derived by each method applied in all studies were compared using non-parametric bootstrap resampling. In addition, using the empirical study data, three different cases of stochastic numerical simulation of underwater visual surveys with multiple detection scenarios were constructed, examining the effectiveness of all methods in assessing the abundance and population density of the case study of a virtual target species. The results of these comparisons revealed significant differences in estimates resulting from the use of different methods in different case studies and sampling conditions, particularly in those studies where detectability was low.Furthermore, in the context of assessing the role of detectability in estimations of range of marine megabenthic organisms, a methodology jointly estimating occupancy and detectability based on multiple observations was applied for the first time in an underwater visual survey scheme in Saronikos Gulf, Greece. Using a sufficient number of observers, the probability of recording false absences was minimized and the occupancy estimation was greatly improved. For the whelk Stramonita haemastoma in the case study area, single-observer occupancy estimates were negatively biased and varied significantly (between 0.64 and 0.89) depending on the observer, while with the proposed methodology, using 5 observers, occupancy was estimated to be 0.93. The probability of false absence was high in the case of single-observer and rather low with any combination of 3 observers (< 0.025), while it dropped to practically zero when using 5 observers. Finally, in the context of minimizing the estimation cost, both in terms of fieldwork and data analysis, a methodological tool was developed to guide the selection of the most economical method of visual sampling and analysis of data for unbiased and accurate estimations of the status parameters of megabenthic organisms of interest. This is based on a multi-criteria analysis process that takes into account the characteristics of the target species in the study area, its behaviour, but also the environmental conditions likely to affect sampling. The criteria to be considered are: A) the population density of the target organism, B) the structural complexity of the organism's habitat in the study area, C) the organism’s cryptic behaviour, and d) the environmental conditions and other parameters which influence the outcome of sampling. Each of these criteria is directly or indirectly related to the detectability of the target organisms and thus influences the choice of the most appropriate visual sampling method. The relevant field data of the studies carried out in the framework of this doctoral thesis were used to calibrate these criteria and based on them, a method of calculating an appropriate indicator was proposed, the values of which corresponded to one of the visual sampling methods assessed in this dissertation.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Issaris, Ioannis
- Ίσσαρης, Ιωάννης
Subjects
dc:subject × 14Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/41694