Skip to content
naulon

Not checkedNot checked for notices yet.

Accounting for detectability in fish distribution models: an approach based on time-to-first-detection

Mário Ferreira, Filomena Magalhães

Frontiers in Marine Science · 2015

doi:10.3389/conf.fmars.2015.03.00235 (opens in a new tab)

We email you when a new notice or a question about this paper appears.

Abstract

Imperfect detection (i.e., failure to detect a species when the species is present) is increasingly recognized as an important source of uncertainty and bias in species distribution modeling. Although methods have been developed to solve this problem by explicitly incorporating variation in detectability in the modeling procedure, their use in freshwater systems remains limited. This is probably because most methods imply repeated sampling (≥ 2) of each location within a short time frame, which may be impractical or too expensive in most studies. Here we explore a novel approach to control for detectability based on the time-to-first-detection, which requires only a single sampling occasion and so may find more general applicability in freshwaters. The approach uses a Bayesian framework to combine conventional occupancy modeling with techniques borrowed from parametric survival analysis, jointly modeling factors affecting the probability of occupancy and the time required to detect a species. To illustrate the method, we modeled large scale factors (elevation, stream order and precipitation) affecting the distribution of six fish species in a catchment located in north-eastern Portugal, while accounting for factors potentially affecting detectability at sampling points (stream depth and width). Species detectability was most influenced by depth and to lesser extent by stream width and tended to increase over time for most species. Occupancy was consistently affected by stream order, elevation and annual precipitation. These species presented a widespread distribution with higher uncertainty in tributaries and upper stream reaches. This approach can be used to estimate sampling efficiency and provide a practical framework to incorporate variations in the detection rate in fish distribution models.

Abstract licence: CC BY

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.