This blog post is part of SMRU Consultings #ThrowbackThursday (#TBT) series. Every Thursday we’re going to posting about projects from way back when, many of which are part of developments that are consented or up and running. This #TBT blog is about our C-POD study conducted at the Sizewell Nuclear power plant site in Suffolk.

The purpose of this study was to investigate the spatial and temporal distribution of cetaceans around the development site. The results of our study were to inform the Environmental Impact Assessment for an extension to the site. We were contracted by CEFAS (one of a number of projects we’ve done in the last few years) to explore the spatial and temporal distribution of cetaceans around the Sizewell site.

We designed an 18 month PAM survey using C-PODs  to detect marine mammals around the Sizewell Nuclear power plant site.Typically, surveys for cetaceans involve labour intensive and costly visual surveys from vessels or planes. But there are limitations to these methods, including the inability for surveyors to see anything in poor visibility conditions such as high sea states, fog, glare, low light etc… Passive Acoustic Monitoring (PAM) is now a widely used alternative to visual monitoring. PAM works by detecting the vocalisations of marine mammals while they are underwater,  allowing us to detect marine mammals when they are not visible at the surface – which is quite often considering many species spent 90% of their time underwater!

One such PAM device is the Chelonia C-POD. These units detect marine mammal clicks (listen to harbour porpoise clicks here). C-PODs are able to detect harbour porpoise clicks to approximately 250m and dolphins to approximately 1km. These units will only detect a marine mammal if it is vocalising – but it has been shown that harbour porpoise echolocate almost continuously while foraging or travelling (Verfuß et al, 2005).

Modelling C-POD data:

The analysis of detection positive hours for each species or species group enables the investigation of temporal changes in their relative (acoustic) occurrence in a specific area (e.g. diurnal or potential impact driven changes). This can be done by modelling the data with respect to a range of environmental and construction-related covariates to assess patterns of usage. This is done using a GAM GEE modelling technique that Cormac at SMRU Consulting has explored during his PhD and we routinely utilise for studies where the data are collected in a time-series and so autocorrelation can be a problem. Autocorrelation is where you have observations close together in time or in space (or both) which violates one of the key assumptions of statistical models (that the observations are independent of one another). This is an issue especially for PAM studies (observations every minute and CPODs deployed within kilometres of one another so both time and space) (see Booth et al., 2013 for more on the method).  We wanted to explore a number of other environmental co-variates (e.g. time of day or night, distance from shore, distance to some oceanographic features, tidal features etc.) in the modelling ensures that any detected effect as a result of marine development is properly assessed against the natural variability in occurrence of cetaceans.

One of the key covariates we developed and included was exploring how self-noise from the CPOD (and noise from the environment) impacted the ability of the CPOD to function and detect. This is to be published shortly, an abstract can be found here.

The results showed that harbour porpoise were the most commonly detected marine mammal at the site, and that they were present throughout the year – particularly at the further offshore monitoring sites where detections were highest in the spring but detected most commonly close to shore in the winter. There was also a relationship with the state of tide (either biological or affecting detection of animals) and day and night (weak peak in the middle of the night and weak trough in the middle of the day). Dolphins were also detected at the site but detection rates were extremely low.

Why is this data important???

Developers are required to characterise development sites in terms of marine mammal presence and site use in order to allow a quantitative prediction of impacts. This is necessary to support Environmental Impact Assessments, Habitats Regulations Assessments and European Protected Species Assessments.

Check out some of our PROJECTS using C-PODs:

Our REPORTS and papers involving PAM and C-POD analysis include: