Frequently asked questions

Please contact us for a quote.  We do give discount for some kinds of work and offer support for indigenous conservation efforts.

SD cards are included but batteries are not.   PODs are built to accomodate either 10 D-cells or  10 or 25 lithium 21700 cells.

• Standard micro-SD cards up to 256GB can be used – a 32GB microSD card – will hold at least 1 year of data.
• We have supplied either Samsung or SanDisk 32GB  MicroSD Class 10 cards. Others will likely work fine but we are confident in these.
• No special formatting is required.

• Power consumption is around 60mW and must come from a very quiet source.
• F-POD batteries & running times
• More information on the rechargeable lithium pack.

PODs contain hydrophones but do not retain digitised sound because it requires too much memory. Instead they do initial analysis in real time to find tonal events and store summary data on each.  This compresses the data stored by a factor of over 1000. Data volumes for a continuous porpoise study using a conventional WAVE file recorder sampling at around 400k/s is about 0.5TB per week.

Continuous digitised sound from the SoundTrap, C-Sound or similar WAVE file recorders will capture cetacean clicks if the sampling rates is around 3x the frequency of the click or higher.  Then a way of analysing those data is needed but no software currently available gives false positives as low as the F-POD without substantial loss of sensitivity.

A combinatinion of a WAVE file recorder sampling at a 96k or less with an F-POD covers many requirements.

No we don’t rent PODs. PODs may be loaned under some circumstances and at Chelonia’s discretion. Second-hand C-PODs and sometimes F-PODs may be available for purchase. Please contact us for more information.

Please see the user manual which provides instructions on common tasks  and feel free to contact us with specific questions or if the user manual isn’t giving you the answers you’re looking for.

Please visit ‘www.phonehome.org’ to send us further details so we can identify the owner and arrange collection.

The latest XXX. You do not need to re-process your files to detect trains.

Yes/No. The number of changes is slowing down! Apologies for the occasional new error!

Please first see our documentation and tutorial videos on data analysis, we can help with specific questions as well as look at your data. <provide link here to relevant page(s) on website>

TO BE UPDATED The original TPOD.exe train classifier was a ‘positive’ system that aimed to identify trains like porpoise trains. This was changed to a ‘negative’ system that aimed to identify trains that were unlike noise trains. This had the advantage that it did not prejudge acceptable train descriptors and improved performance on a difficult species – the Dusky Dolphin. It also showed up some previously unknown trains sources – e.g. WUTS – more clearly. The v1 CPOD.exe classifier was based on the classifier developed for the T-POD, but gained discriminatory power from the better click descriptions (T-PODs only stored the click duration!). The KERNO classifier improved the performance by changing part of the analysis to non-parametric methods to reduce the effect of outliers (clicks that appear to be part of a mathematically defined train, but actually come from a different source). The most useful not-noise / train sources boundary is set on the Q class scale at Mod / Low but in quiet locations the Low and doubtful trains may also be cetacean trains, while in the toughest locations it may be necessary to exclude the Mod trains. The Q classes can be viewed as different thresholds along a Receiver Operating Characteristic (ROC) curve, but these curves do not apply correctly to this system – a different curve would be needed for a different pattern of ambient noise.

10 Alkaline cells will run for 4-5months, lithium primary cells should last about twice that.

The F-PODs can be set to log 1 minute in 2 or 5 or 10 minutes to further extend deployment time. The F-POD also has a feature that lets you set the start time.

Heavy memory use in sites with high levels of substrate transport noise can limit deployments to 2.5 months.

Yes, firstly read the deployment info on our website <xxprovide link here to relevant page on websitexx> and contact us with any further queries. 

No, please see our mooring and deployment pages for information on this <xxprovide link here to relevant page(s) on websitexx>

Yes, but this can give very misleading results as the animals may be silent in this situation. You can easily check that a POD is working on land – see the section xxxStarting the F-POD in the F-POD User Guidexxx.

Yes, see case studies for details.

If you want to study the frequency spectrum of vocalisations, use a hydrophone and suitably fast recording system. You need a sampling rate at least three times the upper frequency of interest, preferably more. So porpoise clicks, for example, need 500k sample/s. On a 16 bit system this will require about 80 GB storage/day and dedicated software to analyse these large volumes of data.

If you want to monitor animal activity or study their echo-location behaviour over periods of time, PODs are easier as the logistics are much simpler, the data volumes much more manageable and accurate automated and standardised post-processing analysis is included.

Porpoises click for a very large fraction of time – probably more than 85%, and other cetaceans in darkness or turbid water probably also click a great deal of the time. A more thorough answer to this important question is needed! XXX

PODs are not a suitable instrument for recording whistles. Whistles above 20kHz could be detected and would appear as a series of maximum length (255 cycles) clicks with minimal silent gaps between, or as shorter events. However, most whistle content is below 20kHz, some whistles show significant spectral complexity.

No. Some visual validation is a good idea, and it is a quick task to look at ten times in a file and view the first ten trains following each. This generally shows that the false positive rates is too low to be a concern. But if you are studying an area where there are no or few porpoises, false positives may have a significant effect on your data and examination of each train classified as a cetacean train may be needed, but will be quick as there will be few of them.

Yes. Porpoises are among the minority of species that produce rather long clicks that contain only high frequencies within a narrow band, usually centred near 130 kHz. C-PODs identify enough of the sound characteristics of each click to distinguish these narrow band high frequency (NBHF) clicks from the shorter, more wideband clicks of dolphins.

Generally no, but a distinction may be possible between larger groups, e.g. distinguishing beaked whales from smaller dolphins.

Please contact us for an up-to-date estimate on delivery times

Yes. Shipping costs is included but all import duties are the responsibility of the buyer.

Please contact us first with the number of PODs you wish to return for servicing or repair. We will reply with a quote, lead time and important shipping information.