3.2 Instrument design

Sensor design

As mentioned in the rationale, two the aims in designing this sensor were: low power and low cost.  The reason behind it being low power is that the sensor could run in a waterproof box continuously, for long periods time in remote areas.  Low cost means that the sensor can be replicated by almost anyone and it doesn't really matter when the sensor is lost or damaged.  Just to be clear, my intention was to make a reliable sensor for around £100 instead of the £1000s that scientific sensors can cost.  As such, Arduino and Raspberry Pi technology are used for data collection and storage.

Conch abundance sensor in a waterproof box

Sensor function

In the simplest terms, the animal density is calculated by counting the number of animals in a defined area.  In this application, we take a photograph of the seabed and the area is calculated from the depth of water and the angle of view of the camera.  To help us gain any understanding of spatial and temporal heterogeneity of the conch density each photograph is stamped with the latitude/longitude and depth data and the jpeg filename has the following format: year, month, day, hour, minute, second, XXXXXX, latitude, longitude, depth (I think the XXXXX is the timestamp from the Arduino RTC, which can drift).  The following posts show circuit diagrams and computer code so that the sensor can be replicated.

Sample photograph taken by the sensor in castle beach pool in St Andrews, with latitude, longitude and depth data stamped.  

Photograph file name "2019,10,14,08,37,30,095208,5620.5109,00247.3261,20.jpg" provides data for analysis.

Filename format: year, month, day,hour,minute,second,XXXXXX,latitude,longitude,depth