Lost SD card found, 2016 expedition review

Introduction

As addressed in the previous blog post, there were a few technical problems with the previous expedition to the lagoon, characterized as insufficient field gear, time, and misplaced data. I am pleased to announce that we have recovered the lost SD card from BIO 4, and we can now go forward with the analysis of the data from that camera trap, a Moultrie D 880 camera trap.


Camera trap Data

In terms of a break down of species recorded, again we have a preponderance of opossums (Didelphis imperfecta), which in this case was again caused by suet cake attraction. Next in numerical order, we have nine photos of coatis, in this case two individuals who are black or dark brown in coloration. These individuals were again attracted by the suet cake, and are actively investigating the bait.

Didelphis in all its abundant glory

coatis congregate around the traps. Note motion blur 

We have nine photos of various small rodents or mice, one of which is a good size and seems to be black in color. The diversity of rodents on the summits of the tepuis is not fully available, and I can not diagnose the species at this time. Six camera trap photos were of the squirrel common to the tepui summits.

Cropped image on top clearly shows the rodents of Auyan. Bottom image contains the same animal, identifiable by its glowing eyes

Outside of mammals, we have some six photos of the tepui tinamou, Cryptuellus ptaritepui,a Venezuelan endemic which is often encountered on Auyan Tepui, and has been consistently recorded on our camera traps north of Campo Lecho. This species seems to be readily abundant on the summit of Auyan, and is frequently photographed in dense forests, as it looks for seeds and food along the forest floor.

Ubiquitous tepui tinamou. 

We also have photographed a bird which I have not identified, which is heavily camouflaged in the black and white image. Indeed, I initially considered this to be a false positive photograph until I looked at it in greater detail.

Heavily camouflaged bird can be seen in the extreme right of the top image.

Capture zone and camera notation

The capture zone in this instance included the same fruits which were photographed during the 2015 expedition to the lagoon area, and remain unidentified. However, it is clear from the photographs that these fruits are not a food source for any of the species photographed, as they remain intact during the entire camera trapping period.

Top photo is from 2015, during our initial explorations of Auyan. Bottom image is from our capture zone. The team placed the traps in the exact location to attempt to monitor frugivors on the summit. 

This capture zone was set up in such a way that it was covered by two cameras: A Moultrie camera and a Bushnell camera. The data presented in this blog is exclusively from the Moultrie camera, which is designed to produce high quality images in color, and has a trigger speed equal to the Bushnell model. Qualitative analysis comparing the two cameras demonstrates that the Moultrie model failed to produce a majority of color photographs. There is also a significant amount of motion blur associated with a number of images, most notably the images of the weasel Mustel frenata. While both cameras photographed this animal, only the Bushnell camera had the shutter speed and resolution to make diagnosing the animal on the image possible. The photos from the Moultrie camera are of such low resolution as to render useful diagnosis of the animal almost impossible.

Moultrie camera Bio 4. Long object ephemeral object is a moving weasel

Bushnell camera weasel photo. Although these are two separate encounters (Moultrie on 5/20 Bushnell on 6/21), in both instances the animals were moving

Quantitative analysis demonstrates that while the Moultrie model produced 111 photos with 12 false positives, the Bushnell model produced 783 photos, with around 15 false positives, and included far more color images at the same setting (three photo rapid fire, high sensitivity).

Conclusion

Two issues have been addressed in this post: missing data and recovery, and comparative analysis between two different models of camera traps. In the case of missing data, the protocol is to catalog and check off that all camera traps have been serviced and their data collected before leaving Auyan. This should be accompanied by photographs of the capture site, and proper labeling of the camera traps and sd cards.
In the case of camera trap performance, the issue is now clear: Bushnell outperforms Moultrie in terms of both qualitative and quantitative measurements. Bushnell offers a higher resolution, better sensor system, and produces fewer false positives. In this case, we will be using  Bushnell products and begin phasing Moultrie products out. We will see how the 180 Moultrie camera preforms at the lagoon entrance latter in 2017, and if it manages to produce high resolution photos we may maintain it in the future. Issues of motion blur and background lighting in the Moultrie camera traps will require further field analysis.