Parasite Safari

General image questions:

1. Why isn't there an option for an "unknown" or "not identifiable" species?

See this awesome explanation from Snapshot Serengeti researchers for an excellent description of the problem. In essence, by asking lots of volunteers to give their 'best guess', we can actually glean quite a bit of information about the image, even if the identifications don't agree with each other. For instance, if a blurred, small, furry thing darts past the camera, volunteers are likely to guess something like "hare" or "jackal" or "genet". From these, researchers can say it's probably a small carnivore or hare, and certainly not an ungulate. So we could likely exclude that image for questions that only pertain to ungulates. But if "unknown" had been an option, the researcher would have 10 classifications that all say "unknown", but with no information about what is actually in the image.

2. Why do some images have water, and others don't?

In order to compare the effect of water on animal behavior, we need to have a control reference! Images without water are our 'Control' sites, which are located at least 1 km from any water source.

3. Why do some images have nothing in them?

Sometimes grass, clouds, and trees can trigger the camera when no animal is present. In some rarer cases, a rapidly-moving animal can trigger the camera, but it has moved out of the frame by the time that the image is actually taken (there is a < 1 second delay).

4. A couple of images seem to have the wrong date and time. Should I be concerned about this?

Occasionally a camera has to be reset, or a new camera isn't properly programmed to the correct date when it's deployed. But don't worry! We check all the image metadata and correct any times that are incorrect. Unfortunately, we can't change the timestamp on the image though!

Specifics of counting:

1. What counts as "drinking" or "grazing"?

If the image is of a watering hole or watering pan, then we count the animal as 'drinking' if we can directly observe the animal taking water. We do not assume that the animal is drinking if it is merely standing next to water.
We count 'grazing' when the animal's mouth is touching the ground to ingest grass or roots. We don't assume that an animal is grazing if its head is just slightly lowered.
We've also had some species-specific questions:

A. Elephants: Can elephants be counted as drinking if I:

i) see their trunks are wet,

We prefer not to count the animal is drinking if the trunk is wet because our goal is to quantify the amount of time that the animal spends drinking. An elephant's trunk might stay wet for a much longer period than it spends drinking!

ii) have their trunks curled into their mouths as they do to drink,

Yes, since we are quantifying time spend actively drinking, this counts.

iii) have water dripping from their mouths?

This is a bit of a gray area, but yes, this can be counted.

iv) How do I tell if they are grazing or just sensing/smelling the ground with their trunks? Only if they're holding vegetation to eat in their trunks?

Usually you can see an elephant ripping vegetation from the ground and then carrying it to its mouth in the next image.

B. Cattle: How do you estimate the total number of cattle when there's a whole herd at the watering hole? Or should I only count the visible ones?

Since there are often large herds of cattle at watering holes, we don't expect you to count each one precisely. A rough estimate is sufficient. If the entire water source is completely surrounded, it is probably safe to assume that 30+ cattle are present. If there are a moderate number, then 21-30 could be estimated. We recognize that often we are underestimating the total number of cows and will account for this in our analyses.

C. Giraffes: Can I assume giraffes are drinking if they have their front legs in the splayed drinking position, even if I don't see their mouths in the water?

To maintain consistency, we prefer to only count animals as drinking if we can see them actively drinking. Each image set is usually part of a longer image sequence, so even if you don't observe the giraffe drinking in your current image set, it's likely to be counted in another image, thereby giving us a more precise measurement of time spent contacting water.

2. Should I count animals in the far distance?

No, you do not need to count those animals. They were unlikely to have triggered the camera, are not interacting with the water, and are probably biased toward larger species.

3. Do I count an animal as grazing if its mouth is touching the ground but there's no vegetation to eat?

Yes, this can be counted as grazing, as it may be especially relevant to warthogs that feed on roots.

4. If the whole animal with its next is bent to the ground but the head is just cut off by the camera, can I assume it's grazing?

In these less common cases, it can be assumed that the animal is grazing if this seems reasonable.

In an ideal scenario, these gray areas wouldn't exist! We thank you for taking the time to help us to make careful classifications, but we also encourage you not to worry when you may be uncertain. We value consistency across images, as this helps us to make unbiased comparisons across our experimental treatments and periods. Thank you!