As part of a wider project investigating the landscape patterns that influence the transmission of Rift Valley Fever in Kenya, we built this Zooniverse campaign to help us map the distribution of livestock and wild animals at our study sites. We have so much imagery it would be difficult for us to do all the locating and counting ourselves, so we’ve turned to citizen scientists for help!
Rift Valley Fever virus is a disease found across Africa and on the Arabian Peninsula. It mostly affects livestock and wild animals, who catch it from the bites of infected mosquitoes. However, humans can get RVF too – usually from handling sick animals. Outbreaks can cause massive disruption to economic and agricultural systems, as well as loss of animal and human lives. As RVF is very strongly influenced by climate and the environment (see our ‘Education section), it is a really important disease to study in this period of global change.
We plan on combining our data on animal distribution with information on where mosquitoes live and breed, as well as features of the environment. From this, we aim to build models that can predict areas where animals exposed to mosquitoes that carry RVF are likely to be.
Knowing where the disease is circulating could be helpful in targeting interventions and prevent big outbreaks – for example, vaccinating animals against RVF is possible but can’t be carried out during an epidemic, only before. Understanding where prevention measures will be most useful means making the most of the resources available and limiting the harm that RVF can cause!
There’s more information on RVF in the “Education” section of our project page, as well as links to some great resources!
If you want to ask us something specific about RVF, chat with on the talk page, and we’ll do our best to answer!
Check the tutorial and field guide for help with the steps and identifying animals! The IDs can be a bit tricky at first but remember – your best guess is still really useful. You can also contact us with your questions on the Talk pages, and we’ll do our best to help!
Short answer: don’t worry about it! Because multiple people will look at each image, we can get a good idea of what's going on in each tile - if there's lots of disagreement or uncertainty, we can always go back to take a closer look ourselves.
Each image tile in the first campaign is a a 90m x 90m square – currently, the images in the project make up over 10 hectares of land, with much more to come! Once we have a better idea of where animals are in these images, we'll cut them down to 15m x 15m tiles to count individuals!
If you are unsure if an image contains an animal or not, but are leaning towards thinking it does, please choose “Yes, I can see animals”. If in the animal counting campaign, then you can select the type of animal you suspect to be and point out where! Each image will be viewed multiple times – adding together lots of people’s best guesses should give us a pretty good idea of where the animals are!
If an image is obscured – for example, there may be some black "no data" areas, swirly areas where we joined our drone pictures together – or covered completely in trees, and you think it would be impossible to see an animal even if there was one, please click “No, I can't see any animals”. We'll mark these places as obscured when we classify the land cover, so we'll know we can't be certain of whether or not there are animals in these areas.
Go to the “Talk” button at the top of the page, and you'll be able to search through questions from other citizen scientists on the message boards! If you can’t find the answer you are looking for, choose the relevant topic (e.g. “Questions for the Research Team” or “Technical Support”) and start a new discussion. We'll check the forums regularly and answer as much as we can!