💩 Bio-Pollution: A 250 % Ratio of Dog Poops Compared to Previous Month

104 views
Skip to first unread message

EwA

unread,
Oct 1, 2020, 9:04:48 AM10/1/20
to earthwi...@googlegroups.com

We don't always post about our monthly Fells bio-pollution data collection results, but this one deserves a special mention: The EwA September Dog Poop records are in and sad news – there has been 250% of the previous monthly count (that is, the current number represents 250% of the August count) of poop that EwA recorded this month (208 counts) compared to August (83 counts), with the same number of citizen scientists reporting. In other words, the sample amount is 2.5 times higher than the previous month. And this does not account for what we could not detect (non-detection probability is likely to be very high), so it's just the tip of the 💩 ...

This month's count of poop is our highest recorded since we started. 182 of those 208 migrated to the map. The ones that could not make it either lacked GPS, or had off GPS metadata (usually due to a sync delay between a user's device and local cell towers), or they failed to migrate for an unknown Google map issue (we suspect a file size issue).

Note that we moved from a monthly map view to a seasonal view of the record. Why? We wanted to visually simplify the count (and ease the reading of the map).

Our structured surveys are up and running (those that follow established routes), and we have less opportunistic data. You can have your routes added if you can guarantee to survey your route at least once a month. 

You can visualize the summer season data in our Dog Poop interactive map (the icon on the left denotes this summer records) > click on any of those in the map and you'll see the associated data and a picture of the real thing when the file could be uploaded. All records are stored and retrievable upon request.

Thanks to all of you who provided dog poop data (sharing your photos) this past month!

We're now starting the September poop data layer. You can help! 🗺️  Share the map - If you don't already share your sightings with us - Have your route recorded! ⚠️  We need pics with exact timestamps and location/GPS metadata. The best way for us to collect your data is that you upload your pics to a shared google album that we'll communicate upon request. Contact us at FellsDPMapping@earthwiseaware.org and ask us to join that album, or for any question that you may have. 

Thanks for your time! - Claire

 
Nature Conservation as a Way Of Life
www.earthwiseaware.org | EwA Newsletter Signup » 

EwA

unread,
Oct 1, 2020, 8:47:54 PM10/1/20
to earthwi...@googlegroups.com
To follow-up on the topic. I did run this evening the Friends of the Fells Science Advisory Committee. During the discussion over the map, it was mentioned that the total count we found overall may not be that high to prevent having -say- bins at park entrances to solve the issue. Well, that's not that simple for plenty of reasons. One big one is that there is so much more that we did not catch. 

The problem is a misunderstanding of the map: the map does not convey a census. It's an under-representative sampling at best. The non-detection probability is likely to be very high. A human eye does not do a good job to find those un-bagged scats for instance. And even poop bags can be hard to find when the first leaves and needles fall on the ground. So, understanding that it's not a census is critical, as well as understanding that we are getting at a lower-bound estimate. In short, the map only shows the tippy-tip of the problem.

Now, can we estimate the total number of dog scat? Probably. Next year we could refine the survey and estimate the associated non-detection probability. We could enroll dogs to do the job! After all, there is precedence. Here's for instance an article to measure the effectiveness of scat detection dogs for detecting forest carnivores. So there... 

Something to chew on (or rather sniff on)...

Otherwise, here's an additional summary of the data acquired so far 

The purple bars represent the monthly field sample counts, and the orange bars are the percentage ratio between the current and previous month validated sample.
Reply all
Reply to author
Forward
0 new messages