Wednesday, April 3, 2013

I answered my eBird question and learned a lot in the process!

Whew! I have a new way to see what birds were reported in an area near an eBird Hotspot. In doing so I was able to re-create some cool data I saw during the GBBC. OK, I got it!

But, it does not allow me to do what I want.

OK, here is what I want:

I want to get a total checklist of birds from one area over time. Case in point, how many birds have ever been recorded from Beaudette Park in Pontiac?

Note: all below references start with Explore Data in eBird.

On my first cut, I used the "All Time First Sightings" and selected Beaudette Park as a Hotspot.  I fugured that any list of all birds must have had a first sighting - right? Yes, this is true.

It gave a total, but was a bit weird because the first recorded checklist was from 1974 and included, for example, Mute Swans in mid-summer as a first sighting. True, and it was the first record of the species at Beaudette, but these guys hang around all year! What it really told me that a life-long birder is entering life-long data into eBird, and this represents one of the earliest checklists there! . Really that is SO cool!!! Kudos to you Jim VanAllen!!! Whew, what a job you are taking on!!! And we will all really benefit from your efforts!!! If everyone entered historical data into eBird, it would start to approach all-time sightings. Yet so many of the wonderful first heroes of birding only kept their records in their field books - hopefully now housed in museums. For a real "first time arrival" (i.e., the earliest recorded arrival),  the Cornell eBird folks would need great funding and an army of data loggers to do this task!

The next record of a Mute Swan was from 2000 which is more like when eBird became a birding tool. It will really be interesting to see how records accumulate as more life-long birders like Jim enter their data into eBird!

The total species recorded this slice was 107 - not too shabby for a small urban park!

The next slice I requested was going to "Arrivals and Departures" and chosing Arrivals. (Hey, if the bird was ever recorded there it had to arrive... right?) This made more sense. The results had the same number of species (107), but it showed (over the same period - 1974 to 2013) the earliest recorded "arrival" (i.e., the earliest date in any year when someone recorded the bird presence into eBird).

I refined it a bit and found that in 2013, so far 47 species are recorded there. Powerful tool!

Yet, I was still uneasy (see previous blog...). I can only query about the Beaudette Park Hotspot to get this data. What if there are personal eBird spots around there? Will the data be combined? The answer is "NO".

So I tried another slice of data. I picked "Point and Range Maps" under Explore Data.

This only works species by species and is another powerful tool. It is what I had seen during the GBBC and was quite fascinated with.

So, I picked a species fairly common at Beaudette Park, but not so common as to overwhelm the data: Hooded Merganser. I kept zooming in until I reached the max zoom with Beaudette centered. The Hotspot flag was there, but also four personal location flags for which "hoodies" had been recorded. YES! That is my answer and complemented the question I was raising before. It allowed me to see the names of the birders who had reported the bird in question on personal hotspots. And - significantly - all four are experienced birders, yet none of their observations had been included on the other "slices" of data I made for the Beaudette Park Hotspot!

So ... I now wonder how many species have really been recorded at Beaudette Park? Maybe it is the same, but I suspect it might be much higher if their personal sites would be merged with the Hotspot!

I know eBird could do a combined "patch" to see the combined total. But I think I can only do "patches" and "yards"  for my own data. One solution is to allow eBird users to create "patches" or data slices by "drawing"  a circle/box/whatever around an area within the Arrivals and Departure tool. I seem to remember this was proposed on eBird Tech Talk awhile back. I also know it will be an intensive (and expensive) software thing to do. But that is what I want. I want all the birds recorded at this park.

Having said that, I realize that this does not address the question I proposed originally in my last blog. It will not help at all at giving a list of all birds at e.g., Stoney Creek Metro Park Nature Center if people are only recording birds as "Stoney Creek Metro Park - Macomb County".  And - see comments in previous blog - I am comfortable at how it works. But, just maybe, if people have the opportunity to draw a circle around an area to combine the data from a trip, they will be more interested in being more specific.

Bottom line:

I think in my last blog, I first had proposed a solution to the wrong question about using Hotspots . Yes, never a solution first - first a question! I think I have  a better idea of what I need, and here proposed the right question. "Ring around an area?"

In the process, I reaffirmed the difficulty of getting people to be more specific. It will always be easier to record birds from a larger area than a specific spot included in the area. I am quite guilty of this myself as well.

Yet, I also learned SO much! Many birders are indeed being more specific! Now an additional question has arisen again about how to combine them for a list of all birds for an avifauna of a special place!!!

Fascinating,  this birding and eBird stuff are ...

1 comment:

Kim Smith said...

You're really going to be an eBird expert soon, Dr. Bob. Like you, I'm sometimes frustrated by having to jump through so many hoops to gather the data I want from their database. I would LOVE the circle option too. Nice post!