The January 16 chart generated some questions. It might be useful to look a little wider range. But how to divide things into useful numbers? What are appropriate time frames? Where to start? After some fiddling I came up with some breaks in the data that fit on the same charts. While not equal, hopefully useful.
The time frames are 1) everything prior to 1960, this covers 80 years and would include the survey efforts in the late 1950's; 2) everything in the years 1960-1999, this covers 40 years and the first statewide survey of the 1990's; 3) everything from the years 2000-2009, basically the quiet 10 years while our iNat data is still limited; 4) everything in the year 2010-2019, this covers 10 years and the bulk of the official statewide survey; and 5) the last 3 years.
Two charts, first on total observations per month across our 5 time spans, then distinct species per month, again across the 5 time spans.
The observation data is a stacked area chart. Remember these are total values for period with a starting representation of 80 years, ending with the most recent 3 years. Compelling evidence that we're doing well on overall observations.
The species data is a simple bar chart. These are the distinct species recorded during months across each period.
In all cases we see activity beginning in March, ramping quickly in late May to peak in Jun, then taper off steadily after July.
It's interesting that our species totals track fairly consistently. Even the period with the fewest observations (the thin green slice) was not far off the other periods. The biggest discrepancy in species numbers is October, perhaps it's just we're out more late in the season in recent years.
Given that we're only seeing (in recent years) about 80% of the documented species, I thought there would be months where the first 2 year spans would have the most species. Not seeing that. Only July has a peak species value for a period before 2000. If there is loss of some species, diversity expression is supplemented by the arrival/discovery of new species.
A couple other random things that turned up as I was thinking about all this:
iNat observations begin to dominate annual numbers in the OOS db starting in 2005. These are all relatively recent additions - this project didn't start until 2017. iNat observations now account for over 80% of the OOS data.
Widow Skimmer was the first species recorded in all 88 counties - that happened in 1991. 22 other species now have records from all 88 counties.
Blue Dasher was the first species recorded in all 88 counties in a year, 2018. 6 other species now have that distinction (Common Whitetail, Eastern Amberwing, Eastern Forktail, Eastern Pondhawk, Fragile Forktail, Widow Skimmer).
Hardin Co was the last county to enter the data set. 3 species were collected on Aug 8, 1958. Several others were added on Aug 9. Hardin Co will be one of the target counties for 2023.
Comentarios
So much good data. Thanks for doing this stuff Jim.
Thank you so very much for taking the time and effort to make such an interesting analysis! I really appreciate your work!
Jim: Thanks again for the analysis. You note the fairly stable number of species over time. I'm impressed that that is the case despite the much greater number of observations in recent years. You'd expect more observations to reveal more species. especially rare ones. Any thoughts on why that isn't the case? The older time frames encompass more years; is it simply that rarer species only show up sporadically, so longer time spans are more likely to pick them up?
Hi David, I picked time periods that seemed to make sense, given the data. As you suggest, this maybe isn't the whole story.
If we pull out numbers based on individual years rather than aggregating, my neat alignments aren't so strong. But, how to compare?
Prior to 1950, every 3 observations recorded a species on a yearly average - not a lot of collecting. In the 1950's, that rate doubled. Then doubled again in the 1990's. So in the 1990's every 11 observations recorded a species, an increase in collecting. In recent years (boom) the observation to species rate is 200 - lots of photos. Following along - we're seeing most of the species that could be expected, including rare things.
More - there are records for 173 species in the Ohio data. 14 of these species have fewer than a dozen records - and several are singles. So, out of 131 years, 14 species are seen maybe once every 10 years. Some of these were likely vagrants - blown or carried in. Non-residents. If we double the recorded records threshold to 24, the number of rare species also doubles, another 15 sp. These are species that are likely here, but so uncommon it takes lots of attention, and/or luck .
So if we reduce our 173 by the 29, we get 144 species that are likely to be encountered on a statewide survey over a reasonable time frame. This is roughly (146) what we have for the first statewide survey (1990's). It's also what we've seen (147) in the last 2 years.
Basically, still thinking on the story.
Agregar un comentario