A major factor which slowed down the compilation of the data was our consideration and comparison of several different ways to geographically partition Long Island Sound (LIS) for mapping and bird reporting. We examined several marine surveys and different types of "grids" (see the post below called "Seabird Data, Locations and Maps") and have finally settled on the following SIMPLE and practical scheme. If you have any questions or further suggestions, please email to me. Here is the proposed approximate partitioning of LIS into nine major sectors, which line up approximately with all the existing CBC Christmas Bird Counts (the green circles) on the CT and NY sides of LIS (except for sector 6 which has no CBC on either its CT or NY sides)...
Of course these 9 sectors were drawn as "oval placeholders" for visual convenience and simplicity, and to communicate the point clearly. These 9 sectors really will be more "rectangular" delineated where possible by longitude lines, with full coverage and PARTITIONING of all of LIS waters, no gaps, and no overlapping areas between them (unlike the approximate oval depictions of them, which are just meant to suggest the concept of 9 sectors)....
 |
| These "oval placeholders" suggest the 9 sectors for reporting. A detailed map with actual sector boundaries is forthcoming. |
Click on this image to enlarge it. Then press Esc key to return.
Green circles are existing Christmas Bird Count areas. Yellow ovals are the proposed nine sectors, for counting and summarizing birds reported in LIS. Of course we can also divide each sector into three sub-sectors, to indicate where in a sector a bird is seen, for example....
5n (sector 5 North, the CT side, closer to New Haven),
5c (sector 5 Center, mid-sound), or
5s (sector 5 South, the NY side, closer to Calverton).
This will give us 9 sectors and 27 sub-sectors, which may be enough granularity for our purposes. With the Old Lyme CBC, for example, we would say that all the LIS birds they report would be assumed within the Sector 7north (the north/CT side of the OldLyme/Orient sector). This seems appropriate, since we have no record of exactly where within the Old Lyme CBC circle these CBC birds were seen in LIS. So lets just say they were seen "in Sector 7north", which is all we actually know, based on the data we received from the CBC.
The nine proposed sectors for LIS are:
WESTERN SECTORS:
1: Bronx-Westchester/NorthernNassau, where both sides are/have CBCs.
2: Greenwich-Stamford/Huntington, where Greenwich-Stamford is a CBC.
3: Westport/Smithtown, where both are CBCs.
MIDDLE SECTORS:
4: Stratford-Milford/RockyPoint, where Stratford-Milford is a CBC.
5: NewHaven/Calverton, where NewHaven is a CBC.
6: Guilford/Mattituck, where neither are CBCs.
EASTERN SECTORS:
7: OldLyme/Orient, where both are CBCs.
8: NewLondon/GardinersIsland, where NewLondon is a CBC.
9: Napatree/Montauk, where both are CBCs.
Our 300 LIS eBird reports (and some other reports) can be much more precise, of course, but it may be useful to aggregate all sightings into these nine sectors (27 sub-sectors) for standardized reporting, for 2013 and future years (to simplify and standardize year-to-year comparisons). We will retain the exact locations of birds (e.g., latitude/longitude, or the exact promontory where birds were seen from) where we know them, especially to facilitate some historical comparisons during future years, but we will create a summary report at this "LCD" lowest common denominator of granularity, the 27 sub-sectors, where possible (and it should be possible in almost every case to do that, for every bird).
The yellow ovals shown on the map above are sector approximations, and the actual sector boundaries will be refined and aligned to the CBC circle limits, where there are CBCs. Every square inch of LIS will be covered by one sector. Note that much of sector 8 and especially sector 9 are outside our LIS study area, but we are defining those sectors for possible future comparisons (we use the long-standing Montauk CBC, in sector 9, as a benchmark for multi-decade comparisons, especially because it is at the ocean gateway into LIS).
These locational sectors will be used in our database for our LIS Bird Count, but we are not currently planning to plot species reports on an actual visual map of LIS, largely because the locational precision of our data is too heterogeneous, and because the GIS work would be too demanding, requiring ArcGIS etc. If anyone would like to take on this piece of work, please let me know.
The heterogeneous nature of our data (combining eBird data, with CBC data, with ad hoc reports, with GPS precision cruises across LIS, with several ferry rides, with longitudinal time-series data from Shippan Pt!, etc) is still a challenge, and we are working on the best ways to deal with it.
