[[File:Portland3HullsLowest.png|right|300px]] An obvious alternative approach is to use the first time (lowest level) that there are three hulls. There is a big difference in the layer numbers for this. See the queries below. Essentially, the HCA algorithm often takes an original hull apart and then takes the resulting hulls apart, giving a quadratic for hull count again layer. But, as is apparent in the image, this leads to big areas that would only be good for overlap analysis and not for identifying individual clusters.
--Three hulls (or fewer) first Lowest,highest, and last time (i.e.lowest-highest, lowest and highest layer)first-after-peak level where there are three hulls
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='2015' ORDER BY year,layer;
--3, 41, 37, 33 (peak at 22,23,24,25)
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='2010' ORDER BY year,layer;
--3,1724, 24, 24 (peak at 18)
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='2005' ORDER BY year,layer;
--7,23, 21, 21(peak at 18)
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='2000' ORDER BY year,layer;
--3,17, 17 , 17 (peak at 12,13,14)
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='1995' ORDER BY year,layer;
--1(1),8(1), 8(1), 8(1) (no peak-just flat but still take the highest layer with the nearest lower number of hulls)
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='1990' ORDER BY year,layer;
--2(2),5(2), 5(2), 5(2) (peak at 2,3,4,5 so 5 is first 'after' peak? but still take the highest layer with the nearest lower number of hulls)
SELECT * FROM DisplayHullsFull WHERE place='Portland' and statecode='OR' AND year='1985' ORDER BY year,layer;
--1(1), 1(1), 1(1), 1(1) (peak at 1 as that's the only layer with a hull but still take the highest layer with the nearest lower number of hulls) [[Portland3HullsLowestHighest.pngright|300px]] Two further options are to find the lowest-highest and the first-after-peak. These often coincide. The lowest-highest finds the highest layer with x hulls and then works back down the layers taking the lowest in the continuous chain of x hulled layers. The first-after-peak finds the first layer with x hulls in or after the layers where there is the peak number of hulls. This last approach is a little problematic because sometimes there isn't a peak - its just flat - and it is inconceivable that there could be two or more peaks.
===Image Analysis===