Repost from Misc. forum. Maybe it belongs over here instead:

Hello all,

I have a large dataset in excel that I need to sort and rank b
abundance. It includes information on the abundances of each of 15
different plant species at each of eight locations, during si
different sampling periods (3 years, fall and spring) with twent
replicates per site (plot number). There are separate site, season
year, and plot columns to identify individual records, as well as on
column for each species.

I need to create rank-abundance profiles for each replicate site durin
each sampling period. For further clarification on what these are, do
google search for "species rank abundance plot". There is an image tha
comes up and many web pages that explain the concept. To do this, I nee
to sort the species data in each row by rank-within-row from mos
abundant to least abundant from left to right. There will be a lot o
zeros on the right since most plots only have 5 - 20 of the 154 specie
in them. The abundance values don't need to be linked with the specie
name headers anymore. I hope that part was clear.

Then I want to calculate the proportion of of the row total abundanc
that each species contributes. Then (it goes on), I want to calculat
mean rank-abundance profiles for each site by sampling dat
combinations by averaging across the 20 replicate plots. I want to d
this part twice - once with the proportion data, and once with the ra
data. Here's how the data is organized. There are 960 rows:

Site Season Year Plot# Species1 Sp2 Sp3 Sp4.... Sp154
A Spring 2003 1 0 5 2 15
A Spring 2003 2 5 0 1 3