Compare Data-set With List & Summarize Matches

1) Background Info

We are trying to summarise some data that has been exported from an ancient database into a poorly delimited csv file. My colleague has imported the csv file into Excel ( attached), and we are trying to work out what to do next.

The dataset contains approx 300 records - each record being called a "sample" and having a unique sample number. NB: I have had to attach a cut-down version with only 3 samples, due to file-size - but it will hopefully give an idea.

Each sample contains 2 types of information that we are interested in (and a lot of irrelevant data besides). The relevant bits are:
- predicted occurrence of various species (given as a percentage for each species)
- observed occurrence of various species (given as positive/negative for each species, where positive is indicated by an asterisk)

2) What We're Trying To Do

We are trying to compare this dataset with a master-list of 80 species names. For each species on the master-list, we want to:
a) check whether it has a "predicted occurrence" value in each of the 300 samples (and if so, record the value in a summary sheet)
b) check whether it was observed in each of the 300 samples (and if so, record this in the summary sheet)

3) The Problem

Unfortunately, the dataset has imported into Excel as a gigantic list: 48000 rows (including loads of blanks) and only 3 columns across. To find the relevant data, it's necessary to:

a) open the attached workbook, and go to the worksheet named "Data"

b) Scroll through the rows, looking for the string "RIVPA" in column A. This tells you where each new sample begins. (The sampleID is stored in the same row as this, in column C. It is mixed up with a load of text, which we will need to separate out at some point, but that's a secondary consideration at the moment.)

c) Scroll down further until you find the text string "Predi" in column A. This indicates the beginning of the data we're interested in, for each sample (i.e. for sample 1, I'm talking about row 58). Count down a further 2 blank rows, and then you find the data itself:
- Column A contains the observed occurrence (a positive result is indicated by an asterisk)
- Column B contains the predicted occurrence
- Column C contains the species name

My colleague was trying to build a summary table, in the worksheet named "Summary". He was using the LOOKUP function to extract the data, but because there are so many samples, it's beyond unwieldy. He asked me if I could write a macro do do the job, and this is where I'm stuck.

How could I set up a macro that can identify where one sample starts and finishes, and where the data is within each sample? What is the most sensible shape for such a macro? If I could get the overarching logic worked out, I could make a start on writing the individual bits of code, but at present I just can't figure out how to begin.

To make things more difficult, the samples are not all the same length, and do not all contain the same list of species as each other. (However, at least there should be no species in the samples that are not in the master-list). The first sample begins on row 5, the second sample begins on row 173, the third on row 340, and so on.