|Simple, but it made me very happy.|
Saturday, 4 August 2012
EMS project: Becoming reacquainted with my friend GIS.
I can't tell you how long I've been wanting to start a little 'diary' of my analytical project that is currently underway. Many moons ago [it seems!] I stated that I had to do a research project as required by my MSc and this was to be done with an external organisation. As I've mentioned a fair few times, I chose to do mine with the EMS management team, specifically working with INCA, Natural England, and the RSPB. I've been officially working on the project for a month now though, unfortunately, have been getting nowhere. I had to wait for data to be sent, I had issues installing my good friend ArcGIS onto my laptop (I even bought myself a new laptop for this purpose), and then there was the issue of me actually becoming reacquainted with the programme. Phew!
My main point of contact for the project has been Maeve, the current EMS project officer and she's brilliant. I think she's received many e-mails/texts/rambles fuelled by frustration, coffee, and on one occasion beer, asking for advice on how to get the data into GIS. I was just not having a good time with it. Essentially I need to create separate shapefiles for each species, both at the start and the end of the bird count surveys. This way I can map abundances. I'm not sure what I'm going to do with the end counts just yet, I think that'll be another "please give me some advice" e-mail, but at present I was just using the 'average' values to get a general idea of the prevalence of bird species at each of the six sites (Hartlepool North Sands, North Gare, South Gare, Redcar Rocks, Seaton Snook, and Bran Sands). Then, alongside this, I'm going to display their likely activity (either feeding or roosting) based on the sum of averages. It gets a little complicated so it's just going to be as a % representation (I think, I'll see how it goes).
The other wall that I hit was to do with getting this data actually into GIS. Those who know me, I'm always trying to better what I've done. I'm never happy with it - perfectionism, it's a blessing and a curse at the same time. Anyway, the way that the previous EMS officer, Katherine, had input the data was manually, inserting each value into each field one by one. Given that I am playing with what, about 600 points per species (if you consider that there are about 10 fields per value as well) I wasn't too happy with manually inputting this data, one by one. Cue a lot of googling and e-mailing GIS 'experts' to try and see whether I could do this. The only solution I could think of was saving a spreadsheet as a database (thank goodness for Open Office, Microsoft Office doesn't do the database extension anymore) and importing this into GIS via the XY (coordinates). The data I had been given didn't have coordinates, and that is why the data had previously been input manually so that each record could be placed where it needed to be on the map.
A few days of wanting to throw my laptop out of the window, I had a solution. I grabbed the XY coordinates off GIS for each of the six sites, assigned them to the appropriate values, and then imported the data in and just smushed around the values myself. Looking back I probably should have done this first off but clearly my noggin didn't want to cooperate.
As a little bit of GIS-practice, I wanted to recreate the European Marine Site (EMS)/ Special Protection Area (SPA) boundary map that I had found. I gathered the SPA data from the JNCC website, opened up my two 1:50,000 OS maps downloaded off Digimap and had a play around. I couldn't find any EMS GIS data and I didn't fancy recreating the polygon shapes myself so I made a duplicate SPA-boundary file and edited the vertices to delete any that weren't part of the EMS. It all sounds very complicated but I knew what I was doing. Threw on a title, a legend, a North arrow, and a scale bar and voila. My first little map produced for a while.
So, now that I've had a really good play around with this and that (pressing far too many buttons for my liking and probably wrecking havoc to the data - only messing, I know what I'm doing!) I feel like I at least have a little bit of knowledge and confidence that I can work with the data. It just took a little while to get going, that's all.
I guess this kind of reiterates that old mantra, "never give up". It's true, you'll get there in the end.