Update 2024-03-12-gis.md

Started doing alt-text, need to finish up some of them tomorrow

bom-website/content/blog/2024-03-12-gis.md

@@ -18,7 +18,7 @@ This blog post will not be an analysis of that data and those maps I made, but a
 
 Step one was to establish a data set to map. For me, it was easy enough. I was already working with the 1656-1680 general bills for some research I did into why these years had more deaths than christenings every year. That does not mean step 1 was complete, though. I needed to finish transcribing the set, which took some time. I can do a typical weekly bill in around 7-10 minutes, but general bills are different from weekly bills. The parishes moved much more on the document, and many disappeared entirely for a few years after the Great Fire. Therefore, I had to be diligent to ensure that the parishes were actually missing from the bill, not that I was just overlooking them. Once I had the full set transcribed and exported, my job was still not done. QGIS (the free mapping software I used) cannot read this document and map it:
 
-{{< figure src="/images/ward-image4.png" caption="" alt="" >}}
+{{< figure src="/images/ward-image4.png" caption="" alt="A screenshot of Google Sheets showing a sheet titled 'BOM Important Generals 1656-1680 by Parish' The X-Axis shows each years, parish, and whether the number represents that parish's burials or plague deaths. The Y-Axis shows the unique identifier for each bill. Burial parish column headers are highlighted in grey, while the row for bill 'Millar-1665-1666' is highlighted in yellow" >}}
 
 There are a few mini steps I need to do to make this usable by QGIS. First, I need to decide which information is important from this long document of stats (you only see the beginnings of the A-parishes, there are MANY more parishes that require scrolling to). I decided I wanted yearly maps, with parish names, and burials, but not cumulative burials for groupings of parishes. The best way I found for getting the data I wanted without destroying this original document showing me numbers for all categories was...transcribing all of it over again, for the most part.
 To re-transcribe this data, I would hide all columns other than burials and the year, and created a whole new spreadsheet with one year and all deaths per parish, but I had to do some fiddly work with the parish names. For QGIS to read my data and the shapefiles together, I would need to have at least one column on each with the exact same entries for a clean join. So I went into a random test map using our shapefiles and found a category to copy. Here are the steps I took:
@@ -29,7 +29,7 @@ To re-transcribe this data, I would hide all columns other than burials and the
 - Check Attribute Table: Go to the Layers panel on the side > Right-click on “WithinTheBills1622” > Open Attribute Table.
 - Choose joinable column. You'll likely want to differentiate between parishes if you’re mapping, so I would choose one of the parish name columns. There are 3 different parish name columns on these shapefiles. I chose the DNB_PAR column, since I knew that DBN (Database Name) values were unique.
 
-{{< figure src="/images/ward-image9.png" caption="" alt="" >}}
+{{< figure src="/images/ward-image9.png" caption="" alt="A screenshot of the attribute table for the 1622 shapefiles. The X-axis shows name and category labels, while the Y-axis shows each parish's delineations based on the X-axis" >}}
 
 - Open an empty spreadsheet. I’d recommend naming it something like “1622 Mapping Template,” but that’s up to you and how organized you are. Then I made two columns: “DBN_PAR” (just like it’s written in the attribute table. It needs to be identical) and “BURIALS”. Then I copied every DBN_PAR item in the column to the matching spreadsheet column, making sure I had the right amount of rows compared to the attribute table. (I used ctrl-c and ctrl-v to make sure the names were identical to their original names. It doesn’t matter if they’re in the right order, but it is important that the names are the exact same.) You should have something like this: