In my analysis I used data obtained from the Government of Yukon, Atlas of Canada via Geogratis, and Agriculture and Agri-Food Canada, including soil texture, effective growing degree days, mean temperature values, a 90m DEM, and the locations of rivers, major roads, and towns. Much of the data was in various projection systems. I ended up using a standard procedure, developed with Alejandro Cervantes' help, whereby after using ArcCatalog to assign layers with their known projection systems I projected them all in ArcGIS to Canada Albers Equal-Area Conic Projection.

Land or  “unchanging” variables

Part of my analysis included the incorporation of what I assumed to be spatially fixed, static variables. Some of the layers which might have been conceptually useful for this ended up being comprehensively restrictive, such as a soil suitability layer from (ldpb), which declared all of the land not suitable for every crop evaluated. This may have either been because of a lack of data which was simply declared as unsuitable, or because the land in the Yukon is relatively unforgiving for agriculture when compared with places like Saskatchewan or interior BC. However, with the knowledge that crops are indeed grown (Ag doc), including potatoes and limited wheat crops, I ignored the soil suitability grade and based my land analysis on raw categories of soil texture, slope, and proximity to water and roads.

My initial data set for the land was:

 
Arc (line) files of Yukon drainage systems from Atlas of Canada Base Maps

Arc files of major roads from Atlas of Canada Base Maps

Soil Inventory (incl texture data) from the Land Potential Database from Agriculture and Agri-Food Canada

90m DEM of the Yukon from Geomatics Yukon, Government of Yukon

These form the basis of the “unchanging” variables which I held constant: Distance from rivers, distance from roads, soil texture, and slope.

Climate variable: Effective Growing Degree Days

 
For EGDDs I constructed a 1990 map of EGDD by ecodistrict using EGDD values from Agriculture and Agri-Food Canada, and combined this with data on mean temperature values by ecodistrict to calibrate a model for approximating EGDD with mean monthly temperature values. Both of these data sets came from Agriculture and Agri-Food Canada.

A note on the scale of my project:

 I recognize that my analysis uses fairly small scale (large coverage area) data and does not necessarily represent detailed, local shifts in soil texture, slope, and other physical characteristics of the land. However, I will compare my results to a map of current agricultural areas provided by the Government of the Yukon for validation and discussion of the accuracy of my analysis. This will allow me to comment both on some of the characteristics of potential Yukon agriculture as well as on the legitimacy of performing such analyses over a broad area.