January 25, 2020
Following our last lecture, we discuss “Why Geography is Important” centering around the primary problems of pattern and scale.
While there is no single natural scale at which any phenomena should be studied, ecological phenomena tends to have a characteristic spatiotemporal domain, which they should be studied at. Examples include species restricted to rocky tidal zones versus a cougar which ranges across a wide range; and serial sexual offenders tending to be marauders, offending within their own neighbourhood. Scale can also be defined by grain: the minimum resolution of data; and extent: the scope of the data. This is related to the MAUP, where spatial analysis of aggregated data have drastically different results at different scales. Gerrymandering is deliberate exploitation of the MAUP, as voting districts are manipulated to minimize minority representation or to create bias towards one party over another.
One geostatistical method we learned about was Kriging. Kriging was developed to model topography and other natural data that have random fluctuation and measurement error which cause irregularities in sampled data values. This means that it has the capability of producing a prediction surface and provide some measure of the certainty or accuracy of those predictions.
This week we began Lab 1: Exploring Fragstats, which was delayed from last week due to the snow.