8. Suppose the input layer shows a county and the overlay layer shows a national forest. Part of the county overlaps the national forest. We can express the output of an Intersect operation as [county] AND [national forest]. How can you express the outputs of a Union operation and an Identity operation?

Union: [county] OR [national forest]

Identity: ([county] AND [national forest]) OR [county]

9. Define slivers from an overlay operation.

Slivers are very small polygons along correlated or shared boundary lines (e.g., the study area boundary) of the input layers on an overlay output.

10. What is a minimum mapping unit? And, how can a minimum mapping unit be used to deal with the sliver problem?

A minimum mapping unit represents the smallest area unit that will be managed by a government agency or an organization. It can deal with the sliver problem by simply removing slivers that are smaller than the minimum mapping unit.

11. Although many slivers from an overlay operation represent inaccuracies in the digitized boundaries, they can also represent inaccuracies of attribute data (i.e., identification errors). Provide an example for the latter case.

[An example of identification errors is the misclassification of land use types, which can result from interpretation errors or from errors in data entry.]

12. Both nearest neighbor analysis and Moran’s I can apply to point features. How do they differ in terms of input data?

Nearest neighbor analysis uses only distances between points as inputs. Analysis of spatial autocorrelation, on the other hand, considers both the point locations and the variation of an attribute at the locations.

13. Explain spatial autocorrelation in your own words.

[Spatial autocorrelation measures the relationship among values of a variable

according to the spatial arrangement of the values. The relationship may be described as highly correlated if like values are spatially close to each other, and independent or random if no pattern is discernable from the arrangement of values.]

14. Both Moran’s I and the G-statistic have the global (general) and local versions. How do these two versions differ in terms of pattern analysis?