Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

Normal Size Small Size show me how

Normal Size Small Size show me how

# Geog 385 Final

Question | Answer |
---|---|

Precision* | exactness of measurement/description |

Accuracy * | closeness of recorded measurements to RL |

Inference | the processes of deriving the strict logical consequences of assumed premises; concerned with making legit. inferences about underlying processes from observed patterns |

Statistical significance (18) | doesn't always equate to scientific significance: with a big enough sample size, statistical significance is often easily achieved. |

3 Types of Spatial Distributions: 1. (18) | 1.Random: any point is equally likely to occur at any location, and the position of any point is not affected by the position of any other point |

3 Types of Spatial Distributions: 2. (18) | 2.Uniform: every point is as far from all of its neighbors as possible; "unlikely to be close" |

3 Types of Spatial Distributions: 3. (18) | 3.Clustered: many points are concentrated close together, and there are large areas that contain very few, if any, points: "unlikely to be distant" |

Mean Center (18) | mean of x and y coords.; called center of gravity or centroid. Minimizes sum of squared distances between itself and all points. Ex: on map equal weights placed at every residence (line of points) |

Median Center (18) | The point which minimizes sum of distances between itself and all other points (point of minimum aggregate travel). No direct soln.; multiple points may meet the criteria. Ex: on map intersection of N/S and E/W line drawn so 1/4 ppl live in each quadran |

Standard Deviation Ellipse (18) | Std. dist. dev. = good single measure of the dispersion of the incidents around the mean center, but it doesn't capture any directional bias; no shape. Ellipse gives dispersion in 2 dimensions |

Spatial autocorrelation (18)* | Tobler's 1st law: near things more related than distant things. Correlation of a variable with itself through space. Correlation between an obs' value on a var. and the value of close0by obs. on the same var. |

Spatial autocorrelation 2 (18)* | The degree to which characteristics at one location are similar to those nearby. Measure of the extent to which the occurrence of an event in an areal unit constrains, or makes more probably, the occurrence of a similar event in a neighboring areal unit. |

Moran's I (18)* | Varies between -1 and 1. 0=no spatial autocorr.; Neg/pos. values indicate neg/pos autocorr. |

Tin Z Tolerance (20) | The maximum allowable difference in z-units between the height of the input raster and the height of the output TIN at any particular location |

Aspect (20) | A directional measure of slope |

Slope (20) | Rate of change of elevation at a surface location |

DEM Sink (or Pit) (20) | Caused by: random errors in source DAM, discontinuities between DEM layers, artifacts of resampling. Cells have a lower value than surrounding cells; water flow stops at the sink |

Spatial Interpolation (21) | procedure of predicting the value of attributes at unsampled sites from measurements made at point locations within the same area. --trying to predict values at some point by using points @ samples location |

Exact interpolator (21) | an interpolator which shows the exact values of the data points (as opposed to approximate/inexact interpolator) |

Global interpolator (21) | Method where all the data points are used to estimate a field |

Local interpolator (21) | Methods which use some subset of data points to locally estimate a field |

Anisotropy (21) | A random process which shows different degrees of autocorrelation in different directions (directional autocorrelation) |

Why use spatial interpolation? (21) | -can't sample every spot in study area -time -money -accessibility -safety -data loss/error/missing records |

Interpolation (22) | prediction within the range of data |

Extrapolation (22) | prediction outside the range of data |

Kriging Requirements* (22) | Assumes spatial variation in variable is too irregular to be modeled by simple smooth function, better with stochastic surface -Requires spatial dependence -Requires 2nd order stationarity (mean and var of sample data must remain invariant in space) |

Deterministic (23) | -Processes whose outcome can be predicted exactly from knowledge -many times can be mathematically described -outcome always the same |

Stochastic (23) | -Processes whose outcome is subject to variation that can't be given precisely by math. formulae -intro. of a random (stochastic) element to model the range of potential solutions -chance process with well-defined mechanisms |

Analog Model (23) | -usu a scaled-down representation of the real world -analogous to actual processes; issues of scale and resolution -ex: wind tunnels, wargarmes, orrery |

Cellular Automata (CA) Models: definition (23) | -Framework for systems experiments -Simplest way to demonstrate complex systems behavior -breaks down interactions into discrete cells that have a set of rules governing their behavior -maps well to GIS raster framework |

Cellular Automata models (23) | 1. set initial conditions: each cell in one of a number of states 2. rules of state transition: determine new cell state at each time step based on states of cell and neighbors |

Cellular Automata: LIFE (23) | -Simulation -each LIFE cell has two states:LIVE/DEAD -each cell exists in neighborhood of 8 cells -count # live neighbors per cell; dead cell w/ 3 live neighbors=live; live cell w/2+ live stays live; otherwise cell=dead |

Types of Weights Matrices: Contiguity (24) | A "neighbor" is defined based on common borders or common corners: -Rook: common borders -Bishop: common corners -Queen: common borders and corners |

Types of Weights Matrices: Distance (24) | A neighbor is defined based on its distance (point to point, centroid to centroid) from each spatial unit |

Types of Weights Matrices: k-Nearest Neighbors (24) | Uses distance but counts only the "k" nearest neighbors |

Weights Matrices: Different representations (24) | Binary: 1 if a neighbor, 0 otherwise Row Standardized: all elements of a row sum to 1 => most common |

Sparse Matrices (24) | -Most weights matrices are sparse; majority of elements have a value of zero. A county in OR would likely not be considered a neighbor of a county in TX -Special SM methods used by software for modeling (R and MATLAB) |

Created by:
Capt_Picard