Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
37 Cards in this Set
- Front
- Back
|
What is a data model?
|
- provides a template for representing the real world in a GIS
- for modeling (spatial) of how the world LOOKS |
|
|
What is a spatial model?
|
- model that represents real PROCESSES
- how the world WORKS - dynamic simulation models (ex. erosion), social processes (ex. movement of cars), optimum alternatives (ex. finding location for a new store) - results of modeling change when the location of objects change |
|
|
What is a time series?
|
- shows the changes that happen to a given landscape over time and processes that occur. ex. deforestation
|
|
|
What is geocomputation?
|
- the applications of computational models to geographic problems
- the way of expressing geographic processes in 0s and 1s |
|
|
What is the difference between analysis and modeling?
|
- Analysis is static (ex. evacuation plans)
-- It leads to hypotheses about systems, looking for patterns one point at a time - Modeling can be dynamic (ex. simulations) -- allows for changing parameters to test hypotheses and evaluate alternative scenarios in multiple stages |
|
|
What are the six types of spatial analysis?
|
1. Queries - to answer simple questions posed by users
2. Measurements - numerical values such as area, shape, distance, and directions 3. Transformations - rasterization, vectorization 4. Descriptive summaries - capture the essence of a dataset in one or 2 numbers (mean, standard deviation) 5. Optimization - select ideal locations for objects give certain well-defined criteria 6. Hypothesis testing - processes of reasoning from the results of a limited sample to make a generalization about an entire population |
|
|
What are static models and indicators?
|
- take multiple GIS inputs of a certain combination and compute useful indices
- indicators are the result ex. Universal Soil Loss Equation predicts erosion from 5 inputs like rainfall and slope |
|
|
What are individual models?
|
- simulate the behaviour of every individual in the system (human systems) ex. every car on the street
|
|
|
What are aggregate models?
|
- used when there are too many individual elements to model (physical systems)
ex. with continuous fluid or sand or water |
|
|
What are cellular models?
|
- model a system using raster
- each cell can be in one of a number of states and can change - depends on state of neighbouring cellls, so if an area around a pixel is deforested then that pixel probably will be too |
|
|
What is the process for calibration?
|
- input creates output
- go to field to validate output - calibrate input data to get this accurate output |
|
|
What is calibration by brute force?
|
- go back to past and then run the model with different inputs and compare with present conditions
|
|
|
What are the four types of GIS operations on a raster?
|
1. Local - determined by the attributes of each cell alone
2. Focal - determined by a cell's neighbours 3. Global - compute properties of the entire raster layer 4. Zonal - apply to all contiguous cells with the same value |
|
|
What are scripts?
|
- sequences of GIS operations that can be stored and shared, written in a scripting language
|
|
|
What is the difference between loose coupling and close coupling?
|
- Loose coupling - the GIS and model exchange data in the form of files
- Close coupling - both GIS and the model read and write to the same file through a common interface (no need for translation) |
|
|
What does Boreal Ecosystem Productivity Simulator (BEPS) do?
|
- models Net Primary Productivity
- mimics plant growth and provides NPP estimates by including inputs like leave area index and land cover, precipitation, temperature, etc |
|
|
What are multicriteria methods and what is one example?
|
- stakeholders often disagree on how to weight different values on a map so multicriteria methods are used to help reconcile differences and reach concensus
- Ex. SAATY's Analytical Hierarch Process - each stakeholder compares each pair of factors and indicates relative importance as a ration - ratings are then combined to produce a consensus set of weights |
|
|
What are three ways to test models?
|
1. Comparison with past history by running a model in the past but forward in time
2. Cross-validation - separate data into 2 using one for calibration and one for validation 3. By experiment - check that it is in line with reality |
|
|
What is sensitivity analysis and how is it done?
|
- used to assess the effects of uncertainty in model parameters
- systematically raise and lower the value of each parameter and observe the effects on model predictions to help identify which parameters are more important Ex. running evapotranspiration with different soil textures and leaf area index to see how they affect each other and to what degree |
|
|
What is the most important input data for hydrological modeling?
|
DEM (Digital Elevation Model) - for looking at slope and elevation to figure out things like run off
|
|
|
What is GIScience ?
|
- computational implementation of geographical concepts and their impacts on society
|
|
|
What is interpolation?
|
- based on an existing point try and figure out what will be at another point using a statistical approach
|
|
|
What is spatial autocorrelation?
|
- Assessment of the correlation of a variable in reference to spatial location of the variable. Assess if the values are related
|
|
|
What is spatial heterogeneity?
|
- the tendency of geographic places and regions to be different from each other.
ex. BC is heterogenous with respect to weather because it is always changing - spatial heterogeneity affects assessment of spatial autocorrelation |
|
|
What are the three important principles for building representations?
|
1. Increased heterogeneity with increased distance
2. some geographic phenomena vary smoothly, others can exhibit extreme irregularity in violation of Tobler's Law 3. High covariance is likely - that patterns of spatial autocorrelation in one variable will be mirrored in another |
|
|
How is spatial autocorrelation determined and what type of values is it assigned?
|
- determined both by similarities in position of spatial objects and by similarities in attributes
- positive = features similar in location are similar in attributes - negative = features similar in location are NOT similar in attributes - zero = attributes are independent of location |
|
|
What is density estimation and how does it work?
|
- density estimation creates a field from discrete objects
- the field's value at any point is an estimate of the density of discrete objects at that point ex. estimating a map of population density (a field) from a map of individual people (discrete objects) |
|
|
What is the Kernel Function? How does it work?
|
- each discrete object is replaced by a mathematical function known as a kernel. Kernels are then summed to obtain a composite 3D surface of density
- Kernel smoothing creates a smooth map of density values in which the density of each location reflects the concentration of points in that surrounding area ( narrow kernels produce bumpy surfaces, wide kernels produce smooth surfaces ) - density of population in relationship to distance from a central point (city centre) versus just per pixel - when the kernel is too small the surface is too rugged and each point generates its own peak, each kernel is isolated from its neighbour, must have wider kernels |
|
|
Draw the following types of samples:
- simple random - stratified - stratified random - clustered - transect - contour |
...
|
|
|
What does Tobler's Law assume with spatial interpolation?
|
- Tobler's Law implies a continuous, smooth, attenuating effect f distance upon the attribute values of adjacent spatial objects
|
|
|
What is the coefficient of determination?
|
R^2 - the proportion of the total variation that is explained by the regression
(dependence of one variable upon another) Ranges from 0.00 to 1.00 0.00 = no correlation 1.00 - perfect correlation - use to create a graph |
|
|
What is the difference between spatial interpolation and density estimation?
|
Spatial interpolation is the estimate of missing parts of a CONTINUOUS FIELD from samples
Density estimation is most often applied to the estimation of point density - creates a clear continuous field from DISCRETE OBJECTS |
|
|
What are the three methods of spatial interpolation?
|
1. Thiessen Polygons
2. Inverse Distance Weighting (IDW) 3. Kriging |
|
|
What are Thiessen Polygons?
|
- network of irregular polygons
- values are constant within each polygon but change sharply when polygon boundaries are crossed |
|
|
What is Inverse Distance Weighting?
|
- the unknown value of a field at a point is estimated by taking an average over the known values, weighting each known value by its distance from the point using Tobler's Law
|
|
|
What is the main problem with Inverse Distance Weighting?
|
- range of interpolated values cannot exceed the range of observed values so ti is important to position sample points to include the extremes of the field which can be very difficult
ex. cannot show the peak of the hill unless you have a point at the peak |
|
|
What is Kriging?
|
- a technique of spatial interpolation grounded in geostatistical theory
- uses a semivariogram: have one fixed point and you meausre the distance between that and other points. That distance is shown as an + on the graph |
