词条 | GeoMod |
释义 |
Simulation inputsGeoMod simulates land change based on a combination of several input requirements. First, users must specify the beginning (time 1) and end time (time 2) for the temporal extent of the simulation and must identify a time step for the simulation.[2] The model also needs the image that depicts the two land categories (e.g. category 1= undeveloped and category 2=developed) at time and a projected quantity of the two categories at time 2.[1] GeoMod also needs a suitability map, which shows the transition suitability of each pixel. The higher values in the suitability map, the more suitable the pixel is for transition the gain of category 2. If the user does not have a suitability map, GeoMod can create one using one or more driver images. A driver image is a categorical image that indicates the distribution of a variable that is believed to influence the change of the land-cover categories.[3] GeoMod can separate the analysis into strata, such as political units, and then simulate change independently within each stratum.[1] If the quantity of change from time 1 to time 2 indicates a net gain of category 2 in a particular stratum, then GeoMod assumes zero gross loss of category 2 during the simulation, meaning GeoMod does not simulate simultaneous gain and loss of an individual category within an individual stratum. GeoMod allocates change based on several user-defined decision rules, which are discussed in section 3. Optional inputsIn addition to the mandatory simulation inputs listed above, there are several additional optional data inputs.[2]
Environmental impact analysisGeoMod can also analyze the environmental impact on the pixels that undergo change within a specified time interval. This optional feature requires an image showing the environmental resource of interest, an image showing the ratio of the potential impact to the environmental resource of interest, and an image showing the ratio of simulated impact to potential impact.[2] If the latter two images are not available, a fixed ratio for the entire study area may be used. GeoMod decision rule for allocating changeDecision rule 1Decision rule 1 is mandatory and assumes a one-way change either forward or backwards in time within each stratum. The model determines which category experiences a net increase and then simulates gross gain in that category and zero gross loss of that category.[1][2] Decision rule 2Decision rule 2 is optional and it concerns regional stratification. Rule 2 can allow for change from category 1 to category 2 in some strata and from category 2 to category 1 in other strata. When using regional stratification, the user has to specify the quantity for each category at time 2 in each stratum. Decision rule 3Decision rule 3 is also optional and it focuses on the neighborhood constraint. GeoMod can geographically limit the simulated change to pixels that are on the edge between category 1 and category 2.[4] The model can thus apply a user-defined minimum search width to constrain where simulated change occurs. Decision rule 4Decision rule 4 is also optional and concerns a suitability map. When simulating the transition from category 1 to category 2, GeoMod simulates change from the category 1 pixels that have the largest suitability values.[1] ValidationPattern validation shows how the simulated change compares to reference change, for cases where a reference map is available for validation. A visual approach makes use of the CROSSTAB module in TerrsSet by comparing three maps simultaneously: Reference time 1, Reference time 2, and Simulated time 2.[2] Additionally, the Total Operating Characteristic (TOC) curve may be used to compare the suitability map to a map of reference change. Terrset has a module called ROC, which can assist in the selection thresholds for the TOC.[5] References1. ^1 2 3 4 5 Pontius, R. G., Cornell, J. D., & Hall, C. A. (2001). Modeling the spatial pattern of land-use change with GEOMOD2: application and validation for Costa Rica. Agriculture, Ecosystems & Environment, 85(1), 191-203. 2. ^1 2 3 4 5 Pontius Jr, R. G., & Chen, H. (2006). GEOMOD modeling. Clark University. 3. ^Dushku, A., & Brown, S. (2003, October). Spatial modeling of baselines for LULUCF Carbon projects: The GEOMOD modeling approach. In 2003 International Conference on Topical Forests and Climate Change:" Carbon Sequestration and the Clean Development Mechanism (Vol. 39). 4. ^de Benito, B. P., & de Peñas, J. G. (2008). Greenhouses, land use change, and predictive models: MaxEnt and Geomod working together. In Modelling environmental dynamics (pp. 297-317). Springer Berlin Heidelberg. 5. ^Batchu, K., & Pontius, R.G. (2003). Using the Relative Operating Characteristic to Quantify Certainty in Prediction of Location of Land Cover Change in India. Trans. GIS, 7, 467-484. Further reading
1 : GIS software |
随便看 |
|
开放百科全书收录14589846条英语、德语、日语等多语种百科知识,基本涵盖了大多数领域的百科知识,是一部内容自由、开放的电子版国际百科全书。