@article{Bradley:2016:10.1016/j.envsoft.2016.04.016,
author = {Bradley, AV and Duarte, Rosa IM and Pontius, Jr RG and Ahmed, SE and Araujo, MB and Brandão, Jr A and Câmara, G and Carnerio, TGS and Hartley, A and Smith, MJ and Ewers, RM},
doi = {10.1016/j.envsoft.2016.04.016},
journal = {Environmental Modelling and Software},
pages = {229--240},
title = {SimiVal, a Multi-Criteria Map Comparison Tool for Land-Change Model Projections},
url = {http://dx.doi.org/10.1016/j.envsoft.2016.04.016},
volume = {82},
year = {2016}
}

Download

TY  - JOUR
AB  - The multiple uses of land-cover models have led to validation with choice metrics or an adhoc choice of the validation metrics available. To address this, we have identified the majordimensions of land-cover maps that ought to be evaluated and devised a SimilarityValidation (SimiVal) tool. SimiVal uses a linear regression to test a modelled projectionagainst benchmark cases of, perfect, observed and systematic-bias, calculated by rescalingthe metrics from a random case relative to the observed, perfect case. The most informativeregression coefficients, p-value and slope, are plot on a ternary graph of ‘similarity space’whose extremes are the three benchmark cases. This plot provides a rigorous similarityassessment against these extremes and other projections. SimiVal is tested on projectionsof two deliberately contrasting land-cover models to show the similarity between intra- andinter-model parameterisations. Predictive and exploratory models can benefit from the tool.
AU  - Bradley,AV
AU  - Duarte,Rosa IM
AU  - Pontius,Jr RG
AU  - Ahmed,SE
AU  - Araujo,MB
AU  - Brandão,Jr A
AU  - Câmara,G
AU  - Carnerio,TGS
AU  - Hartley,A
AU  - Smith,MJ
AU  - Ewers,RM
DO  - 10.1016/j.envsoft.2016.04.016
EP  - 240
PY  - 2016///
SN  - 1873-6726
SP  - 229
TI  - SimiVal, a Multi-Criteria Map Comparison Tool for Land-Change Model Projections
T2  - Environmental Modelling and Software
UR  - http://dx.doi.org/10.1016/j.envsoft.2016.04.016
UR  - http://hdl.handle.net/10044/1/31248
VL  - 82
ER  - 

Download