Land-cover change model validation by an ROC method for the Ipswich watershed, Massachusetts, USA

Pontius, Robert G.; Schneider, Laura

doi:10.1016/s0167-8809(01)00187-6

Cited by 737 publications

(342 citation statements)

References 21 publications

Supporting

Mentioning

329

Contrasting

Unclassified

Order By: Relevance

“…Instead of using the statistical parameter wald to interpret the Logistic regression result, we chose relative wald ratio to quantify the contribution of driving factors for better understanding the driving forces on farmland changes. The ROC was chose for the simulation validation of Logistic regression model for its practicality, effectiveness and wide use (Pontius et al, 2001).…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Farmland changes and the driving forces in Yucheng, North China Plain

Zhao

Fan

2012

J. Geogr. Sci.

View full text Add to dashboard Cite

Taking Yucheng, a typical agricultural county in Shandong Province as a case, this study applied Logistic regression models to spatially identify factors affecting farmland changes. Using two phases of high resolution imageries in 2001 and 2009, the study obtained the land use and farmland change data in [2001][2002][2003][2004][2005][2006][2007][2008][2009]. It was found that the farmland was reduced by 5.14% in the period, mainly due to the farmland conversion to forest land and built-up land, although part of forest land and unused land was converted to farmland. The results of Logistic regressions indicated that location, population growth and farmer income were main factors affecting the farmland conversion, while soil types and pro-curvature were main natural factors controlling the distribution of farmland changes. Regional differences and temporal-spatial variables of farmland changes affected fitting capability of the Logistic regression models. The ROC fitting test indicated that the Logistic regression models gave a good explanation of the regional land-use changes. Logistic regression analysis is a good tool to identify major factors affecting land use change by quantifying the contribution of each factor.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The ROC (Relative Operating Characteristics) curve (Pontius et al, 2001;Pontius, 2002) is used to test the fitness of each regression equation. When ROC value is bigger than 0.7, the regress result is accepted, meaning that the determined driving factors have good explanatory power (Neilson, 1995;Manel et al, 2001;Bai et al, 2004).…”

Section: Logistic Regressionmentioning

confidence: 99%

Farmland changes and the driving forces in Yucheng, North China Plain

Zhao

Fan

2012

J. Geogr. Sci.

View full text Add to dashboard Cite

show abstract

“…To this end, the Cohen (Cohen's Kappa) concordance index was used with success for its simplicity and clarity of interpretation (Carletta 1996;Hagoort et al 2008;Pontius 2000) through which the two compared maps have revealed a fair degree of conformity from both quantitative (number of pixels in the vineyard) and spatial points of view (i.e., checking the exact location of each pixel in both maps): two maps can have the same percentage of pixels for each class of land use, but different spatial allocations of these quantities. The study of spatial inconsistency is complex; it derives from a shift in position between pixels of different categories (Pontius and Schneider 2001). Index K was calculated to be equal to 0.9728.…”

Section: Validation Of Resultsmentioning

confidence: 99%

Use of cellular automata in the study of variables involved in land use changes

Riccioli

Asmar

et al. 2012

Environ Monit Assess

View full text Add to dashboard Cite

The study of changes in land use has been included lately in territorial processes in order to optimize future management decisions. The different functions that the territory plays (production, aesthetic, and natural functions, etc.) make planning choices more difficult. This work focuses on the selection and combination of a set of indicators to analyze the variables through which the changes occur in the land used in viticulture for wine production. The proposed approach makes use of the Geographic Information System (GIS) in the development of a map of land use scenario. It is applied to a case study through a model involving cellular automata (CA) implemented with maps of suitability for viticulture and Markov chains. The use in this case of the CA is aimed at validating the scenario map in order to deduce the variables and the orientations of the farmers in the field of wine production.

show abstract

“…For example, the ROC could be used to compare an image of modelled probability for deforestation against an image of actual deforestation (Pontius & Schneider 2001).…”

Section: Relative Operating Characteristicmentioning

confidence: 99%

Mapping regional forest fire probability using artificial neural network model in a Mediterranean forest ecosystem

Şatır

Berberoğlu

Dönmez

2015

Geomatics, Natural Hazards and Risk

162

View full text Add to dashboard Cite

Forest fires are one of the most important factors in environmental risk assessment and it is the main cause of forest destruction in the Mediterranean region. Forestlands have a number of known benefits such as decreasing soil erosion, containing wild life habitats, etc. Additionally, forests are also important player in carbon cycle and decreasing the climate change impacts. This paper discusses forest fire probability mapping of a Mediterranean forestland using a multiple data assessment technique. An artificial neural network (ANN) method was used to map forest fire probability in Upper Seyhan Basin (USB) in Turkey. Multi-layer perceptron (MLP) approach based on back propagation algorithm was applied in respect to physical, anthropogenic, climate and fire occurrence datasets. Result was validated using relative operating characteristic (ROC) analysis. Coefficient of accuracy of the MLP was 0.83. Landscape features input to the model were assessed statistically to identify the most descriptive factors on forest fire probability mapping using the Pearson correlation coefficient. Landscape features like elevation (R D ¡0.43), tree cover (R D 0.93) and temperature (R D 0.42) were strongly correlated with forest fire probability in the USB region. KEYWORDSForest fire probability and hazard; landscape feature; weighting; artificial neural network; Mediterranean region; fire weather index

show abstract

Land-cover change model validation by an ROC method for the Ipswich watershed, Massachusetts, USA

Cited by 737 publications

References 21 publications

Farmland changes and the driving forces in Yucheng, North China Plain

Farmland changes and the driving forces in Yucheng, North China Plain

Use of cellular automata in the study of variables involved in land use changes

Mapping regional forest fire probability using artificial neural network model in a Mediterranean forest ecosystem

Contact Info

Product

Resources

About