Comparing five modelling techniques for predicting forest characteristics

By Gretchen G. Moisen, T. Frescino. 2002

486 citations recorded by the source corpus.

Subfields: Environmental Engineering, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/efd2ee8219c035207d868c77e87051da516d2f19

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Environmental Engineering

In Nature and Landscape Conservation

Citations in the corpus

Can species distribution modelling provide estimates of population densities? A case study with jaguars in the Neotropics (2012, background): “…al., 1984), random forests (RF) (Breiman, 2001), generalized boosting models (GBM) (Friedman, 2001), Multivariate Additive Regression Spline (MARS) (Moisen & Frescino, 2002), mixture discriminant analysis (MDA) (Manel et al., 1999), artificial neural networks (ANN) (Segurado & Araújo, 2004) and…”
Multi-criteria decision analysis in natural resource management: A critical review of methods and new modelling paradigms (2006, background): “…the development of approaches that formally analyse qualitative decision problems such as: artificial neural networks (e.g. Blackard and Dean, 1999; Moisen and Frescino, 2002; Liu et al., 2003), knowledge bases (e.g. Reynolds et al., 1996, 2000), and expert systems (e.g. Store and Kangas, 2001;…”
Newer Classification and Regression Tree Techniques: Bagging and Random Forests for Ecological Prediction (2006, background): “MARS also does not lend itself well to interpretation of species–environment relationships and has been used infrequently in ecosystem science except for mapping certain vegetation characteristics (Prasad and Iverson 2000b; Moisen and Frescino 2002; Munoz and Felicisimo 2004).”
Comparative performance of generalized additive models and multivariate adaptive regression splines for statistical modelling of species distributions (2006, methodology): “…using models capable of fitting non-linear relationships between species occurrence or abundance and environment identified in other studies (e.g., Moisen and Frescino, 2002; Olden and Jackson, 2002; Muñoz and Fellicı́simo, 2004; Moisen et al., this issue) is emphasized again here by the…”
Predicting species distributions from museum and herbarium records using multiresponse models fitted with multivariate adaptive regression splines (2007, background): “…interesting because it gives comparable predictive performance to other nonlinear regression methods — such as generalized additive models (GAMs) — that have been shown to be useful for modelling ecological data, yet MARS is much faster than GAMS (Moisen & Frescino, 2002; Leathwick et al ., 2005).”
Novel methods improve prediction of species' distributions from occurrence data (2006, unknown): “…GAMs, artificial neural networks, and MARS) are comparable in terms of predictive ability, and are often superior to methods such as traditional single decision trees (Ferrier and Watson 1997, Elith and Burgman 2002, Moisen and Frescino 2002, Muñoz and Fellicı́simo 2004, Segurado and Araujo 2004).”
Predicting climate‐induced range shifts: model differences and model reliability (2006, result): “With few exceptions, previous studies have found very little consistency in the performance of these alternative approaches ( Moisen & Frescino, 2002; Robertson et al., 2003; Thuiller, 2003; Segurado & Arau´jo, 2004).”
Spatial models of delphinid (family Delphinidae) encounter rate and group size in the eastern tropical Pacific Ocean (2006, methodology): “In a comparison of modeling techniques,Moisen and Frescino (2002) found that GAMs built on real (as opposed to simulated) data performed marginally better than o rees, a pred e is t ions t stat tial p ., 2 inu with t trees, w rived b , 2 G ironm tion f umb ize n ETP dolphin data supportedAu and…”
Predicting tree species presence and basal area in Utah: A comparison of stochastic gradient boosting, generalized additive models, and tree-based methods (2006, methodology): “Predictive mapping using GLMs and GAMs pertaining to forest inventory applications is illustrated in Moisen and Edwards (1999), Frescino et al. (2001), and Moisen and Frescino (2002).”
Using presence‐only modelling to predict Asian elephant habitat use in a tropical forest landscape: implications for conservation (2010, methodology): “Forest cover data were derived from three Landsat Enhanced Thematic Mapper Plus satellite scenes from 2005 and 2006 using a classification regression tree algorithm (Lawrence & Wright, 2001; Moisen & Frescino, 2002).”
Are niche‐based species distribution models transferable in space? (2006, methodology): “…an identical number of degrees of freedom allowed for each predictor in each technique, we expect GAMs to provide higher fits overall (as shown by Moisen & Frescino, 2002) but potentially at the cost of being less generalizable than GLMs to other situations, that is, showing weaker…”
Constrain to perform: Regularization of habitat models (2006, methodology): “…by comparing the performance of different statistical methods with varying degrees of flexibility like linear discriminant analysis, logistic regression, generalized additive models, artificial neural networks or classification and regression trees(Manel et al., 1999; Moisen and Frescino, 2002).”
Support vector machines for predicting distribution of Sudden Oak Death in California (2005, background): “…Manel et al., 1999), classification and regression tree analyses (De'ath and Fabricius, 2000; Fabricius and De'ath, 2001; Kelly, 2002), genetic algorithms (Stockwell and Peters, 1999), and artificial neural networks (Manel et al., 1999; Spitz and Lek, 1999; Moisen and Frescino, 2002).”