Keyword: logistic regression

Modelling post-fire tree mortality: Can random forest improve discrimination of imbalanced data?

Predicting post-fire tree mortality is a major area of research in fire-prone forests, woodlands, and savannas worldwide. Past research has relied overwhelmingly on logistic regression analysis (LR) that predicts post-fire tree status as a binary outcome (i.e. living or dead). One of the most problematic issues for LR (or any classification problem) occurs when there is a class imbalance in the training data.

Spatial relationship of resident and migratory birds and canopy openings in diseased ponderosa pine forests

A method is described for predicting the spatial distribution of individual birds using presence data. The approach is demonstrated using a statistical habitat association model developed for resident and migratory birds on a 12 ha plot of ponderosa pine (Pinus ponderosa) heavily infested with southwestern ponderosa pine dwarf mistletoe (Arceuthobium vasinatum subsp. Cryptopodum (Englemann) Hawksworth and Weins).

Random forests for classification in ecology

Classification procedures are some of the most widely used statistical methods in ecology. Random forests (RF) is a new and powerful statistical classifier that is well established in other disciplines but is relatively unknown in ecology.

Predicting post-fire tree mortality for 12 western US conifers using the First-Order Fire Effects Model (FOFEM)

Accurate prediction of fire-caused tree mortality is critical for making sound land management decisions such as developing burning prescriptions and post-fire management guidelines. To improve efforts to predict post-fire tree mortality, we developed 3-year post-fire mortality models for 12 Western conifer species - white fir (Abies concolor [Gord. & Glend.] Lindl. ex Hildebr.), red fir (Abies magnifica A.

Predicting post-fire tree mortality for 14 conifers in the Pacific Northwest, USA: Model evaluation, development, and thresholds

Fire is a driving force in the North American landscape and predicting post-fire tree mortality is vital to land management. Post-fire tree mortality can have substantial economic and social impacts, and natural resource managers need reliable predictive methods to anticipate potential mortality following fire events.

Multiple-scale prediction of forest loss risk across Borneo

Context: The forests of Borneo have among the highest biodiversity and also the highest forest loss rates on the planet.

Mountain pine beetle attack in ponderosa pine: Comparing methods for rating susceptibility

Two empirical methods for rating susceptibility of mountain pine beetle attack in ponderosa pine were evaluated. The methods were compared to stand data modeled to objectively rate each sampled stand for susceptibly to bark-beetle attack.

Habitat use by mountain quail in Northern California

We studied habitat use by Mountain Quail (Oreortyx pictus) at four sites in northern California. Vegetative cover types (macrohabitats) were used in proportion to availability. Significant microhabitat variables which distinguished used from available microhabitat structure included proximity to water and tall, dense shrubs.

Seasonal resource selection of Canada lynx in managed forests of the northern Rocky Mountains

We investigated seasonal patterns in resource selection of Canada lynx (Lynx canadensis) in the northern Rockies (western MT, USA) from 1998 to 2002 based on backtracking in winter (577 km; 10 M, 7 F) and radiotelemetry (630 locations; 16 M, 11 F) in summer. During winter, lynx preferentially foraged in mature, multilayer forests with Engelmann spruce (Picea engelmannii) and subalpine fir (Abies lasiocarpa) in the overstory and midstory.