You are here

Keyword: random forest

Improved predictions of deforestation in Borneo

Science Spotlights Posted on: October 12, 2017
A collaborative team, led by RMRS Research Ecologist Samuel Cushman, has produced a substantial breakthrough in advancing predictive modeling of drivers and patterns of deforestation. The method combines multi-scale optimization with machine-learning predictive modeling to identify the drivers of deforestation and map relative future deforestation risk.  

Combined effect of pulse density and grid cell size on predicting and mapping aboveground carbon in fast‑growing Eucalyptus forest plantation using airborne LiDAR data

Publications Posted on: July 19, 2017
LiDAR measurements can be used to predict and map AGC across variable-age Eucalyptus plantations with adequate levels of precision and accuracy using 5 pulses m− 2 and a grid cell size of 5 m.

Multiple-scale prediction of forest loss risk across Borneo

Publications Posted on: May 24, 2017
Context: The forests of Borneo have among the highest biodiversity and also the highest forest loss rates on the planet.

Multi-scale habitat relationships of snowshoe hares (Lepus americanus) in the mixed conifer landscape of the Northern Rockies, USA: Cross-scale effects of horizontal cover with implications for forest management

Publications Posted on: January 10, 2017
Snowshoe hares (Lepus americanus) are an ecologically important herbivore because they modify vegetation through browsing and serve as a prey resource for multiple predators. We implemented a multiscale approach to characterize habitat relationships for snowshoe hares across the mixed conifer landscape of the northern Rocky Mountains, USA.

Shape selection in Landsat time series

Media Gallery Posted on: September 15, 2016
The tree canopy patterns in these time series Landsat map images, taken over a section of the central Rocky Mountains near Steamboat Springs, Colorado, provide information about canopy lost to disturbance events.

Shape selection in Landsat time series

Science Spotlights Posted on: September 01, 2016
There is new methodology for fitting ecologically feasible “shapes” to time series of Landsat imagery for modeling, mapping, and monitoring annual forest disturbance dynamics. Through a case study of fire, harvest and bark beetle outbreak, scientists illustrate how resultant fitted values and parameters can be fed into empirical models to map disturbance causal agent and tree canopy cover changes coincident with disturbance events through time.

Sagebrush scent identifies species and subspecies

Science Spotlights Posted on: August 24, 2016
Big sagebrush (Artemisia tridentata) is the dominant plant species across much of the Western U.S. and provide critical habitat and food for many endemic species, including the threatened greater sage-grouse. Sagebrush habitat is imperiled due to disturbances and increased wildfire frequency due to exotic annual grasses. Identification of big sagebrush subspecies is difficult, but critical for successful restoration. Researchers discover that volatiles emitted by sagebrush species and subspecies differ in consistent ways and can be used to accurately identify plants.

Signals of speciation: Volatile organic compounds resolve closely related sagebrush taxa, suggesting their importance in evolution

Publications Posted on: July 15, 2016
Volatile organic compounds (VOCs) play important roles in the environmental adaptation and fitness of plants. Comparison of the qualitative and quantitative differences in VOCs among closely related taxa and assessing the effects of environment on their emissions are important steps to deducing VOC function and evolutionary importance.

Vegetation, topography and daily weather influenced burn severity in central Idaho and western Montana forests

Publications Posted on: October 06, 2015
Burn severity as inferred from satellite-derived differenced Normalized Burn Ratio (dNBR) is useful for evaluating fire impacts on ecosystems but the environmental controls on burn severity across large forest fires are both poorly understood and likely to be different than those influencing fire extent.

Random forests and stochastic gradient boosting for predicting tree canopy cover: Comparing tuning processes and model performance

Publications Posted on: October 06, 2015
As part of the development of the 2011 National Land Cover Database (NLCD) tree canopy cover layer, a pilot project was launched to test the use of high-resolution photography coupled with extensive ancillary data to map the distribution of tree canopy cover over four study regions in the conterminous US. Two stochastic modeling techniques, random forests (RF) and stochastic gradient boosting (SGB), are compared.