DOI: 10.1016/j.foreco.2017.05.038
Scopus记录号: 2-s2.0-85019972519
论文题名: Predicting post-fire tree mortality for 14 conifers in the Pacific Northwest, USA: Model evaluation, development, and thresholds
作者: Grayson L.M. ; Progar R.A. ; Hood S.M.
刊名: Forest Ecology and Management
ISSN: 0378-1127
出版年: 2017
卷: 399 起始页码: 213
结束页码: 226
语种: 英语
英文关键词: Classification errors
; Logistic regression
; Modeling
; Post-fire tree mortality
; Scott guidelines
Scopus关键词: Fires
; Forecasting
; Managers
; Models
; Regression analysis
; Classification errors
; Logistic regression models
; Logistic regressions
; Mortality threshold
; Pseudotsuga menziesii
; Scott guidelines
; Tree mortality
; Western United States
; Forestry
; accuracy assessment
; beetle
; coniferous tree
; error analysis
; forest fire
; forest management
; model validation
; morphology
; mortality
; natural resource
; regression analysis
; threshold
; Pacific Northwest
; Abies amabilis
; Calocedrus decurrens
; Chamaecyparis lawsoniana
; Coniferophyta
; Larix occidentalis
; Picea engelmannii
; Pinus contorta
; Pinus ponderosa
; Pseudotsuga menziesii
; Scolytinae
; Thuja plicata
; Tsuga heterophylla
英文摘要: 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. Current fire mortality models are limited to a few species and regions, notably Pinus ponderosa and Pseudotsuga menziesii in the western United States. The efficacy of existing mortality models to predict fire-induced tree mortality is central to effective forest management. This study validated 54 logistic regression mortality models from seven published articles and two sets of mortality guidelines from two sources. Survival and a suite of fire injury metrics were monitored for 3654 trees representing 14 species that burned in fires between 2002 and 2009 in the Pacific Northwest, USA. Tree species included Abies amabilis, A. concolor, A. grandis, A. lasiocarpa, Calocedrus decurrens, Chamaecyparis lawsoniana, C. nootkatensis, Thuja plicata, Pinus contorta, P. lambertiana, P. monticola, Picea engelmannii, Larix occidentalis, and Tsuga heterophylla. Existing logistic models adequately described post-fire mortality of A. concolor, A. lasiocarpa, C. decurrens, C. lawsoniana, L. occidentalis, P. engelmannii, P. contorta, and P. lambertiana. We also evaluated predictive accuracy of two published mortality guidelines that apply to species in the Pacific Northwest. In addition to validating existing models, we also developed new logistic regression models and simplified mortality guidelines, or thresholds. We created new logistic regression models for species with adequate sample size and which had no existing species-specific model (A. amabilis, A. grandis, P. monticola, and T. heterophylla). Most recommended models contained a crown scorch term and either a cambium injury term or a bark beetle infestation term. New post-fire mortality thresholds were developed for A. amabilis, A. concolor, A. grandis, P. contorta, P. lambertiana, P. monticola, P. engelmannii, L. occidentalis, and T. heterophylla. We were not able to validate or develop acceptable logistic mortality models or thresholds for C. nootkatensis or T. plicata. Injury to cambium and crown were both significant predictors in all but one set of new thresholds. The validation of existing models and guidelines allows managers to determine which models will likely perform best and identifies knowledge gaps where no adequate models exist to predict post-fire tree mortality. The new logistic regression models and threshold guidelines provide improved accuracy, with simpler application for fire and forest management. © 2017 Citation statistics:
资源类型: 期刊论文
标识符: http://119.78.100.158/handle/2HF3EXSE/64276
Appears in Collections: 影响、适应和脆弱性
There are no files associated with this item.
作者单位: USDA Forest Service, Rocky Mountain Research Station, Fire, Fuel, and Smoke Science Program, 5775 US Highway 10 W, Missoula, MT, United States; USDA Forest Service, PNW Research Station, La Grande Forestry and Range Sciences Laboratory, 1401 Gekeler Lane, La Grande, OR, United States
Recommended Citation:
Grayson L.M.,Progar R.A.,Hood S.M.. Predicting post-fire tree mortality for 14 conifers in the Pacific Northwest, USA: Model evaluation, development, and thresholds[J]. Forest Ecology and Management,2017-01-01,399