Empirical modeling of cutblock edge windthrow risk on Vancouver Island, Canada, using stand level information

doi:10.1016/S0378-1127(00)00620-4

Forest Ecology and Management

Volume 154, Issues 1–2, 15 November 2001, Pages 117-130

https://doi.org/10.1016/S0378-1127(00)00620-4 Get rights and content

Abstract

Winter storms routinely damage stands along clearcut boundaries on Northern Vancouver Island, disrupting forest management plans. Empirical modelling of windthrow risk is more suitable than mechanistic modelling for the western hemlock (Tsuga heterophylla (Raf.) Sarg.), redcedar (Thuja plicata (Donn)) and amabilis fir (Abies amabilis Forbes.) stands in this area which vary in composition, structure, density and age. Cutblock edge windthrow in a 900 km² study area was mapped using 1:15000 scale aerial photographs. A geographic information system (GIS) database for 3000–50 m long by 40 m deep cutblock edge segments was assembled using readily available topographic and inventory information. Logistic regression models for predicting probability of damage for various levels of damage severity were evaluated. Key variables in these models included site quality, stocking, boundary orientation, time since harvest and topographic exposure. The models correctly predicted the outcome of 71–76% of test segments, and adequately fit using χ² tests for test data grouped by stand characteristics. These models can be entered into GIS map calculations to produce landscape level maps of windthrow hazard, or into spreadsheets for evaluating the probability of damage for alternate cutblock design scenarios.

Introduction

Windthrow refers to the uprooting or breakage of trees by wind, and results from the interaction of climatic, topographic, stand, tree and soil factors. The risk of damage is increased by management interventions such as thinning or clearcutting which increase the exposure of residual trees to wind (e.g. Harris, 1989, Ruel, 1995). Results from a province wide census by the British Columbia (BC) Ministry of Forests indicated that a volume of timber equivalent to 4.2% of BC’s annual allowable cut was windthrown in 1 year. This damage disrupts stand and forest level management plans (Mitchell, 1995).

Documenting the relationship between windthrow occurrence and biophysical and management factors enables damage prediction and development of lower risk harvesting plans. The different approaches to windthrow risk assessment can be broadly classified as: observational, mechanistic or empirical. In observational methods, the presence of factors known to be associated with higher incidence of damage is tallied. The relative risk of windthrow is considered to increase with the number of risk indicators observed (e.g. Ruth and Yoder, 1953, Alexander, 1964, Harris, 1989, Stathers et al., 1994). Mechanistic models predict the likelihood of damage based on an evaluation of the critical windspeed for tree failure, and the probability of such a wind occurring at a given location. These models are built by winching trees to determine resistance, placing tree crowns in wind tunnels to determine drag, and using wind tunnel, or numerical modelling to determine local modification of the regional wind regime. Mechanistic models have been successfully developed for single species uniform canopied stands which become increasingly unstable as they increase in height (e.g. Smith et al., 1987, Peltola and Kellomaki, 1993). These models must be calibrated in field studies on a range of sites (Quine, 1995).

For stands with more complex structure and composition, the empirical approach is better suited. In this approach regression models are built which relate presence or magnitude of wind damage in sampling units to their environmental and management attributes. This method has been used for predicting stand edge damage in boreal black spruce strip cuts (Elling and Verry, 1978, Fleming and Crossfield, 1983) and in riparian reserves in coastal Oregon (Steinblums et al., 1984, Andrus and Froehlich, 1992). In these studies, the data were collected using field sampling, and coefficients of determination ranged from 0.57 to 0.93. Fridman and Valinger (1998) developed models for predicting snow and wind damage using tree, stand and site characteristics from permanent sample plots in Scots pine stands in Sweden. Their best model correctly predicted 81% of the undamaged and 82% of the damaged plots. Because windthrow results from the interaction of multiple factors, these equations should not be applied in locations with biophysical or management characteristics different from those used in building the model.

For the construction of empirical models, information about large numbers of sample units is necessary. Geographic information systems (GISs) provide the opportunity to assemble stand or landscape level information, generate spatial variables, analyze, and map the results. Wright and Quine (1993) used a GIS to analyze damage caused by a strong wind event in a forest in North Yorkshire. They divided the mapsheets into $50×50 m^{2}$ cells, added a layer with three classes of wind damage (none, partial, total) determined through ground mapping by field foresters, and analyzed the association of windthrow with individual topographic, soil and stand variables contained in other GIS layers. While they found associations between damage and specific variables, they did not build a predictive model. Moore and Somerville (1998) in New Zealand used GIS software to run a mechanistic model with a tree/stand critical windspeed component and an airflow component. They concluded that their model provided reasonable predictions for a test area with non-complex terrain but performed poorly in more complex terrain due to inaccurate prediction of windspeeds.

The primary objective of this study is to use readily available landscape and stand level information to produce empirical models for prediction of windthrow damage along the boundaries of proposed harvest units. A secondary objective is to investigate the relationship between site, stand and management factors and damage severity.

Section snippets

Description of study area

The study area is located at the northern tip of Vancouver Island bounded by Johnston Strait to the East and Quatsino Sound to the West. Six 1:20 000 mapsheets, covering an area of 900 km² were selected as representative of the range of biophysical conditions in this landscape. This area is in the Nahwitti Coastal Lowland physiographic region bordering on the northern extension of the Vancouver Island Ranges to the south and west (Lewis, 1985). Low pressure systems originating over the Pacific

Results

Several of the stand variables are correlated, reflecting linear relations between stand age and stocking, and height of mature stands and site quality (Table 6). Ground slope increases with elevation, and topex scores increase in areas with steeper slopes. The more mountainous condition of the west coast is reflected in the correlation of coast with elevation and TOPEX2000.

The major stand types are also associated with specific site features. The χ² tests indicated that the distribution of

Discussion

Boundary orientation relative to damaging winds is commonly associated with higher windthrow risk (e.g. Ruel, 1995). There is a clear directionality to the damage in this area of Northern Vancouver Island, with boundary orientation and position of segment relative to opening centroid both contributing to increased damage. The vulnerability of boundaries is consistent with the storm wind directions recorded at the local AES stations with a more easterly component on the east side of the island

Conclusion

Windthrow results from complex interactions of climatic, site and management factors. GISs combined with aerial photograph detection of wind damage provide an opportunity to develop windthrow risk models and maps for large landscape units. The large datasets produced in this approach enable detection of the association of damage with site and management features which is not easy to determine from field observations of a limited number of sites. By using endemic damage associated with cutblock

Acknowledgements

Funding for this project was provided by Forest Renewal British Columbia. Western Forest Products Limited provided logistical support, GIS and aerial photograph information. Arnold Moy and Eric Aubert assisted with programming. Jerry Maedel of the UBC GIS/FIRMS lab provided advice and assistance throughout the project.

References (32)

R.R Alexander
Minimizing windfall around clear cuttings in spruce–fir forests
For. Sci.
(1964)
C.J Anderson et al.
Root extraction force measurements for Sitka spruce
Forestry
(1989)
Andrus, C., Froehlich, H., 1992. Wind damage in streamside buffers and its effect on accelerated sedimentation in...
J Becquey et al.
L’existence de zones de stabilite des peuplements. Consequences sur la gestion
Revue Forestiere Francaise
(1987)
K.W Cremer et al.
Effects of stocking and thinning on wind damage in plantations
NZ J. For. Sci.
(1982)
A.E Elling et al.
Predicting wind-caused mortality in strip cut stands of peatland black spruce
For. Chron.
(1978)
Environment Canada, 1994. Canadian Monthly Climate Data and 1961–1990 Climate Normals. CDROM. Environment Canada...
ESRI, 1996. Using ArcView GIS. Environmental Systems Research International Inc., Redlands, CA, 340...
Fleming, R.L., Crossfield, R.M., 1983. Strip cutting in shallow-soil upland black spruce near Nipigon, Ontario. III....
J Fridman et al.
Modelling probability of snow and wind damage using tree, stand, and site characteristics from Pinus sylvestris sample plots
Scand. J. For. Res.
(1998)

B.A Gardiner et al.

Field and wind tunnel assessments of the implications of respacing and thinning for tree stability

Forestry

(1997)

Hannah, P., Palutikof, J.P., Quine, C.P., 1995. Predicting windspeeds for forest areas in complex terrain. In: Coutts,...

Harris, A.S., 1989. Wind in the forests of southeast Alaska and guides for reducing damage. USDA Forest Service...

Hosmer, D.W., Lemeshow, S., 1989. Applied Logistic Regression. Wiley, Toronto, Ont., 307...

O Laiho

Susceptibility of forest stands to windthrow in Southern Finland

Folia Forestalia

(1987)

Lewis, T., 1985. Ecosystems of the Quatsino Tree Farm Licence (T.F.L. 6). Unpublished Report prepared for Western...

Cited by (97)

Predicting the risk of tree fall onto railway lines
2024, Forest Ecology and Management
Tree fall onto railway lines puts passengers at risk and causes large economic losses due to disruption of train services and damage to infrastructure. Railway lines in Germany are vulnerable to tree fall because of the large number of trackside trees that exist in that country with approximately 70% of all railway lines being tree-lined. In this paper we first tested whether a hybrid-mechanistic tree wind damage model, ForestGALES, could identify the sections of the railway network affected by tree fall in two federal states of Germany, Northrhine-Westphalia (NRW) and Thuringia (TH). We secondly tested whether the model, in combination with meteorological forecast models, could predict where tree fall occurred during a damaging windstorm. We used information on tree characteristics derived from LiDAR and aerial photography along the railway line network in NRW and TH to calculate the critical wind speed (CWS) at which damage is expected to happen for every individual tree as a function of its size and species, and the underlying soil. The railway network was then divided into 500 m sections and the statistics of the CWS, tree height, and species composition (broadleaf/conifer mix) within each section were calculated. Analysis of past tree fall events recorded by Deutsche Bahn AG (DB) showed that there was a significantly lower minimum CWS and significantly greater maximum tree height in sections that had recorded damage. In a second step we compared the calculated CWS values for all trees against downscaled wind speed assessments across the two federal states during Storm Friederike (named Storm David internationally) on 18 January 2018 and tested the ability of the model to discriminate sections with recorded damage during the storm. Excellent model discrimination was found with an AUC value of 0.82 and an overall model accuracy of 74.2%. The first test showed that the ForestGALES model with precise individual tree information can identify the sections of a railway network most vulnerable to tree fall. The second analysis showed, for the one storm tested, that the ForestGALES model when combined with predicted storm wind speeds can identify the most probable sections of the railway network to experience tree fall during an approaching damaging storm. Such information could be of value in firstly planning remedial work along railway lines, and secondly preparing the railway network ahead of a major storm.
Poleward migration of tropical cyclones induced severe disturbance of boreal forest above 50°
2023, Science of the Total Environment
With global warming, tropical cyclones (TCs) are moving to northern latitudes with devastating effects on boreal forests and significant ecological and socioeconomic consequences in the northern hemisphere. Recently, TCs disturbances have been documented in the northern temperate and even the southern boreal forest zone. Here we report and quantify the impact of TC Lingling (2019), which damaged the boreal forests >50° latitude in a remote area of Sakhalin Island, Northeast Asia. A multi-step algorithm was used to identify disturbed forested areas together with Sentinel-2 imagery to recognize windthrow patches caused by TCs and evaluate tree species composition. We found extensive damage to boreal forests caused by TC Lingling, with forested area losses of >80 km². The affected areas mainly belonged to zonal dark coniferous forests, which account for 54 km² of windthrows. In contrast, a lower impact was recorded in deciduous broadleaf and larch forests. TC Lingling caused a high proportion (>50 %) of large gaps (>10 ha), however, gaps of such extent have not been previously recorded in these dark coniferous forests. Hence, our study highlights the potential of TCs as the new disturbance agent responsible for extensive disturbances of boreal forests at more northern latitudes than previously thought. This implies the significant role of TCs in disturbance regimes and boreal forest dynamics. We suggest that continued poleward migration of TCs may lead to an unprecedentedly large area of disturbed boreal forests resulting in complex changes in diversity and ecosystem functioning. Our findings are crucial for identifying potential shifts in boreal forest structure and dynamics under ongoing global climate change and altered forest disturbance regimes.
Tropical cyclones moving into boreal forests: Relationships between disturbance areas and environmental drivers
2022, Science of the Total Environment
Tropical cyclones (TCs) are common disturbance agents in tropical and subtropical latitudes. With global warming, TCs began to move to northern latitudes, with devastating effects on boreal forests. However, it remains unclear where and when these extraordinary events occur and how they affect forest structure and ecosystem functioning. Hence knowing which geomorphological features, landforms, and forest types are most susceptible to severe wind disturbance is vital to better predict the future impacts of intensifying tropical cyclones on boreal forests. In October 2015, catastrophic TC Dujuan hit the island of Sakhalin in the Russian Far East. With a wind speed of 63 m·s⁻¹, it became the strongest wind recorded in Sakhalin, damaging >42,000 ha of native forests with different levels of severity. We used high-resolution RGB satellite images, DEM-derived geomorphological patterns, and the U-Net-like convolutional neural network to quantify the damaged area in specific landform, forest type, and windthrow patch size categories. We found that large gaps (>1 ha) represent >40 % of the damaged area while small gaps (<0.1 ha) only 20 %. The recorded canopy gaps are very large for the southern boreal forest. We found that the aspect (slope exposure) is the most important in explaining the damaged area, followed by canopy closure and landform type. Closed-canopy coniferous forests on steep, west-facing slopes (typical of convex reliefs such as ridges, spurs, and peaks) are at a much higher risk of being disturbed by TCs than open-canopy mountain birch forests or coniferous forests and broadleaved riparian forests in concave reliefs such as valley bottoms. We suggest that the projected ongoing poleward migration of TCs will lead to an unprecedentedly large area of disturbed forest, which results in complex changes in forest dynamics and ecosystem functioning. Our findings are crucial for the development of mitigation and adaptation strategies under future changes in TC activity.
Long-term cumulative impacts of windthrow and subsequent management on tree species composition and aboveground biomass: A simulation study considering regeneration on downed logs
2021, Forest Ecology and Management
Citation Excerpt :
Windthrow risk is influenced not only by stand age but also by the wind direction, wind intensity, forest structures, topography, and so on. Previous studies have reported that the risk of windthrow is greater in individual trees with larger diameters at breast height or tree heights (Rich et al., 2007) and in stands that have homogeneous structures (Jalkanen and Mattila, 2000; Mitchell et al., 2001; Morimoto et al., 2019b). On the other hand, predicting windthrow risk is very challenging because it varies greatly among regions and among windthrow events (Dobor et al., 2019).
Post-windthrow management delays forest biomass recovery by altering the situation of disturbance legacies and can change the species composition. Although the short-term effects of post-windthrow management have been well studied, we do not have enough knowledge about the long-term effects of post-windthrow management on species composition and biomass recovery. Those effects associated with an increase in the windthrow frequency are also unknown. Although forest landscape models can effectively evaluate these effects, conventional models do not represent the regeneration process on downed logs, which is essential for simulating forest succession. We focused on hemiboreal forests in northern Japan and aimed to (1) incorporate the regeneration process on downed logs into LANDIS-II, which is one of the most used forest landscape models; (2) evaluate the long-term effects of post-windthrow management on tree species composition and aboveground biomass recovery; and (3) evaluate the associated long-term effects of interactions between post-windthrow management and increased windthrow frequency. We incorporated the regeneration process on downed logs into LANDIS-II by regulating the probability of the establishment of species that depend on dead wood, such as spruce, according to the availability of well-decayed dead wood. The incorporation of this process resulted in simulations of trends in species composition and aboveground biomass recovery after post-windthrow management that were more accurate than those produced by the original model. In the modified LANDIS-II simulation, reductions in dead wood and advanced seedlings due to salvage logging had little effect on the tree species composition or aboveground biomass recovery; however, the complete destruction of advanced seedlings by scarification induced a delay in aboveground biomass recovery and a shift to birch-dominated forests that continued for 100 years. In addition, the reduction in dead wood due to salvage logging decreased the number of seedlings, especially of dead wood-dependent species, that established after windthrow. When the windthrow frequency doubled, this decrease in seedlings induced a delay in aboveground biomass recovery, and a substantial decrease in dead wood-dependent species biomass occurred after a subsequent windthrow event. However, after the second windthrow event and following scarification, the forest recovered in the same way as after the first windthrow because the destruction of advanced seedlings and understory plants, namely, dwarf bamboo (Sasa spp.), by scarification reset the site conditions. To conserve the species composition and aboveground biomass of hemiboreal forests under climate change, which is expected to increase windthrow frequency, salvage logging and scarification should be avoided.
An evidence-based review of the effectiveness of riparian buffers to maintain stream temperature and stream-associated amphibian populations in the Pacific Northwest of Canada and the United States
2021, Forest Ecology and Management
Citation Excerpt :
Wind damage is strongly associated with buffer orientation relative to the predominant storm direction (i.e., southeast, south, southwest in the Pacific Northwest) and local conditions (including wind fetch length resulting from the size of clearcut units; Kramer et al., 2001; Mitchell et al., 2001; Beese et al., 2019). Reductions in windthrow mortality are feasible when site and landscape factors are considered in harvest unit plans (Kramer et al., 2001; Mitchell et al., 2001; Beese et al., 2019). The temperature regimes of headwater streams were spatially and temporally variable.
We employed a systematic evidence review to evaluate empirical scientific evidence for the effectiveness of buffering headwater (typically non-fish-bearing) streams to maintain stream temperature and stream-associated amphibian populations in the Pacific Northwest of Canada and the United States. To address our synthesis objective, we identified thirteen temperature, seven amphibian, and two temperature/amphibian primary research studies that met objective inclusion criteria. We evaluated external validity for how study treatments inform or were linked to causal factors influencing temperature response (a) and how the sampled population represented or provided inference to an intended target population or landscape (b). The evidence indicated substantial variability in the temperature response to streamside buffers. The effect size for the mean 7-day maximum temperature metric showed a positive association when comparing no-buffers (clearcut) to treatments with wide buffers (≥30 m). However, this effect varied substantially and overlap existed in effect sizes among no-cut buffers, no-cut plus variable retention buffers, and no-cut patch buffers all ≤20 m wide. Large variability in effect size among treatments obscured any potential trend between effect size for the seasonal (summer) mean daily maximum temperature metric and buffer width. Shade was correlated with temperature response within several studies, but direct comparisons of treatment effectiveness among studies as a function of shade was confounded by different measurement methods. The evidence also indicated that variation in temperature response among studies may be associated with multiple factors (geology, hydrology, topography, latitude, and stream azimuth) that influence thermal sensitivity of streams to shade loss. For amphibians, we found mixed evidence for relationships between population responses and buffers maintained along streams after forest harvest. Specifically, we did not find evidence to support the contention that positive population responses are associated consistently with larger buffers. Also, considerable uncertainty exists about which environmental covariates reliably explain variation in amphibian population responses. Collectively, our results indicate that evidence is weak to address questions most relevant to policy discussions concerning effectiveness of alternative riparian management schemes. Future studies should test effectiveness of alternative treatments with either experimental or purposefully structured observational studies to develop tools and derive guidelines for how to achieve management goals based on site and landscape characteristics.
Risk assessment of forest disturbance by typhoons with heavy precipitation in northern Japan
2021, Forest Ecology and Management
Citation Excerpt :
The estimated value of the probability by modelling did not indicate the actual probability of forest disturbance at this event but rather its relative probability. As for explanatory variables, we first selected the crucial factors identified by previous studies focused on windthrow risk assessments in mountainous regions dominated by cool-temperate and boreal forests (Kramer et al., 2001; Mitchell et al., 2001; Nakajima et al., 2009). In addition, we selected a series of precipitation indicators identified by a previous study focused on shallow landslides in forests caused by heavy rainfall (Dou et al., 2019).
Under future climate regimes, the risk of typhoons accompanied by heavy rains is expected to increase. Although the risk of disturbance to forest stands by strong winds has long been of interest, we have little knowledge of how the process is mediated by storms and precipitation. Using machine learning, we assess the disturbance risk to cool-temperate forests by typhoons that landed in northern Japan in late August 2016 to determine the features of damage caused by typhoons accompanied by heavy precipitation, discuss how the process is mediated by precipitation as inferred from the modelling results, and delineate the effective solutions for forest management to decrease the future risk in silviculture. In the results, we confirmed two types of behaviours in the model: one represents the same process as that of forest disturbance by strong wind, which has been widely studied, and another represents a unique process mediated by storms and precipitation that has not been previously investigated. Specifically, the ridges that received strong wind from the front side had the highest risk of disturbance. Precipitation increased the probability of disturbance in forest stands, and its effect was dependent on the dominant species composition. Our hypothesis regarding treefall mediated by storms and precipitation is that rainwater flows into the gaps around the tree root systems during sway and the introduction of rainwater below the root-soil plate decreases the root anchorage. The species-specific vulnerability to rainfall may depend on the volume of lateral roots. Modelling the disturbance risk helped us to examine the kinds of factors that were related to exposure and vulnerability that should be managed to effectively decrease the risk of disturbance by typhoons during future uncontrollable hazards. It is recommended to avoid silviculture on the ridges of plateaus considering the high risk estimated in this area. In addition, species with dense lateral roots would be suitable for planting because they may have high resistance to typhoons with heavy precipitation.

View all citing articles on Scopus

View full text

Article preview

Forest Ecology and Management

Abstract

Introduction

Section snippets

Description of study area

Results

Discussion

Conclusion

Acknowledgements

References (32)

Minimizing windfall around clear cuttings in spruce–fir forests

For. Sci.

Root extraction force measurements for Sitka spruce

Forestry

L’existence de zones de stabilite des peuplements. Consequences sur la gestion

Revue Forestiere Francaise

Effects of stocking and thinning on wind damage in plantations

NZ J. For. Sci.

Predicting wind-caused mortality in strip cut stands of peatland black spruce

For. Chron.

Modelling probability of snow and wind damage using tree, stand, and site characteristics from Pinus sylvestris sample plots

Scand. J. For. Res.

Field and wind tunnel assessments of the implications of respacing and thinning for tree stability

Forestry

Susceptibility of forest stands to windthrow in Southern Finland

Folia Forestalia

Cited by (97)

Predicting the risk of tree fall onto railway lines

Poleward migration of tropical cyclones induced severe disturbance of boreal forest above 50°

Tropical cyclones moving into boreal forests: Relationships between disturbance areas and environmental drivers

Long-term cumulative impacts of windthrow and subsequent management on tree species composition and aboveground biomass: A simulation study considering regeneration on downed logs

An evidence-based review of the effectiveness of riparian buffers to maintain stream temperature and stream-associated amphibian populations in the Pacific Northwest of Canada and the United States

Risk assessment of forest disturbance by typhoons with heavy precipitation in northern Japan