The firestorm of 2003 has increased community awareness of the dangers of forest fires within the wildland urban interface. The interface fires that year demonstrated the vulnerability of homes immediately adjacent to the forest. While the probability of fire within West Arm Provincial Park, hereafter referred to as the ‘Park’, is not as high as that for Okanagan Mountain Provincial Park, there is risk to the community associated with a large interface fire. BC Parks is actively engaged in developing strategies to manage and mitigate this fire risk.

As part of the strategic planning currently underway, B.A. Blackwell and Associates Ltd. were retained by BC Parks to identify a Wildland Urban Interface zone associated with the community of Nelson in West Arm Provincial Park and carry out fuel hazard assessments of this zone and prepare fuel mitigation prescriptions where required. This work has included determination of fire behaviour potential, values at risk, and suppression capability in the Park as well as in the adjacent communities of Nelson and Harrop Proctor. Following the preliminary risk assessment phase, the planning effort has been focused on the design and engineering of a strategic fuel treatment in an area of the Park that will provide fire suppression personnel with an effective anchor to marshal resources, conduct a back burn and/or mobilize equipment in the Park.

Integrating suppression planning with ecological and recreation values is key to the development of an effective fuel reduction prescription for a Park. There are many considerations in designing effective fuel treatments to protect the ecological and recreational values present in the Park. During the development of this plan, forest health, access, treatment impacts, fire history, natural disturbance regimes and ecology were considered in respect to protecting the adjacent communities and the community watershed in the Park. Properly designed and implemented, fuel treatments can complement ecosystem restoration goals by reintroducing disturbance which emulates the results of historic natural disturbances.

To undertake this project we utilized methods and data developed as part of a Provincial Fire Risk Assessment project, undertaken in 2004, on behalf of the Union of B.C. Municipalities and the Ministry of Forests and Range Protection Branch. The data were updated to incorporate more specific fuel typing of the Park based upon data collected in the field. The Provincial fuel typing was also revised to address differences in the composition and structure of forests not well represented within the Canadian Forest Fire Behaviour Prediction System (CFFBPS). These adjusted fuel types were used in modelling fire behaviour in the Park to aid in determining appropriate fuel break locations and treatment prescriptions.

The results of this project provide a strategic level analysis that will assist in the development of comprehensive interface management plan and emergency response plans for the Park which may be required in the event of a large wildfire.

1.1 Project Area

West Arm Provincial Park encompasses 25,319 ha. Its south western boundary abuts the north eastern municipal boundary of Nelson. The Park extends north along the shore of Kootenay Lake to Harrop and reaches south close to Whitewater Winter Resort (Figure 1). The Park encompasses a range of habitats, from lakeshore to alpine. It plays an important role in representing the Southern Columbia Mountains (SCM) Ecosection and includes the Interior Cedar Hemlock and Engelmann Spruce-Subalpine Fir Biogeoclimatic zones. To ensure that relevant data adjacent to the Park was considered in planning, a 2 km wide buffer was created around the Park.

Figure 1. Overview of West Arm Provincial Park.

Access to the Park is quite limited. Whitewater Winter Resort provides some hiking and ski touring access. The foreshore is primarily boat accessible where stream fans exist. Three gravel roads provide limited access including: Svoboda Road that runs up Five Mile Creek; a 1.4 km road to Goddard Hill from Harrop; and a forest service road to the confluence of Midge and Ketetl Creek. There are also several major trails including one that runs along Lasca Creek and another along the water pipeline that extends to the City of Nelson.

Recreation use in the Park includes mountain biking, hiking, back-country skiing, hunting, wildlife viewing as well as water activities along the lake shore such as boating, fishing and swimming. Most activity occurs on the lower slopes adjacent to Nelson

The Park also contains the community watersheds which supply drinking water to the City of Nelson. The watershed is important in maintaining the hydrological functions that determine water quality, quantity and timing of flows.

There are several known First Nations archaeological sites in the Park. Most sites are located along the foreshore, but only the area within 750 m of Kootenay Lake has been surveyed according to the Park Management Plan (MELP 2000). There are also archaeological sites related to European settlement, found primarily along the foreshore. An Archaeological Overview Assessment was done for the park in 2006, which identified probable areas of importance for First Nations.

1.2 Ecology

The Park is located in the Interior Cedar Hemlock (ICH) and Engelmann Spruce Subalpine Fir (ESSF) biogeoclimatic subzones (Braumandl and Curran 1992). Within these zones are: Interior Cedar Hemlock dry warm (ICHdw1); Interior Cedar Hemlock moist warm (ICHmw4); and 4 variants of Engelmann Spruce Subalpine Fir wet cool (ESSFwc) (5, 6, wcp, wcw). Table 1 summarizes the areas for each variant.

Table 1. Summary of BEC subzones and variants in West Arm Provincial Park.

Biogeoclimatic Subzone and Variant	Area (ha)
ESSFwc 5	3557
ESSFwc 6	10306
ESSFwcp	784
ESSFwcw	4682
ICH dw 1	2593
ICH mw 4	3166

Climate in the ICHdw1 and mw4 has hot to very hot moist summers, very mild winters with little snowfall and shallow snow packs. No data exists for the ESSF subzones in the Braumandl and Curran field guide to site identification (1992).

As no variant data exists for the ICHmw4, the grid for the ICHmw2 was used for all site series related interpretations (pers. comm. Regional Ecologist).

Figure 2. Biogeoclimatic subzones and variants in West Arm Provincial Park.

2.0 Historic Fire Regimes and Stand Structure

To date there have been no empirically-based fire regime and forest structure studies conducted in and adjacent to West Arm Provincial Park. The current understanding of historic fire regimes has come from an interpretation of disturbance dynamics as they relate to the biogeoclimatic classification system. The Biodiversity Guidebook (Ministry of Environment and Ministry of Forests 1995) describes disturbance agents and their effects on ecosystem structure by biogeoclimatic subzone and variant and uses a numerical classification system of Natural Disturbance Types. The predominant disturbance agent in the classification system is fire, although other critical disturbance agents are factored into the system. Ecosystems with historically frequent fire regimes are classified as NDT4, while ecosystems with increasingly long fire intervals are classified as either NDT3 or NDT2. Using this classification system, Holt and Machmer (2005) suggest that the Park’s ecosystems be classified as either NDT3 or NDT2. The ICHmw4, ESSFwc5, and ESSFwc6 variants are all classified as NDT2 – very infrequent, stand-replacing disturbance events (Figure 2).

The mid-elevation ICHmw2 and ESSFwc4 variants are classified as NDT2, but are considered to vary in historic frequency by site. Warm aspects and dry sites are thought to be characterized by a fire frequency of 100 years, while adjacent cool aspects and cool sites are thought have frequencies in excess of 500 years (Holt and Machmer 2005). Lower elevation ecosystems in the Park are classified as NDT3, which is characterized by stand-replacing disturbance events with a frequency of 150-200 years. This classification would apply to the ICHdw subzone. Holt and Machmer (2005) suggest that warmer, westerly and south-westerly aspects likely include a mixed-severity fire regime with a frequency range of 30-100 years.

These long fire intervals suggest that the natural disturbance regime in the Park is likely dominated by moderate to long intervals of low intensity gap-replacement stand dynamics attributable to agents such as insects, fungi, and wind that operate on a continual basis. These dynamics would be interrupted by infrequent stand-replacing disturbances such as wildfires of various sizes, outbreaks of bark beetles, defoliating insects, and root diseases (Holt and Machmer 2005). Stand structure under this natural disturbance regime would include areas of even-aged forest where shorter interval, stand-replacing fires are common, and structurally-complex forest where very long fire intervals prevail. There is enough evidence, however, to suggest that another interpretation of fire history and historic forest structure is possible. The interpretations of Holt and Machmer (2005) are based entirely on the Biodiversity Guidebook plus local opinion. The Biodiversity Guidebook natural disturbance classifications are not based on empirical data in most cases, but on professional judgment. However, determining the historic disturbance regimes and resultant forest structures in West Arm Provincial Park is complicated by the large-scale fires that occurred in the late 1800’s and early 1900’s.

A number of lines of evidence are available that can be used to gain a more accurate picture of the fire ecology history of West Arm Provincial Park, including: age structure data, point-source fire chronologies, historic photographs, empirical data from adjacent sites, and predictive models. In 2003 a predictive model of Historic Natural Fire Regimes (HNFR) was developed by Blackwell et al. for the southern third of the province, including West Arm Provincial Park (Figure 3). This model incorporated up to date empirical historic fire regime data from BC, AB, and the adjacent United States, terrain factors affecting fire behaviour, and professional judgment. Another significant difference of the model is the recognition and delineation of mixed-severity fire regimes. The model resulted in 10 potential fire regimes which replace 4 natural disturbance types. For West Arm Provincial Park the primary difference is the parsing of the landscape into 5 fire regimes, shown in Figure 3, instead of 2 natural disturbance types.

Figure 3. Historic natural fire regimes within West Arm Provincial Park.

Using the HNFR model, the lower slopes of the Park are characterized by frequent mixed-severity fire. These areas are comprised of Douglas-fir, ponderosa pine, western larch, lodgepole pine, western white pine, western redcedar, western hemlock, and grand fir. A number of deciduous species can also be found here. This area is characterized as high frequency due to the regional climate being conducive to fire start and spread. The mixed-severity characterization comes from the greater fire intensity associated with higher between-fire fuel loads on productive sites in the west Kootenays, and the high number of fire-intolerant species.

Progressing upslope, fire frequency diminishes leading to greater between-fire fuel accumulations and greater fire severity. Fire intensity is also affected by increasing slope angle and slope uniformity. With increasing elevation comes decreasing fire tolerance, and higher general fire vulnerability due to slower growth rates (shorter trees with small diameters). The complex terrain of West Arm Provincial Park aids in fire regime heterogeneity. Under this model the maximum fire-free interval would be in the 150-200 year range, not the 500 year figure as suggested by the Biodiversity Guidebook and Holt and Machmer (2005). The higher frequency is the result of a number of factors including: the proximity of the Park to higher frequency fire regimes, the lack of significant barriers to fire spread both within and outside the Park, the latitude of the Park, and the lack of terrain features that would ameliorate local fuel characteristics.

This picture of historic fire regimes and forest structure in West Arm Provincial Park is still a predictive model, not unlike the model of NDT’s proposed in the Biodiversity Guidebook. It does, however, include the interpretation of empirical studies from adjacent areas. An example of this is the small fire history analysis conducted by Quesnel and Pinnell (2000) for the West Arm Demonstration Forest immediately across the lake from West Arm Provincial Park. The authors found a very frequent fire regime that maintained open forests comprising ponderosa pine and Douglas-fir. However, because fire has been absent on this site for the past 100 to 120 years, forest density has increased with much of the density comprised of fire-intolerant grand fir, lodgepole pine, and western redcedar. While the study conducted by Quesnel and Pinnell (2000) describes conditions on a warm aspect, these conditions should also apply to the cool aspect across the lake in the Park due to proximity. This would suggest that the NDT designation proposed by Holt and Machmer (2005) is too conservative regarding fire frequency. The study results obtained by Quesnel and Pinnell (2000) were used in the development of the HNFR predictive model.

Gaining an understanding of historic fire regimes and forest structure is best accomplished using solid empirical data derived through disturbance chronologies. Disturbance chronologies document historical disturbance frequency and severity via fire scar interpretation and cross dating (Dietrich and Swetnam 1984, Agee 1993). However, as a stand-alone analytical tool, they often yield only crude inferences about associated vegetation community structure (Hessburg et al. 1999). Historic stand reconstructions, through tree ring and cohort analysis and stem mapping (Habeck 1990, Covington and Moore 1994, Arno et al. 1995), provide spatially and temporally precise information about composition and structure (Hessburg et al. 1999).

Both fire scar analysis and tree ring and cohort analysis were applied to West Arm Provincial Park in the fall of 2007. However, fires that occurred near the turn-of-the-century as well as the preceding timber harvest, made the collection of disturbance and forest structure evidence difficult. Although several wildfires burned through West Arm Provincial Park in the late 1890’s and early 1900’s, exact dates have not yet been established. Early photographs and oral histories point to a probable fire in 1896 and a number of subsequent fires in 1911 (www.city.nelson.bc.ca/html/founding.html). From a landscape perspective, the result was heavily denuded hill slopes (Figure 4).

Multiple disturbance events such as wildfires create problems in terms of collecting disturbance chronology data. A single fire would have resulted in large numbers of dead trees, live trees with scars, and patches of live trees which would all have provided empirical data that would enable researchers to reconstruct past processes and patterns. However, the reburn of the initial burned areas would have killed and consumed large areas of the remaining live and dead trees, leaving little evidence of conditions that prevailed prior to the mid- to late-1800’s.

Figure 4. Photograph of the City of Nelson in 1898. The mountainsides surrounding Nelson have all been heavily impacted by logging and wildfire.

In the fall of 2007 we collected a number of increment cores, fire scar samples, and cross-sections in order to develop a number of lines of evidence that will enable us to piece together historic disturbance and forest structure. A total of 38 cores were collected from western white pine (1), lodgepole pine (12), western redcedar (3), western larch (10), and Douglas-fir (12). We targeted large diameter trees with structural elements consistent with older trees. These included bark char, large limbs, extensive clear bole, flat or broken top, and flat, platy bark on western larch. Most cores were collected from shade-intolerant tree species. The age of these trees suggests that they germinated following a disturbance that created growing space, and/or a bare mineral soil germinating substrate (western larch). The western redcedar cores were collected from snags. All other cores were from live trees.

One of our potential lines of evidence is cohort establishment. Species such as western larch require very open canopies for growth and a bare mineral soil seedbed. The same is true for western white pine. Douglas-fir and lodgepole pine are less dependent on a bare mineral soil seedbed and, to a certain extent, are more tolerant of shade.

In addition to the cores, we collected 6 western larch cross-sections, 2 western larch fire scar samples, and 3 western redcedar fire scar samples. All of these samples were from dead trees. The western larch cross-sections are intended for fire-origin cohort analysis, while the western larch fire scar samples, each exhibiting two scars, are intended to provide solid dates for the turn-of-the-century fires. The western redcedar snags are unique in that they exhibit scars from three fire events, two of which we hope to establish dates for (Figure 5). Each scar sample indicates the occurrence of a low-intensity fire that scarred the tree but did not kill it. Based on the outside ring date, the scars also suggest the incidence of a fire that killed the tree. A third fire or fires began the decomposition process by partially consuming the snags (live trees killed by fire are typically charred but do not exhibit bole consumption).

Figure 5. Partially consumed western redcedar snag in West Arm Provincial Park.

We were able to successfully cross-date five of the western redcedar samples and four of the western larch samples, although in many cases the dates are approximated due to low correlations (Figure 6). The samples were assessed against both a western larch and Douglas-fir regional chronology with many samples sharing a higher correlation with larch than Douglas-fir. No western redcedar chronologies exist for this area. As part of this project we did not have the time or budget to develop a local ring-width chronology. Future work in the West Arm Provincial Park should include the development of multi-species chronologies. Of the four larch samples, one sample contained two fire scars dated 1894 and 1917. This sample came from the steep northeast aspect overlooking Five Mile Creek at the far eastern end of the treatment unit boundary. The remaining larch and redcedar samples came from within the large lodgepole pine type in the center of the treatment unit. The remaining three larch samples provided pith dates that we suspect correspond to wildfires.

The ages of the nine fire scar samples are highly variable with the western redcedar samples ranging in age from 150 to over 200 years. Several samples, SC4-THPL and UNK-THPL, are estimated to have germinated in the early 1600’s and died in the mid-1800’s. Two other redcedar likely germinated in the late 1600’s and died in the late 1800’s. Most of the larches are considerably younger than the redcedar at 100 to 150 years, except for sample SC7-LAOC which is estimated at over 200 years of age. The five redcedar samples were charred while none of the larch samples exhibited charring. We have high confidence in two fire dates from a western larch; 1894 and 1917. These are represented in Figure 6 by the narrow vertical bars. Three other fire “periods” are indicated by wide shaded areas. These are 1620 to 1655, 1775 to 1805, and 1835 to 1860. A number of lines of evidence converge within each of these “periods” to suggest that a wildfire occurred. Within the period 1620 to 1655 we have three of our nine samples germinating. Western larch is highly reliant on a bare mineral soil seedbed and lots of light for germination and growth. Western redcedar, on the other hand, is shade-tolerant and is not as reliant on a bare mineral soil substrate for germination. Western redcedar does, however, germinate better on bare soil than on duff (Edwards and Leadom 1988). The second period, 1775 to 1805, corresponds to approximate germination dates for two larch and one redcedar. Four fire dates from scars on redcedar and larch also fall within this period. The last period, 1835 to 1860, corresponds to two fire scars recorded on western redcedar plus approximate death dates for all five redcedar sampled.

Thirty-nine increment cores were collected in the Five Mile Creek area in an attempt to determine estimates of cohort ages and to possibly identify a number of disturbance dates (Figure X). Two distinct stand types were sampled: a mesic area containing Douglas-fir, western redcedar, western larch, trembling aspen, western hemlock, grand fir, western white pine, paper birch, and lodgepole pine; and, a drier site containing predominantly lodgepole pine with scattered western larch and Douglas-fir. The mesic site is on the west end of the treatment area, while the dry lodgepole pine-dominated stand type occupies the eastern three quarters of the area. One core was from western white pine, ten of the cores were from western larch, twelve cores were from Douglas-fir, and ten were from lodgepole pine. Core samples PIMO-1, LAOC-1 through 6, and all 12 PSME samples were collected in the mesic stand type. All remaining core samples, including all PICO samples, were collected in the dry lodgepole pine stand type. An additional six cores were from western redcedar snags. All trees were cored at DBH and ages indicated in Figure 6 are not corrected to germination date.

Two fire dates and two fire “periods” are superimposed on the core age data in Figure 7. The 1894 fire appears to correlate well with the germination of 58% of the trees sampled. This includes all of the lodgepole pine sampled, the one western white pine, five of the ten western larch, and three of the twelve Douglas-fir. This fire also shows up on a number of older Douglas-fir and western larch cores as injuries. The fire appears to have influenced stand dynamics on both the mesic and dry stand types. The 1917 fire, on the other hand, does not appear to have influenced stand dynamics on either site. There are no cohorts associated with it, the trees sampled did not contain any external evidence of it in the form of a catface, and it does not show up as an injury on the cores. This would suggest that the fire that scared the recorder tree did not spread to the treatment area (this snag was located in the Five Mile Creek draw just to the east of the treatment area).

Two fire “periods” also appear to have influenced stand dynamics over time, the earlier 1775 to 1805 period especially. Cohorts of both western larch and Douglas-fir appear to have been established shortly after each event. Western larch and Douglas-fir both germinate and grow well following disturbances that create a bare mineral soil seedbed and that reduce competition for light.

The combination of age data and fire scar data provides clues to the historic fire regime, but not a clear picture of it. The two fires dated 1894 and 1917 were likely human-caused and likely burned through an abnormal fuel complex of timber harvest slash. Extensive harvest in the late 1800’s was associated with railroad and mining needs – material for rails and timbers. One of these fires, the 1894 fire, resulted in much of the young cohort seen in the forests immediately adjacent to Nelson today, including the large proportion of lodgepole pine being attacked by the mountain pine beetle. The three fire “periods” indicated in Figure 6 and Figure 7, suggest that historically, fires were not uncommon in this area of the Park, occurring on average every 60 to 150 years. Some were high-severity, creating good germination and growth conditions for Douglas-fir, western larch, and western redcedar, and some were low-severity, scarring but not killing western redcedar. With this relatively frequent, mixed-severity fire regime, fire-tolerant species would be favoured over intolerant species and those requiring longer periods of time to reach reproductive age or for the site to develop special germination conditions. These conditions, and species assemblages, were likely present on the landscape but were fluid in space and time. The former list of species includes western larch, Douglas-fir, trembling aspen, paper birch, and lodgepole pine. The latter list includes western redcedar, western hemlock, western white pine, grand fir, and Engelmann spruce.

2.1 Future Fire Regimes

Climate (temperature, precipitation, and topography) combines with other variables to influence the vegetation that can grow in a given place. It also influences the timing, severity, and extent of fires (the natural fire regime) (Wells 2007). As climate changes, species growth, regeneration, and dominance will shift. For example, in much of the Pacific Northwest, lodgepole pine shares dominance with Douglas-fir at low/mid elevations and with subalpine fir at higher elevations. At low elevations, warmer, drier summers may translate into more favourable growing conditions for lodgepole pine than Douglas-fir (Hermann and Lavender 1990, Case and Peterson 2005), and at higher elevations increased temperatures and reduced snowpack may favour subalpine fir growth and regeneration. Climate warming is expected to increase the frequency of fires (decrease the fire return interval or fire interval) (Running 2006, Swetnam and Westerling 2007, Westerling et al. 2006).

“The effects of warming are not smooth or steady, but discontinuous in space and time. In general, the places and times that were coolest to start with have warmed the most; nighttime low temperatures, winter temperatures, and temperatures at high latitudes and high altitudes. Not only rising mean temperatures are of concern, but also an increase in the variability of the climate that might lead to more extreme weather events (Wells 2007).”

Of particular concern to fire scientists is the potential confluence of warming temperatures, high fuel loads, and impending drought in some areas. Some of these scientists predict that the area burned in wildfires will double or even triple over the next 50 years (Bartuska 2007, Wells 2007). High fuel loads are problematic in the Park where recent disturbances have resulted in large-scale tree mortality. The 2003 Ketutl Fire was not salvage logged, meaning all dead trees will eventually fall creating high fuel loads. The ongoing mountain pine beetle epidemic is also resulting in unnaturally high fuel loads. A reburn of the Ketutl Fire area or an initial burn through the beetle affected areas could result in significant damage to soil productivity and local hydrology.

One scenario linked to warming climate in the West Kootenays involves the distribution and productivity of lodgepole pine, and its consequences on other native species. The effect of a warmer climate on disturbances may have a greater affect on lodgepole pine distribution and productivity than direct climatic impacts on tree growth. Warmer, drier summers will increase the likelihood of fires (McKenzie et al. 2004), which could lead to changes in the distribution and abundance of plant species. While lodgepole pine is relatively intolerant to intense fires, regeneration immediately following fire is typically dominated by lodgepole pine because it can disperse large quantities of seeds from nearby trees and often has serotinous cones that open due to heating from fire (Lotan and Perry 1983). Warmer, drier summers may also lead to increased outbreaks of insects such as the mountain pine beetle (Logan and Powell 2001). Because it is likely that the current trend of warming temperatures will continue into the future (IPCC 2001), many Pacific Northwest forested ecosystems, especially drier systems, may experience reduced soil moisture, increased water stress and altered disturbance regimes. Extended summer drought over decades could significantly affect which tree species are the most productive and abundant (Case and Peterson 2007).

Potential future forest succession dynamics in the West Kootenays under a warming climate/increased disturbance scenario is graphically represented in Figure 8. Vital attributes theory postulates that a small number of life history attributes termed “vital attributes”, can help us predict the behaviour of plants in disturbed environments (Roberts 1999). Vital attributes pertain to the potentially dominant species in a particular community. Three main groups of vital attributes are recognized relating to the method of persistence of species during a disturbance, and to their subsequent arrival, to their ability to establish and grow to maturity following the disturbance, and to the time taken for them to reach critical stages in their life history (Noble and Slatyer 1980). In Figure 8, most dominant tree species would be able to persist on site with a historically median fire interval of 100 years. Some would be able to survive a disturbance, take advantage of less competitive germination and growing conditions, and start to produce large quantities of viable seed at a young age. This would certainly be the case for Douglas-fir, lodgepole pine, trembling aspen, western larch, and paper birch. Other species would be more disadvantaged by being killed by the disturbance, and having to seed in from outside the burn area. Species such as western hemlock, western redcedar, Engelmann spruce, subalpine fir, and grand fir, would also grow slower and take longer to produce large quantities of viable seed. Even with these disadvantages a historic fire interval of 100 years would still enable them to persist on the site.

Figure 8. Critical life stage ages for a number of West Kootenays tree species are graphed relative to the historic fire interval and a potential future fire interval under a warming climate.

Under a warming climate scenario, and a significant decrease in the fire interval, a large number of tree species would potentially be displaced by species more adapted to a higher frequency of disturbance. The vital attributes of Douglas-fir, western larch, trembling aspen, paper birch, and lodgepole pine would leave these species more adapted to an increased disturbance frequency scenario. Lodgepole pine, with its habit of producing large quantities of viable seed at a very young age, and sealed in serotinous cones, could dramatically increase its distribution and stand proportion under this scenario.

3.0 Fire Environment

The fire environment is described in the following section and includes fire weather, fire causes and frequency, and fuels and forest health factors which are currently influencing fuel types in and adjacent to the Park.

3.1 Fire Weather

The Canadian Forest Fire Danger Rating System (CFFDRS), developed by the Canadian Forestry Service, is used to assess fire danger and potential fire behaviour. The Ministry of Forests and Range (MOFR) maintains a network of fire weather stations during the fire season that is used to determine fire danger on forestlands within the community. The information is commonly used by municipalities and regional districts to monitor fire weather information provided by the MOFR Protection Branch to determine hazard ratings and associated fire bans and closures within their respective municipalities. Key fire weather parameters summarized as part of the analysis included:

· Drought Code: The Drought Code represents the moisture in deep, compact organic matter with a nominal depth of about 18 cm and a dry fuel load of 25 kg/m². It is a measure of long-term drought as it relates to fire behaviour.

· Days above Danger Class Rating IV and V: The Danger Class Rating is derived from fire weather indices and has 5 classes: 1) Very Low Danger; 2) Low Danger; 3) Moderate Danger; 4) High Danger; and 5) Extreme Danger.

It is important to understand the likelihood of exposure to periods of high fire danger, defined as Danger Class IV (high) and V (extreme), in order to determine appropriate prevention programs, levels of response, and management strategies.

Fire danger within the study area can vary significantly from season to season. Figure 9, Figure 10 and Figure 11 are compilations of available weather station data, dating back to the early 1900’s, within the ICHdw, ICHmw, and ESSFwc biogeoclimatic units (representative of the study area). The Figures provide a summary of the total number of Danger Class IV and V-days from May through to August of each year and show that fire danger can fluctuate substantially between years. On average, the number of Danger Class V-days within the ICHdw, ICHmw, and ESSFwc over the most recent 30 year period in the data is respectively 17, 12, and 6 per year. The decrease can be explained by the increasing elevation boundaries for these ecosystems. Typically, the most extreme fire weather occurs between late July and the third week of August.

Figure 9. Seasonal variability (May-August) in the number of Danger Class IV and V-days within the study area as described by the regional climate of the ICHdw (1904-2003).

Figure 11. Seasonal variability (May-August) in the number of Danger Class IV and V-days in the study area as described by the regional climate of the ESSFwc (1909-2003).

A summary of historic drought codes provides a similar comparison (Figure 12, Figure 13, and Figure 14). A drought code that exceeds 350 is considered high and is associated with high fire behaviour. A drought code exceeding 500 is considered extreme. Based on annual averages, drought code values commonly exceed 350 in the ICHdw, while drought code values in the ICHmw infrequently surpass 350 and very seldom surpass 350 in the ESSFwc. A comparison of monthly, rather than seasonal, values reveals that the low values in May and June reduce the seasonal average. During the months of July and August, values commonly exceed 500.

Figure 12.Summary of seasonal (May-August) drought codes by year for the ICHdw (1904-2003).

Figure 13. Summary of seasonal (May-August) drought codes by year for the ICHmw (1895-2003 – note missing data 1899-1965).

Figure 14. Summary of seasonal (May-August) drought codes by year for the ESSFwc (1909-2003).

3.2 Recorded Fire History

Recent fire history, between 1950 and 2003, in the Park is dominated by lightning caused fires (Table 2). No human caused fire was over 4 ha in size, while the largest lightning caused fire was in 2003 and consumed 7916 ha of forest. A marked decline in human caused fires has occurred over the last 3 decades (Table 3).

Table 2. The number of fires by cause and size in West Arm Provincial Park.

Cause	Fire Size <=4 ha		Fire Size > 4 ha		Summary
Cause	Area	Number of Fires	Area	Number of Fires	Total Fires	Total Area
Campfire	0.5	5			5	0.5
Equipment use	0.3	2			2	0.3
Fire use	2.1	7			7	2.1
Miscellaneous	2.7	7			7	2.7
Railroads	0.1	1			1	0.1
Smoker	0.9	8			8	0.9

Submitted by