General Pyromorphology cont.

“Not only the life forms of individual plants but also community characteristics, such as productivity, food chain length, and broad features of patterns of energy flow, are determined primarily by the local physical environment rather than by evolutionary history.” Levin, 1999.

4.2 Overview

Manifesting in multiple morphologies, wildfire inhabits a variety of synergistic states. This chapter discusses the qualitative differentials between these states, the global and local context thereof, together with the fire behaviours, fire ecologies, including physiological biotic traits as are associated therewith, and which will inform the material and information constructs that will be developed for the new Wildland Urban Interface paradigm.

4.2.1 Fire Regimes: Biomes as Biotic and Abiotic Assemblages

As one might expect of a relatively young science, there are several fire regime classification systems in use in North America alone, and as yet, no general consensus on which system to use in ecological literature. The seminal classification system was published in the proceedings of the 1972 annual meeting of The Ecological Society of America (Wright and Heinselman, 1973). Documenting studies of the role of fire in conifer forests, the paper gave precedence to fire frequency and intensity whereupon identifying fire regimes, specifying the following variants: Long-interval crown-fire regimes; Short-interval crown-fire regimes; Long-interval surface-fire regimes; Short- interval surface-fire regimes, and “all possible combinations of the above” (Ibidem, p.11). The authors identified fire as the over-arching ecological organiser within the study areas, while also acknowledging the complexity of the symbiosis between the abiotic and biotic elements involved. Their paper stressed the matter that not “anywhere in the world” was research underway to document fire regimes in their totality (Ibidem, p.5). In the absence of the technological advances that have occurred between then and now, Wright and Heinselman estimated it would take “centuries” to complete the task (Ibidem, p.5). However, the pair were part of a growing ecological school of thought that recognised the role of disturbance as imperative to ecosystem integrity and functioning, whereupon in the absence thereof not merely would many species generally demise, but go extinct. The very same year that the above paper went to print, fellow ecologist C. S. Holling published his seminal paper on resilience and stability of ecological systems (Holling, 1973), which though a work apart, evidenced the same general thinking around the role of disturbance in ecosystems [Fig. 21]. Collectively, Wright, Heinselman, and Holling understood the need for a shift from a principally quantitative interrogation of the natural world, to a qualitative approach. The parties also recognised that whereupon a disturbance becomes manifest within an ecosystem, it may align to not merely one successional trajectory, but to one of several, which, at that time, was a veritable ecological epiphany.

In the intervening years since the publication of Wright and Heinselman’s paper, various authors have added or subtracted to the number of regime types described, while also editing the defining characteristics of the schemata (Barrett et al, 2010; Morgan, 2001; Smith and Brown, 2000; Frost, 2000; Hardy, 1998a, 1998b; Agee, 1993; Crutzen and Goldammer, 1993; Pickett and White 1985; Kilgore, 1981). Differences of opinion with regard to the significance of the many variables in shaping fire regimes underpin the variances in the several classification systems. Whereas Heinselman’s system is principally shaped by the fire frequencies and intensities (Heinselman, 1973) later studies have placed emphasis on biotic assemblages (Davis et al, 2001), or on fire severity (Agee, 1993; Quigley et al, 1996; Hardy et al, 2001). Furthermore, whereas some classification systems are born primarily of ecological observations, others bring greater emphasis to anthropogenic factors. Therein, the classification system adopted tends relate to the end-purposes of the study concerned.

On the whole, the authors of the several hundred fire ecology journal papers, articles, and books reviewed for the purposes of this study employed a tri-part classification system in which fire regimes are described as being low-severity, mixed-severity, or high-severity in type. In some instances, authors added moderate-severity to this trio, thus engaging a quadri-part descriptive system, but more commonly ‘moderate’ is used interchangeably with ‘mixed’, (Agree, n.d). Occasionally authors utilised the Fire Regime Groups [FRG] specified by the LANDFIRE Program, which is one of a suite of more than twenty products delivered as part of a partnership between the U.S Department of Interior, the U.S Forest Service, and the Nature Conservancy, FRG harnesses the rich and varied quantitative data sets produced by Earth satellite and terrestrial observation ICT infrastructure.

Launched in 2001, in such fashion as one might expect of a software program, LANDFIRE and its various assets, have so far been updated on five occasions, in each instance by means of integrating new insights and technological capabilities. At the time of writing LF 2014 [LF_1.4.0] is the latest ‘update’ (LANDFIRE, 2014, 2016). FRG, as was Wright and Heinselman’s, is a 5-part fire regime classification system born of analysis of vegetation dynamics, fire spread, fire effects, and spatial context, that utilises empirical data, but within a qualitative approach. However, whereas Wright and Heinselman’s system featured two long-term [35-200 years] fire return intervals, two short-term [<35 years] fire return intervals, and “all” combinations thereof, FRG replaces the latter with a yet longer [>200 years] fire return interval of “any” severity. A further variation of the 5-class fire regime system retains the same duration for the short-term return interval [0-35 years], but reduces the term of the long-term interval [35-100 years], together with that of the fifth class [100> years] (Agee, 2005). A yet more recent study likewise concurs that whereupon datasets spanning fire size, frequency, intensity, season, and extent are analysed, fire regimes align to one of five classifications. Furthermore, the study demonstrates fundamental constraints within each of the five regime variants, wherein only certain combinations of fire characteristics are possible, described as akin to the ecological trade-offs found in plant physiology (Archibald et al, 2013). Technically, the LANDFIRE fire regime classification system is the most rigorous and fit for future purpose, given that its overall design schematic can accommodate for advances in Earth observation systems and analysis, the latter of which are subject to rapid and ongoing technological advances (Asner et al, 2012; Saatchi et al, 2011; Shugart et al, 2010; Lefsky et al, 2010; Running et al, 2009). However, a “stochastic, spatially complex disturbance process” (Morgan et al, 2001) wildfire manifests with such great variability as to render both FRG, and all the above classification systems unable to quantify fire regimes in absolute terms. Given that most of the fire ecology literature reviewed in this study aligned to the tri-part ‘low, mixed, or high’ severity system, it is this to which the fire ecology case studies, and the wildland urban interface theoretical design works born thereof, are ascribed in this thesis. The method utilizes the fire severity trio as an organising paradigm through which we can interrogate the systemic nature of wildfire (Agee, 1998).

4.2.2 Forests: The Biogeographic Realms of Fire Regimes.

Low-severity, mixed-severity, and high-severity fire regimes are indigenous to the Western United States, which comprised several mid-latitudinal biome [56] types, falls within the Neoarctic [57] biogeographic realm [58] (Udvardy, 1975). More specifically, the fire regimes are endemic to the ecoregions [59] the WWF classifies as Temperate Coniferous Forest, and Mediterranean Forests, Woodlands, and Scrubs (Olson et al, 2001; Olson and Dinerstein 2002, WWF, 2012a, 2012b, 2012c), thus both are subclasses of the dominant terrestrial ecosystem: forests [60] (Pan et al, 2013). Covering an estimated 5.9 billion hectares (ha) in the pre-industrial age (Adams, 2012), and roughly 3.8 billion ha worldwide [30.6% of Earth’s terrestrial surface] today, (FAO, 2015; Lindquist et al, 2012), forests account for 75% of terrestrial gross primary production (Beer et al, 2010), and 92% of biomass worldwide (Pan et al, 2013). A greater volume of CO2 is stored in forest biomass and soils than in Earth’s atmosphere (Pan et al, 2011). Approximately 42% of the sum thereof is stored in living forest biomass, the remainder stored in necromass [61] (Pan et al, 2013).

In consequence of the albedo effect [62], forests sizeably alter the reflectance properties of Earth’s surface (Feddema et al, 2005; Avissar and Werth, 2005), therein further impacting upon atmospheric and climatic conditions, and to the extent that deforestation can indefinitely alter local and global weather patterns (Miller and Cotter, 2013). Peaking at a gross rate of roughly 16 million ha p/a [8.3 million ha p/a net whereupon expansion through planting and/or natural processes is accommodated for], deforestation was at its highest in the 1990s, whereas today the gross has dropped to an estimated 13 million ha p/a [net 5.2 million ha p/a] (Adams, 2012). In totality, the latest global assessment estimates that 129 million ha of forest has been lost since 1990 (FAO, 2015). However, the biological integrity of the world’s forests is varied: whereas some remain in pristine condition, others are in various states of degradation, the former resilient to disturbance events, the latter less so.

The distribution of the world’s forests is coupled with the climate (Davis et al, 2001; Davis and Shaw, 2001; Jackson et al, 2000; Prentice, Bartlein, and Webb 1991). Multiple studies evidence that, as average global temperatures rise, treelines are shifting in response (IPCC, 2007; Parmesan, 2006); in some instances, headed upwards (Liang et al, 2016), in others polewards (Iverson and McKenzie, 2014; Zhang et al, 2013), and in others still, longitudinally (Fei et al, 2017). Several-fold are the biological reproductive processes and ecological phenomena that enable flora to migrate by means of mitigating shifting local and global environmental conditions, and details thereof will be discussed in the following chapter.

A recent study of ground-based surveys and satellite imagery indicates there to be approx. 3 trillion trees worldwide, of which 1.30 trillion are in the tropics and sub- tropics, 0.74 trillion are in the boreal forests, and 0.66 trillion are in the temperate regions (Crowther et al, 2015). 93% of the world’s forested areas are comprised primary or secondary forest, the former that which has been subject to comparatively few human disturbances, the latter that which has naturally regenerated (FAO, 2015). The remaining 7% of forests are planted (Ibidem); therein generally contain sizeably less biodiversity than natural forests. An estimated 80% of living forest biomass worldwide is aboveground, with the remaining 20% belowground (Cairns et al, 1997; Jackson et al. 1996, 1997). However, the ratios thereof vary considerably from one ecoregion to the next. Whereas the percentage of belowground living biomass is relatively high in tropical deciduous [25%] and boreal forests [24%], it drops to 16% in tropical evergreen forests, and to just 15% in temperate coniferous forests (Jackson et al, 1996). Disturbances, both natural and anthropogenic, are so very wide-spread within forests that at any given time, a sizeable portion thereof will be recovering from one or several such events. A relatively recent study indicated that 60%> of the some 3.8 billion ha of forests worldwide are in the process of recovering from a disturbance (FAO, 2006).

In totality, 40% of the Earth’s terrestrial surface is covered by fire-prone ecosystems (Bond, Woodward, and Midgley, 2004), of which forests are the dominant biome- type. Ecosystems affiliated with low-severity fire regimes are more commonly found on the Western side of continents within 30-40° latitude and with a Mediterranean climate, including California and Northern Baja, Central Chile, Mediterranean Basin, Western Cape South Africa; and Southwest and South Australia. Whereas, ecosystems affiliated with mixed and high severity fire regimes are associated with both these, and with other regions worldwide, including in the northern latitudes. For example, analysis of data including charcoal distribution in lake sediments evidences both mixed and high severity fire regimes indigenous to the Artic tundra, such that vast swathes of forest burn with a periodicity spanning decades to millennia (Sheng Hu, 2015; Kolden and Rogan, 2013; Mack et al, 2011; Jones et al, 2009). Therein, forests constitute the foremost spatiotemporally dynamic of the several primary terrestrial biome-types; so intimately integrated to Earth’s abiotic systems as to be inseparable thereof, and no less so than in relation to fire [Fig. 22].

4.2.3 Forests and Fire Regimes of the United States

Prior to British colonization of the Americas [63], forests covered an estimated 420 million ha [approx. 46%] of the United States, of which 104 million ha has subsequently been subject to land-conversion, leaving an estimated 310 million ha of forested land [approx. 33.8%] (Oswalt, 2012) [Fig. 23]. Together with the forests of Central America, this constitutes 17% of forests worldwide (Pan et al, 2013). 56% of U.S. forests are under private ownership, of which 75% is divided between several million individuals and families, with the remaining 25% owned by corporations, partnerships, and tribes (Oswalt, 2012). 75% of forests under public ownership are in the western United States, of which the majority have protected status (Ibidem), whereas the contrary is true of forests under private ownership.

Omitting agricultural, urban/developed, and barren lands, 48% [240 million ha] of Federal and non-Federal territory in the conterminous United States falls within the historical range for biomass composition, structure, and fuel loadings. Whereas, 38% [190 million ha] lies moderately outside thereof, with 15% [73.6 million ha] of lands classified as significantly outside of their historical range (USDA, 2015). Given biotic assemblages are coupled to their fire regimes, the above statistics essentially indicate that, whereupon wildland fire takes hold in these lands, currently, roughly 50% are likely to manifest levels of fire severity on a par with that which the historical record suggests to be probable. However, the biomass composition, structure and fuel loadings of the other 50% or so of lands suggests that, to a lesser or greater extent, whereupon wildfire manifests there is an increased probability that the level of fire severity may vary from the historical record.

63% of the vegetated lands of the conterminous United States [640 million ha] historically experience frequent fires [0-35-year occurrence] (Hardy et al, 1998a, 1998b, 2000), and it is at these sites that the fire regimes, therein fire behaviour and severity, are changing most (Morgan et al, 1996, 2001). Distinct patterns are starting to emerge in the spatiotemporal shifts underway (Parisien et al, 2012), and these are born of the tight coupling between the variables within the regional bio-chemical complexes. The data evidences that some biome types are currently experiencing a greater degree of change than others, and in particular dry forests, woodlands, and grasslands, each of which are native to the western United States. Given that 1% of all fires in the conterminous United States currently account for 98% of the area burned (Strauss et al, 1989; Morgan et al, 2001), the recent data suggests we might reasonably assume this ratio will alter in the years and decades to come, and that the biome-types indigenous to the study region are more, not less likely to manifest change, and despite the fact that, historically, fire has been the foremost natural disturbance in the region (Spurr and Barnes, 1980).

Mediterranean Forests, Woodlands, and Scrubs

Globally, this ecoregion type sustains some 10% of flora species (WWF, 2012b). Comprised an assemblage of woody, herbaceous, and graminoid [64] plants, Mediterranean forests, woodlands, and scrubs are amongst the foremost fire-prone ecosystems on Earth (Keeley et al, 2012). Climatically, long, hot, and dry summers are juxtaposed against mild and relatively wet winters, the latter enabling the accumulation of biomass, the former converting the sum thereof to tinder-dry fire- ready fuel. Manifold functional traits [65] that enable flora to flourish within ecosystems classified under this ecoregion type, and in particular those that enable coexistence with fire, pre-date the onset of the Mediterranean climate, and, within the case study region, by a period of at least 50 million years (Lamont and He, 2017; He, Lamont, and Manning, 2016). Covering an area of 4.7 million ha, California Chaparral and Woodlands is one of five variants of the Mediterranean shrublands ecoregion worldwide (WWF, 2016b). Ascribed critical/endangered status, anthropogenic threats to the region include overgrazing by cattle; logging; introduction of exotic species; dams and hydrological disruptions of various kinds; urbanisation; and myriad actions that affect the historical fire regime.

Temperate Coniferous Forest

Populated by members of the genus Pinus, the fossil record and phylogenetic [66] analysis pinpoint the emergence of the taxonomic family that populate this ecoregion to the Middle Carboniferous [340-310 mya] (He et al, 2016; Wang and Ran, 2014). A fire-preserved Pinus fossil dated to the Early Cretaceous [140 – 130 mya] evidences fire as the evolutionary driver of functional traits in coniferous forests since at least that date (Falcon-Lang, Mages, and Collinson, 2016). Generally, the climatological realm in which this highly flammable evergreen ecoregion persists is temperate, known for warm summers and cool winters, the latter endowed with suffice levels of precipitation to sustain the highest levels of biomass to be found in a terrestrial biome (WWF, 2016a). Compared to some other forest ecoregions, such as tropical rainforests, Temperate Coniferous Forests are structurally simple, generally comprised just two layers: understory and overstory. Covering an area of 29.5 million ha, the Pacific Temperate Rainforests are one of seven variants of the Temperate Coniferous Forest ecoregion worldwide (WWF, 2016a). Orientated North to South, The Rocky Mountains stretch from New Mexico to Canada, which at a distance of some 3,000 miles makes them the second-longest mountain range in the world. Centred on the Yellowstone Plateau, the forests of the South Central Rockies span nearly 16 million ha, and are home to world’s oldest and most renowned national park. Climatically, these forests experience the strongest seasonality of any in this ecoregion type, juxtaposing relatively brief but dry, continental-style summers against, at times, bitterly cold, and in higher elevations, snowy winters, though the impact thereof varies sizeably with the topology of the landscape. 83% protected (Hoekstra et al, 2010), the South Central Rockies forests are classified as vulnerable, with logging, hard-rock mining, oil and gas development, WUI development, livestock grazing, and the introduction of exotic species listed as the primary threats to the integrity of its biotic assemblages (WWF, 2016c).

4.2.4 Fire Regime Biotic Schemata

The three quantitatively and qualitatively distinct fire regime classes of low-severity, mixed-severity, and high-severity exhibit intrinsic behaviours that are shaped by energetic constraints (Archibald et al, 2013) [Fig. 24]. Fundamentally, fire regimes express a dichotomy between ecoregions dominated by woody vegetation, and those dominated by graminoids. The latter taxon, which is populated by an estimated 12,000 species worldwide (Christenhusz and Byng, 2016), is comprised a mix of annual and herbaceous perennial grasses (OST, 2017) of which the physiology enables coexistence with fires as frequent as every 1-3 years (Archibald, 2013). In contrast, ecoregions dominated by woody species, such as members of the genus Pinus, are physiologically adapted to fires of frequencies ranging from decades to centuries.

Analysis of datum documenting historical fire regimes, together with the topography of the sites thereof, indicates a general correlation between fire regimes and landscape gradients. Fuel-driven, the low-severity fire regime is affiliated to lower elevations (Agee, 1997). Weather-driven, the high-severity fire regime is affiliated to higher elevations (Bessie and Johnson, 1995; Agee, 1997). Whereas, the mixed-severity fire regime can be fuel or weather driven, but is most commonly found on mid to high elevations (Agee, n.d). While the physiological traits that enable flora to coexist with fire predate the climatic conditions of the present (Lamont and He, 2017; He, Lamont, and Manning, 2016), the synecological character of the symbiotic relationships that have evolved between flora and fire are evident whereupon historical fire regime records are interrogated. In the United States, and the wider Neoarctic realm, the historical fire regimes of the Holocene [11.7 tya to present], and their affiliated dendrological chronologies, illustrate an affinity between biotic assemblages predominantly populated by Oak [including Quercus garryana and Quercus wislizeni], Ponderosa Pine [Pinus ponderosa], Mixed Confer, and to a lesser extent Douglas-fir [Pseudotsuga menziesii] and the low-severity fire regime (Agee, n.d, 1998; Gray and Riccius, 1999; Foster, 1998). In contrast, species including Subalpine fir [Abies lasiocarpa], Whitebark pine [Pinus albicaulis], Mountain hemlock [Tsuga mertensiana], and Lodgepole pine [Pinus contorta] (Pierce and Taylor, 2011; Agee, n.d, 1998) are associated with the mixed-severity and high-severity fire regimes.

Thus, while fire’s inherent stochasticity limits our capacity to predict precisely when and where a wildland fire will become manifest, whereupon we know a site’s topography, and more particularly, its elevation, together with its biomass composition, and we triangulate this datum with meteorological forecasts, we can reasonably assess the probability of a low, mixed, or high-severity fire at a given site. If expressed through the medium of a pyro-polygon, a fire behavior triangle forms comprised fuel, weather, and topography (Agee, 1997, 1998).

In order of historical frequency, the fire regimes are described below:

Low-severity

The low-severity fire regime is associated with biomes comprised open canopy structures (Jenson and McPherson, 2008) of limited tree density and basal area [67] (Agee, n.d; Baker 2017), therein forests, shrublands [68], and grasslands of which the biomass structure has a lower overall load of hydrocarbons than those affiliated to mixed and high severity regimes. This regime is characterised by very frequent, but relatively low-intensity fires that swiftly burn through understory vegetation, such as leaf-litter, logs, and low-lying branches, as well as snags [69], but leave soils unheated, and the overstory vegetation generally unaffected. Hence, the low-severity fire regime is sometimes referred to as non-lethal, killing no more than <20% of basal area (Agee, 1993, 1994). Spatially, low-severity fires tend be small, (Agee, 1998, 2005); burning out comparatively quickly compared to mixed and high severity fires. Visually, the low-severity fire regime creates relatively homogeneous landscapes, within which the legacy of one fire subtly blurs into that of the next (Agee, 2005). While individual fires tend have little apparent effect on the biome, cumulatively they sustain the landscape mosaics that are fundamental to the survival of the species adapted to this regime type (Agee, 1998). Dendrochronological techniques, including cross dating of tree rings and fire scarring [70], have enabled the creation of hypothetical reconstructions of historical low-severity fire regimes. While the spatial complexity of fire spread within the landscape renders the accuracy thereof uncertain (Jensen and McPherson, 2008), analysis of datum relating to the case study area suggests that once fire has burned through an area it will rarely return within 3 years (Heyerdahl, 1997; Agree. n.d), the exception thereto being some graminoid-dominated ecoregions (Archibald, 2013).

Mixed-severity

The mixed-severity fire regime is associated with biomes comprised landscape mosaics that evidence combinatory fire severities; a hybrid of low, moderate, and high-severity fires; thus the regime constitutes the foremost complex of the fire regimes (Agee, 2005). While technically, the mixed-severity fire regime is associated with an average 25-75% biomass topkill [71] (LANDFIRE, 2014), its legacy is a highly heterogeneous, patchwork quilt of a landscape comprised juxtapositions of burned and unburned vegetation. Common in the mid-elevation forests of the western United States (Jensen and McPherson, 2008; Agee, n.d), the mixed-severity regime supports floral and faunal diversity, as it provides a rich and complex array of biotic habitats, both living and dead. Many species benefit from the abundant coarse woody debris [72] [CWD] born both of mixed-severity fires past (Agee, 2005), together with wide- ranging other biotic disturbances, including windfall, heat stress, and insect outbreaks. In the case study region, this regime averages an MFRI of 25-75 years (Agee, n.d). The timing and behaviour of the mixed-severity fire regime is shaped by a generally wider spectrum of influences than its low and high counterparts. While biomass condition [fuel moisture-levels, structure, and composition, the latter two both shaped by fires past] and weather are generally cited to be the foremost influencers in this regime, topography also plays a fundamental role (Jensen and McPherson, 2008; Agee, 2005; Schoennagel et al, 2004). The mixed-severity fire regime constitutes a microcosm of the wider fire regime family, sometimes manifesting low-severity patches on lower elevation North/East facing slopes, and high-severity patches on higher elevation South/West facing slopes (Agee, 2005; Taylor and Skinner, 1998).

High-severity

The high-severity fire regime is principally associated with biomes that are largely comprised swathes of even-aged stands [73] at the late-successional stage, therein biomass structures that are susceptible to crown fire behaviour (Agee, 1998). The exception thereto is subalpine forests where harsh environmental conditions and limited seed distribution hinder regrowth in the aftermath of a fire, and/or other natural disturbance (Agee, n.d). Occurring at higher elevations, and widespread in boreal (Agee, 1998; Johnson, 1992), subalpine (Agee, 1998; Agee and Smith, 1984), and wet coastal forests (Heinrichs, 1983) of the western United States, this regime is characterised by relatively infrequent, but high-intensity fires, of which the MFRI historically averages 75-100 years (Agee, 1998). Usually weather-driven (Agee, 1998, 1997; Bessie and Johnson 1995), the high-severity fire regime typically burns through many thousands of hectares (Agee, n.d), and especially whereupon winds widely distribute sparks and firebrands, therein igniting spot fires far ahead of the fire front (Scott and Reinhardt, 2001). Extreme weather conditions tend result in high-severity fires burning through biomass at all successional stages (Bessie and Johnson, 1995; Romme and Despain, 1989; Romme, 1982). Whereas, more usual weather tends result in high-severity fires slowing or stopping at the boundaries of early successional-stage stands [i.e. where the fuel-load lessens] (Agee, 1998; Romme and Despain, 1989; Despain and Sellars, 1977). In either scenario, the tree mortality rate is high, and especially given that the physiology of the species affiliated to this regime is generally not adapted to survive the intensity of the fires it commonly manifests (Agee, 1998). Hence, why the high-severity fire regime is associated with stand-replacing fires, and thus sometimes referred to as lethal (Agee, 2005). However, whereupon the regime resides within the usual parameters of the historical MFRI, ‘lethal’ is somewhat of a misnomer, in so far as the biota affiliated to this regime has adapted to reproduce in the aftermath of a fire (Hutto, 2008).

4.2.5 The Flaming Quartet: Four Types of Fire

Reverting to the words of “fire laureate” (Omi, 2005, p.84) Stephen Pyne, fire’s “character” is formed by its “context”, and no less so than in regard of the biomass stratum through which it is burning. While the National Fire Danger Rating System recognises four types of fire, fire ecologists generally describe three (Scott and Reinhardt, 2001): Ground, Surface, and Crown, with Spot the lesser referenced of the quartet. In turn, these are attributed to ground, surface, or aerial fuels that are grouped into four size classes: Grass, Litter, and Duff of less than 1⁄4 inch [0.64cm] diameter; Twigs and small stems of 1⁄4 - 1 inch [0.64 - 2.54cm] diameter; Branches of 1-3 inches [2.54 – 7.62cm] diameter; and large stems and branches greater than 3 inches [7.62cm] diameter (Jenkins, 2004). More generally, fuels are referred to as being light or heavy, the former attributed to fuels with a diameter of 1⁄2 inch [1.27cm] or less, thus accommodating grass, litter, duff and smaller twigs and stems, the latter referring to larger twigs, stems, branches, logs, and tree trunks (Schneider and Breedlove, n/d). Light fuels dry out faster than heavy fuels, which in combination with their higher surface area to volume ratio [74], thus higher volume to oxygen ratio means they also ignite several times faster, as do flat fuels (Ibidem). Timelag refers to the time it takes for “a dead fuel particle to lose or gain 63% of the difference between its initial moisture content and its equilibrium moisture content at a constant temperature and relative humidity” (Knapp, 2007). Each of the four fuel size classes has an affiliated timelag: less than 1⁄4 inch [0.64cm] diameter = 1 hour timelag fuels; 1⁄4 to 1 inch [0.64 – 2.54cm] diameter = 10 hour timelag fuels; 1-3 inch [2.54 – 7.62cm] diameter = 100 hour timelag fuels; and greater than 3 inch [7.62cm] diameter = 1000 hour timelag fuels (Jenkins, 2004). Furthermore, size and shape underpin how and why fuel stuffs may move within a wildfire. For example, the flatness and surface area to volume ratio of pine needles endows them with aerodynamic properties that increase the probability that, upon ignition, they will be carried, as firebrands, within a wildfire plume, to thereon fall several miles from the main fire (TDA, 2002). In contrast, though too heavy to travel any great distance, pinecones are perfectly shaped to spread fire upon igniting, falling and tumbling down a slope (Ibidem).

In toto, there are seven fuel characteristics that are indicative of how a fire will behave upon ignition: as above, fuel size and shape, and moisture content, together with loading, compactness, chemical properties, horizontal continuity, and vertical arrangement, of which the first five are the foremost influential (Ibidem). Fuel loading refers to the dry weight of combustible fuels per unit area [i.e. kg/m 2 or t/ha] (CIFFC, 2003), and is usually restricted to surface fuels of less than 3 inches [7.62cm] diameter (Jenkins, 2004). Fuel loadings vary greatly across landscapes: grassland areas typically produce 1-5 tons per acre [t/a]; shrublands 20-40 t/a; logging slash 30- 200 t/a; and timber 100-600 t/a (Ibidem). However, the highly heterogeneous nature of biotic assemblages manifests significant variations in the quantities thereof. Biochemically, the molecular make-up of plants varies by species, and in some instances, by season. At dry weight plants are principally comprised of carbon. Nonetheless, outside of severe drought conditions, water usually accounts for 80-90% of a plant’s weight (Boundless, 2016). Conversely, stripped of its former capacity for homeostasis, though levels rarely drop below 3-4%, dead biota is subject to significant fluctuations of moisture content (Jenkins, 2004). Thus, within ecosystems, and especially those that are located in the mid to high latitudes where seasonality is greatest (NASA, 2017), biota’s ratio of CO2 to H2O is one of relative spatiotemporal flux. Furthermore, while, at dry weight, all terrestrial vegetation largely constitutes the organic [carbon-based] compound cellulose [C6 H10 O5], structural composition, therein density, varies across species. For example, at dry weight Spruce [Picea glauca] is comprised 39.5% cellulose, Eucalyptus [Eucalyptus camaldulensis] 45%, Jack Pine [Pinus banksiana] 45.2%, and Quaking Aspen [Populus tremuloides] 50.2% (GIT, n.d). Hence, while cellulose has a melting point of 467°C (Dauenhauer, Krumm, and Pfaendtner, 2016), content variance across taxonomic groups manifests relative divergence in the combustibility of woody species.

Compactness refers to the spacing of fuel particles [i.e. forest litter], of which the closer the proximity, the lower their surface area exposure to air flow, thus the slower the heat spread, as there is less oxygen to sustain the combustion process (Barr and Eversole, 2003). In consequence, fire usually spreads more slowly when burning through a bed of pine needles than in open grassland (TDA, 2002). The fifth fuel facet that significantly influences wildfire behaviour is referred to as chemical properties, but specifically relates to whether vegetation contains volatile substances, such as oils, resins, wax and pitch, that act as natural fire accelerants, or contra, constitute a high mineral content, which subdues fire spread and intensity, thus act as fire retardants (Jenkins, 2004). Examples of plants that contain highly flammable chemicals include the aptly named Burning Nettle [Urtica Urens], California Sagebrush [Artemisia californica], Indian Tabacco [Nicotiana bigelovvi], Cypress [Cupressus sp], and Juniper [Juniperus sp]. Thus, metaphorically speaking, when it comes to assessing the combustibility of forests [and other fire-prone habitats] it is imperative to see the [molecular-level composition of] the wood for ‘the trees’.

In addressing the final two fuel characteristics of consequence in anticipating a wildfire’s behaviour we need move from the micro to the macro level: to the level of the landscape. Horizontal continuity refers to the distribution of fuel across the horizontal plane and encompasses all levels of the fuel complex. The characteristic is indicative of where a fire will spread, the rate thereof, and whether it will burn through surface fuels and/or aerial fuels (Ibidem). Whereupon fuel is distributed continuously fire may spread uninterrupted for miles on end. Per contra, landscapes that are strewn with patches of bare mineral soil, rock outcroppings, and other natural, and anthropogenic barriers tend not sustain fire for great distances, with the exception thereto being the advent of strong winds that facilitate fire spotting (TDA, 2002). Horizontal continuity is of particular significance in forest fires, because the distribution of aerial fuel in the canopy impacts upon moisture content in surface fuel and on near-surface wind speeds, therein fire’s rate of spread (Ibidem). Fuel’s perpendicular continuity [vertical arrangement] influences the biomass strata to which a fire ascends: when vertically continuous, fuel forms a ladder that enables fire to rise up to the canopy, but, in the absence thereof, a fire will remain at the surface. In the instance that wildfire meets uninterrupted fuels its forward rate of spread can reach speeds of up to 10.8km/p/h [6.7 mph] in forested habitat, and 22km/p/h [14 mph] in grasslands (Knowling, 2016). Though theoretically, anywhere between 0-100% of fuel may burn, fire tends consume 5-25% of standing trees in a forest fire, between 5- 95% of vegetation in a shrubland fire, and, if cured, approaching 100% in a grassland fire (Jenkins, 2004).

Ergo, biota’s biochemical composition and distribution defines the probability of a wildfire’s ignition, likely spread and intensity thereafter, together with the likelihood of spotting, torching, and crowning. Each of the 7 fuel characteristics influences between two and four of these behaviours: ignition and intensity is impacted by moisture content, loading, compactness, chemical content, and size and shape; spread by all the above, and horizontal continuity; spotting by moisture content, size and shape and horizontal continuity; and torching and crowning by vertical arrangement and moisture content.

In order of elevation, the four fire types are described below:

Ground

‘Slowly, but surely’ is the mantra of ground fire, which gradually works its way through the densest of the biomass stratum: organic soils, deep duff [75], roots, rotten buried logs, etc (Jenkins, 2004; Scott and Reinhardt, 2001). While slow, ground fire persists, often smouldering or creeping, through comparatively high moisture loads (Jenkins, 2004; Frandsen, 1987, 1991), altering soil structure and hydrology, while causing injury to above ground biomass (Scott and Reinhardt, 2001). Usually ignited by a surface fire, once started, they are difficult to assess, thus control, as they tend have little, if any flame and smoke (Schneider and Breedlove, n.d).

Surface

Burning through grass, duff, understory shrubs, downed and lower branches, and tree stumps (Schneider and Breedlove, n.d), surface fires are the most common of the four fire types (Jenkins, 2004). Their rate of spread and intensity depends upon the biomass composition, both in structure and species. However, the fuels that feed a surface fire are less compacted than that of a ground fire, which is one of several factors that underpin a differential in the rate of spread between the two. In the absence of aerial fuels (i.e. forest canopy), the behaviour of a surface fire is subject to elevated drying by solar radiation and heating, and to stronger wind speeds (Ibidem). Unlike ground fires, surface fires are easy to spot, but their controllability is dependent on the specifics of the site and meteorological conditions (i.e. wind speeds and direction, humidity and fuel moisture content).

Crown

If ground fires are akin to a flaming tortoise, crown fires are the hare. Occurring in diverse forest ecoregions of the United States (Agee, 1996), crown fires typically spread several times faster than surface fires (Rothermel, 1983), and yet faster still than ground fires. Known for being unpredictable and volatile (Schneider and Breedlove, n.d.), crown fires burn through the canopy, consuming fuels both live and dead (i.e. foliage, lichen, and fine branchwood) of which the moisture content is higher, but the density lower than in the fuels found at the surface (Scott and Reinhardt, 2001). Comprised three subcategories, crown fires are classified as passive, active, or independent (Ibidem; Van Wagner, 1977):

Passive crown fires exhibit relatively diverse behaviour, which in some instances becomes manifest as a single tree torching out, and in others extends to a group of several (Scott and Reinhardt, 2001). Particularly common in forests with an understory comprised share-tolerant conifers, though, in and of themselves passive crown fires are small in size, the radiation they emit tends accelerate the flame front’s rate of spread, while their embers are prone to igniting spot fires downwind (Ibidem).

Active crown fires engulf both the surface and canopy, to form a continuous wall of flame that rises from the fuel bed upwards (Ibidem). Radiation from burning surface fuels drives crowning, and vice versa, as radiation and spotting from crowning accelerates the rate of spread of fire at the surface (Ibidem).

Independent crown fires burn in the absence of surface fires, occurring rarely (Van Wagner, 1993), they form whereupon high winds lick crowning on steep slopes covered in vegetation with low moisture content (Scott and Reinhardt, 2001).

Passive, active and independent crown fires are not mutually exclusive, as factors including spatial variability in the landscape and temporal variability in winds speeds causes crown fires to metamorphose between the three variants: a fire behaviour which is termed an intermittent crown fire (Ibidem). While, technically, the term canopy fire applies to scenarios in which the entire canopy stratum is alight [active fires], with ‘crown’ applying to passive fires, fire ecologists tend to apply the latter term more or less generically (Ibidem). Within the study region, the Chaparral shrublands of California are particularly associated with crown fires, which are recognised as having played a significant role in the evolution of the region’s biota (USGS, 2015).

Spot

Propagated by flying or rolling sparks or embers, spot fires develop beyond the perimeter of the main fire (NOAA, 2016) and commonly manifest in consequence of unstable meteorological conditions [i.e. wind, fire whirls, and/or convection within wildfire plumes] (Andrews 1996; Schneider and Breedlove, n.d). Several fuel types are prone to spotting, including moss and lichens in the canopy, and needles, leaf litter, cured grasses, rotten and dead snags and foliage, and logging slash at the surface (Jenkins, 2004). Spotting can occur over short, intermediate, or long distances, ranging from several meters to tens of kilometres beyond the main fire (Albini, 1983).

But, no matter the many potentialities in a wildfire’s behaviour, the steadfast fundamentals of thermodynamics are of universal applicability whereupon postulating on possible trajectories: water vaporizes at 100°C, therein whatsoever its previous state, whereupon woody matter is exposed to temperatures thereto it will dry to a fire- ready state; the first phase of combustion, pyrolysis [76] occurs in woody matter at 230°C, releasing flammable gases in the process; at 380°C woody matter heats sufficiently to smoulder, thereon igniting at 590°C; 800°C is the temperature to which, through the process of heat transfer [convection and radiation], a wildfire front can preheat the atmosphere before it (Knowling, 2016). Hence, when bearing witness to these, and to many other physical processes that are bound to particular thermodynamic parameters, we can read a wildfire’s behavior in a not dissimilar way to that of an art historian decrypting the meaning of an Old Masters painting: the picture before us is encoded with signs and symbols that are known to the expert eye.

4.2.6 Pyrophyta: Systema Naturæ per ignem regnis [77]

“The typical pattern of development varies from system to system, but it is the sharing of different states of successional development that maintains diversity in virtually any ecosystem. The natural forest becomes a tapestry of patches in different stages of succession, and hence a tapestry of diversity” Levin, 1999.

Biota exhibits wide-ranging evolutionary responses to fire. Collectively known as ‘pyrophytes’, some species selected morphological means of co-existence, while others chose biochemical, physiological, phenological, and/or behavioural. Pyrophytes are classed as either ‘passive’ or ‘active’, depending on whether their fire- defences are principally armoury, as is the case for the Cork oak (Quercus suber) or biochemical in nature, as for the California lilac (Ceanothus L.). In some instances, pyrophytes have synced their reproduction with the frequencies, intensities, and resultant severities of wildfire, and these are classed ‘pyrophiles’, of which Knobcone pine (Pinus attenuata) is one. However, whatsoever functional traits were selected, as is always the case in biology, a trade-off was involved. While many are the means by which plants may be classified, the foremost relevant with respect to this study [78], (sensu Vogl, 1974) assigns flora to one of five “modes of persistence” (Rowe, 1983, p.140): Invaders; Evaders; Avoiders; Resistors; and Endurers.

In order of the fire frequency with which they are principally affiliated, the functional groups are described below:

Endurers

Dubbed “phoenix species” (Ibidem, p.144), endurers regenerate upon the passage of fire. Their resilience resides in their perennating parts [rhizomes, roots and root crowns], which, excluding in the instance of ground fires, tend be protected by both humus [organic] and mineral soils (Ibidem). Survival is coupled to the vertical positioning of the belowground organs, the depth of which correlates with the proportions of the humus stratum (Ibidem). Ecoregions that incur low frequency, but high-severity fires will tend accumulate more biomass, therein deeper duff, of which a consequence is the “zone of maximum biological activity and nutrient release” (Ibidem, p.145) moving upward, with the perennating parts following suit. Given that mineral soils are relatively retardant to fire, but organic soils are significantly less so, a result thereof is that endurers may be all but eliminated from a site that experiences a high-severity fire. Thus, their functional traits primarily befit endurers to the low and mixed-severity fire regimes, and to a high rate of fire frequency. The exception is a subclass that Rowe describes as “superspecies” (Ibidem, p.145), which, though primarily surviving by means of resprouting, are nonetheless prevalent in ecoregions experiencing a range of fire frequencies. Some species, including many fire-adapted shrubs, have both the capacity for resprouting and postfire seedling recruitment. Termed ‘postfire facultative seeders’ (Keeley, 2012), we might posit these to be ‘superspecies’.

Evaders

Also referred to as “bankers” (Rowe, 1983, p. 142), evaders have evolved the ecological equivalent of an insurance policy. While the species within this functional group succumb to fire, they do so having stored seeds that rapidly germinate upon its passing. One variant store their seeds in the canopy, the other in organic and mineral soils. When stored aerially, serotinous [79] cones and fruits shield the seeds from the flames. When in the ground, the soil itself performs this task. In both instances, an abundance of seeds remain dormant until fire-related mechanisms release them from their protective casing (Ibidem). Their survival strategy harnessing the abundance of space and nutrients within post-fire landscapes, evaders are adept at enduring high- severity fires. Like endurers, evaders fall into two subclasses: one comprised semi- tolerant and shade-tolerant perennials [80], the other shade-intolerant ephemerals [81], the former subclass storing their seeds over prolonged periods of time, the latter in “one shot” deposits during “brief post-fire flowerings” (Ibidem, p. 142). Thus, the subclass populated by perennials is commonplace in ecoregions where fire frequency is low, and vice versa. While Rowe proposed evaders as being primarily adapted to short and intermediate fire cycles, subsequent studies evidence that the mosaic nature of mixed and high severity fires enables this group to co-exist with long fire cycles (Wallace and Christiansen, 2004). Furthermore, other studies assert that whereupon fire becomes especially frequent some perennials within this class have not sufficient time to accommodate the length of their reproductive cycle (Nijhuis, 2012).

Resisters

In keeping with Rowe’s analogies, in a nod to novelist Roderick Thorp’s character John McClane, we might dub resisters ‘die hards’. While juvenile plants have low resistance to fire, adults are protected by an array of pyro-armoury. Resisters defences include thick bark to protect against ground fires; self-pruning and peripheral foliage that help prevent against fire scaling from surface to canopy [fire ladders] (Rowe, 1983); and in some species, biochemical mechanisms, such as flammable oils in their exterior parts, which create the biological equivalent of a flashover [82]. Resisters resilience varies from one species to another, with those of the genus Pinus and Eucalyptus amongst the most [die] hardy. Factors that contribute to the heterogeneity in resilience limits amongst species include variations in the distribution of defences. For example, whereas some resisters select for defence against surface fires, others are adapted to withstand crown fires, the former’s pyro-armoury restricted to the base of its trunk (Jackson, Adams & Jackson, 1999), the latter’s protecting the plant to its crown (Pausas, 2015). Rowe suggested that resistors are principally affiliated to fire cycles of intermediate duration. This notwithstanding more recent studies have illustrated some species within this functional group are well adapted to relatively frequent fires (Ibidem).

Invaders

Their propagules dispersed by the wind, invaders are the “fugitives” (Rowe, 1983, p. 141) of the functional quintet. Prolific in their populous, invaders are amongst the first on the post-fire scene. Present in the aftermath of fires of any frequency and intensity, invaders may be found in quantities large or small in wide-ranging ecoregions (Ibidem). Generally shade-intolerant species, invaders tend flower and fruit in abundance in the period immediately following a fire, thereon dwindle as other functional groups re-establish. Thus, their populations pulse with perturbations, including both wildfires and other categories of natural hazards, including volcanic eruptions.

Avoiders

The antithesis of the above four functional classes, avoiders are the laggards of the ecological arena. Featuring no known evolutionary adaptions for co-existence with fire, avoiders are late successional species within landscapes where fire is infrequent. Mainly shade-tolerant mesophytes [83], most often, their propagules will only persist where invader, evader, resistor, and/or endurer species are well established, therein humus and aerial coverage have accumulated (Ibidem). Hence, while all flora persists in a state of symbiosis, avoiders tend do more so than do most others.

However, an interdisciplinary researcher working within and across the disciplines of botany, forestry, and landscape ecology, the creator of the above classification system, the late Stan Rowe, recognised that functional groups persist within an “open geographic system” (Ibidem, p. 147). Therein, in and of itself, autecological [84] information provides not a flawless means of modeling future ecosystemic states. Rowe also acknowledged the possibility that, in some instances, functional traits that are advantageous in fire-prone landscapes may, in the initial instance, have evolved in response to other environmental factors. Citing examples, Rowe referenced a study that posited that high concentrations of volatile substances in a species, which, under his system, fell into the resistor group, might primarily have evolved as a herbivory defense (Chapin and Van Cleve, 1981). He also highlighted how many species exhibit not one, but two or more stratagems for coexisting with fire, therein while assigned a structural-functional type, are akin to hybrids (Rowe, 1983). Nonetheless, Rowe’s ‘modes of persistence’ provides a scientifically plausible, coherent and elegant framework to use when examining the nature of the relationships at play within fire- prone landscapes. While all five of the functional types will be referenced in later chapters, those of primary interest to this study are the three that feature fire-response mechanisms: endurers, evaders, and resisters, the former two of which may more generally be described as fire-tolerant, the latter fire-resistant, and species with neither properties, as fire-intolerant.

4.2.7 On the Tendency of Species to form Resilience [85]

“Adaption is the act of bending a structure to fit a new hole. Evolution on the other hand, is a deeper change that reshapes the architecture of the structure itself – how it can bend – often producing new holes for others”. Kelly, 1994.

In diagrams of the phylogenetic tree of life, fire is conspicuous by its absence, and particularly so given that, in some species, pyroendemics, both seedling germination (Moreira and Pausas, 2012) and successful recruitment, are limited to “immediate postfire environments” (Keeley and Pausas, 2016, p. 1). However, dated molecular phylogenetic [86] and Bayesian [87] ancestral state reconstructions, together with the fossil record, make evident the emergence of fire-related functional traits within the coevolution of fire and flora (He and Lamont, 2017, 2016). Furthermore, enduring, evading, and resisting were the first three of the five modes of persistence to evolve.

In no particular order, the five foremost functional traits of fire-resilient species are described below:

Pyriscence

Originating in the common ancestor of the conifer (Ibidem), pyriscence, a type of serotiny [Fig. x], is a functional trait found in endurer species in ecoregions including Mediterranean forests, woodlands, and scrubs, and temperate coniferous forests. Evident in members of the genus Pinus since the Late Carboniferous [332> mya], pyriscence evolved in angiosperms no later than 74 mya (Ibidem). Seed release in pyriscent species is fire-stimulated. In conifers and some angiosperms the release mechanism is the melting of the resin that holds the protective parts together (Ibidem) [Fig. 25]. In the genus Pinus, pyriscence is a heritable trait, which recent studies suggest is encoded via genes into enzymes, and of which the expression level varies throughout the evolution of species (Ibidem). Thus, the temperature at which the resin in pyriscent cones and fruits melts is variable over time, which, over generations, enables pyriscent species to adapt to changing fire conditions [i.e. match the melting point to the intensity of shifting fire regimes, and in turn, the environmental scenarios as underpin them]. However, some species stoke their own fires, having tightly coupled pyriscence with abscission. In this instance, the litter produced therefrom is highly flammable, an example thereof being the genus Banksia (He, Lamont, and Downes, 2011). Regardless of whether abscission is present or otherwise, in the Northern Hemisphere, together with an abundance of nutrients, light, and space, upon germination, pyriscent species benefit from the removal of allelopathic leaf litter, therein, part removal of pre-existing hierarchical ecological structures that may inhibit their species populous.

Pyrogermination

Several-dozen plant families found in fire-prone Mediterranean-type climate (MTC] ecoregions possess fire-stimulated germination (Baskin and Baskin, 2014; Keeley, 2012: Keeley et al, 2012), which for the purposes of this thesis shall be termed ‘pyrogermination’. Potentially an exaptation derived from earlier evolutionary changes in primary metabolism (He and Lamont, 2017), pyrogermination has been identified in 2,500 species (Bradshaw et al, 2011) spanning a “vast phylogenetic range” (Keeley and Pausas, 2016 p.3) [Fig. 26]. Of the various hypothesis of origin, at the time of authorship, the foremost plausible appears to be that of convergent evolution arising manifold times across diverse taxa in temporally and spatially divergent MTC ecoregions (Ibidem; Pausas and Keeley, 2009). For example, in the Poales species Anarthriaceae-Restionaceae, fire-stimulated germination was already well established by 91 mya (He and Lamont, 2017). Whereas, in the flowering plant family Proteaceae, of which genera include the national flower of Australia, the Banksia, and the genus Embothrium, which is more commonly known as Chilean firebrush, pyrogermination evolved 81 mya (Ibidem). In the study region, California, pyrogermination is prevalent in many annual species, most of which are pyroendemics, as are a significant number of woody species (Keeley and Pausas, 2016). Primarily associated with evader species, the trait is also exhibited in many invaders.

The trait is triggered by a variety of autecological mechanisms, including:

• Heat cracking of seedcasings, which enables the process of imbibition [88] to occur (He and Lamont, 2017).

Response to chemical signals emitted in smoke, charate, and ash upon the breakdown of organic compounds (Ibidem). For example, when living, some plants use cyanide as a defence against herbivorous animals. However, when these same plants burn they emit the nitrogenous compound cyanohydrin (glyceronitrile), which has been found to stimulate germination in pyroendemics (Flematti et al, 2011; Downes et al, 2014). Additionally, a class of organic molecules known as butenolides have been identified as highly active stimulants (Flematti et al, 2004; van Staden et al, 2004), and studies evidence that butenolide derivative Karrikins [KAR1, KAR2, KAR3, KAR4, and, to a lesser extent, KAR5 and KAR6], of which the origin is strictly combustion (He and Lamont, 2017, p.11), trigger the germination of dormant pyroendemic seeds (Keeley and Pausas, 2016; Nelson, D. C., et al, 2012; Daws et al, 2007). However, KARs and glyceronitrile are but one of several compounds found to stimulate germination, others include another nitrogenous compound, nitrous oxide, and the hydrocarbon ethylene (He and Lamont, 2017). Having been transported in smoke and residues, the compounds are absorbed by soils, thereon, the seeds that are stored in those soils, ultimately reaching the embryos, their proteins, and the genes therein, the latter of which are encoded to respond to the compounds (Ibidem). The earliest report documenting the process thereof was published just a decade ago (Keeley and Pausas, 2016), thus the field is one of active enquiry, with many questions left as yet unanswered. However, studies suggest that pyrogermination cued by chemical signals likely incurs combinations of compounds to catalyse the process, but that butenolides [KARs] are central thereto (Ibidem). In some instances, pyrogerminatic species, as we might call them, have been found to tightly couple chemical signalling with flammability, thus ensuring that enough ethylene is emitted to trigger pyrogermination (He and Lamont, 2017).

Abscission

A process of “co-ordinated breakdown of the cell wall matrix at discrete sites and at specific stages during the life cycle of a plant” (Roberts et al, 2000, p. 223), abscission is a common trait in floral species, which facilitates the shedding of parts for reasons including reproduction, conservation of resources, and defence. Appearing in the conifer lineage in the Carboniferous, branch abscission and healing (He and Lamont, 2017, Looy, 2013), and the sparse crown that is a consequence thereof, is a trait associated with resisters.

Retardant Rhytidome

Foremost in the pyro-armoury of resisters, rhytidome is the outer of a tree’s two bark layers. Mainly comprised of dead tissue, but penetrated by periderms [cork layers], rhytidome is hypothesised to have evolved by means of protecting plants from the elements [the climate hypothesis] and/or from pests, infections herbivory species [the biotic hypothesis] (Pausas, 2015). In resisters, rhytidome is thicker, as is the overall bark assemblage, this being a trait that protects the inner, meristematic89 tissues from fire (Pausas and Keeley, 2009). The genus Pinus provides evidence of the evolution of thick bark as a means of protection from fire [the fire hypothesis] [Figs. 27, 28], as lineages that are prominent in fire-prone ecoregions are endowed with thicker bark than are those that are present elsewhere (Pausas, 2015; Keeley and Zedler, 1998).

Resprouting

As described above, resprouting is an endurer hallmark. Abundant in its taxonomical distribution, the trait is present in several ancient woody lineages, including that of ‘living fossil’ Ginkgo biloba (Pausas and Keeley, 2009), a gymnosperm, which has remained largely unchanged in over 200 million years (Cohn, 2013). Whereas pyriscent and pyrogerminating species may endure the passing of one fire, but not another whereupon the interim between the two is short, resprouters will usually persist (Agee, 1998). Hence, why Ginkgo, and its evolutionary descendants are well placed to endure for many millennia to come.

4.2.8 Vitai Lampada Tradunt [90]: Six Fire-Resilient Specimens

“At the festival of Attis, a sacred pine tree was cut down and carried in a procession, its branches hung with violets, signifying nature’s endless cycles of death and renewal”. Ronnberg and Martin, 2010.

A selection of species that exhibit the traits of endurer, evader, and resister are described below:

Endurers

Quaking aspen

A shade intolerant pioneer species [91], Quaking aspen (Populus tremuloides) is the most widely distributed tree in the northern United States (Little, 1971). Enduring fires at frequencies of 3> years, the species flourishes in myriad alluvial variants, including the mineral soils that fires create (Perala, n.d). Within a year, Quaking aspen seedlings have capacity for reproduction by resprouting, this being a trait that is expressed copiously in mature stands (Jones, 1974), and that is stimulated by fire. Prolific seeders, once established, stands usually yield a heavy seed crop every 4-5 years, with lighter crops in the interim (Perala, n.d). While individual trees and ramets [92] ‘live fast, die young’, usually surviving no more than two centuries at most (Hunt, 1986; DeByle and Winokur, 1985), the clone colony from which they sprout may be millennia old (Kemperman and Barnes, 1976).

Canyon live oak

A native of the western United States, the Canyon live oak (Quercus chrysolepis) endures wide-ranging terrains and soil types. But, as its name implies, it is particularly well adapted to sheltered canyons. Found in several MTC ecoregions, including chaparral and mixed conifer forests, the species flowers from 15-20 years of age, in some instances yielding acorns each year, but with good seed crops averaging every 2-4 years (Thornburgh, n.d). Seeds mature and fall within one season, generally dispersing within the immediate vicinity, though sometimes dispersed further by fauna or by falling down a steep slope (Ibidem). While its seeds germinate in a variety of seedbed types, their survival rate is higher when soil is moist, covered by leaf litter, and an overstory is present (Ibidem). However, while a fire can eradicate these conditions, thus, seedlings rarely germinate post-fire (SDSUF, 2004), Canyon live oak is a resprouter adept at repopulating disturbed sites (Talley and Griffin, 1980). A process stimulated by injury, resprouting is more vigorous in healthy parent specimens (Thornburgh, n.d). Other genus members that exhibit this rapid post-fire repopulation trait include the Gambel oak (Quercus gambelli) and the California black oak (Quercus kelloggii).

Evaders

Lodgepole pine

A ubiquitous species of which the name originates from its use as a structural component of Native American tipi shelters, Lodgepole pine (Pinus contorta) has several morphologically and ecologically distinct subspecies, of which the Rocky Mountain-Intermountain Race (P.contorta var. latifolia) is the variant of particular interest in this study. Producing plentiful seeds from the age of 5-10 years, with good crops averaging as often as 1-3 years, the species protects its faunal offspring in both serotinous and non-serotinous cones (Lotan and Critchfield, n.d), the selection therefrom shaped by the regional historical fire regime [HFR]. In the former instance, cones may remain closed for decades, which given annual seed production can run as high as 790,000 seeds per ha (Ibidem), provides P.contorta with a robust insurance policy against extinction, so long as the HFR remains in situ. Favouring germination in bare mineral soils or distributed duff in landscapes devoid of shade and competing flora, this evader species can repopulate disturbed sites in great abundance, hence commonly found in monotypic stands at sites including Yellowstone National Park (Ibidem).

Knobcone Pine

A fast-growing West Coast indigene, Knobcone pine (Pinus attenuata) is the region’s most widely dispersed closed-cone species (Howard, 1992). Residing at the interface of MTC chaparral and woodland, P.attenuata is shade intolerant and short-lived, surviving <100 years (Horton, 1960). A pioneer species that prefers higher elevations, its cones are sealed by resin that requires temperatures of 203°C> to melt, thus pyrogermination is its sole means of reproduction (Ibidem). Remaining attached to the tree for its life-term (Zedler, 1986; Vogl, 1967) the cone begins to disperse its contents within 1-12 hours of a fire’s passing, continuing the process thereof for 4> years (Vogl, 1973). Hence, the Knobcone pine is dependent on stand replacing crown fires for its survival. The species open and multi-trunked form helps propagate its affiliated fire regime (Ibidem), and vice versa: fire both enables the mechanisms of distribution of its seeds and the rapid germination thereof; studies suggest that germination of Knobcone pine seedlings may require the elevated soil pH that fire creates (Ibidem). Their length the longest of the Californian closed-cone pines, morphologically, P.attenuata’s seeds, which form when the parent plant reaches 10- 12 years old, are designed for dispersal (Howard, 1992). Tolerant of nutrient-poor soils, post-fire, this thin-barked evader species commonly forms even-aged stands of which the density widens as the elevation level increases (Ibidem).

Resisters

Coulter pine

A shade intolerant inhabitant of mixed conifer and mixed chaparral MTC ecoregions (Cope, 1993), the Coulter pine (Pinus coulteri) produces the largest cones of any conifer species [Fig. 29]. Typically found on south-facing slopes and ridges at elevations of 300 – 2,100 meters (Kral, 1993), the species thrives in mineral soils and open canopy. A resistor species, mature P. coulteri have pyro-armoury in the form of thick, furrowed bark of sufficient depth to protect against relatively frequent, low to moderate intensity surface fires (Cope, 1993). Fast-growing and living to <100 years (Ibidem), the species produces both serotinous and non-serotinous cones from the age of 10-15 years onwards, bearing good seed crops every 3-6 years (Krugman and Jenkinson, 1974). Pyriscence is markedly more frequent in P. coulteri cohabiting with chaparral, Canyon live oak, and Sargent Cypress [Cupressus sargentii) communities (Cope, 1993). Whereupon the trait is present, copious seedlings may germinate, such that, in the immediate years after a fire’s passing, stands tend be even-aged and dense, that latter in part due to seedlings’ dispersal being limited due to the great size of the cones (Ibidem). Listed near-threatened (IUCN, 2013), the species distribution has been in decline for some decades, with changes to regional fire regimes posited as the primary causation (Ibidem).

Ponderosa pine

An autochthonous species of western North America, Ponderosa pine (Pinus ponderosa) has evolved into five morphologically distinct subspecies (Callaham, 2013), the matter thereof reflecting the wide expanse and relative diversity of its territory. Inhabiting elevations from sea level [California, 150 – 1070m] to <3050m [southern Rockies], P.ponderosa forests shift to progressively higher altitudes along a north to south axis (Oliver and Ryker, n.d). Shade intolerant, this resistor species commonly lives to between 300 – 600 years old (Ibidem). Studies of the basal scars in old growth forests evidence that prior to European colonisation P.ponderosa stands commonly burned at intervals of 5-20 years (James, 1979). Endowed with some of the most formidable pyro-armoury of any fire-adapted flora species, mature P.ponderosa usually withstand multiple surface fires thanks to defences including thick bark, which forming from 3>5 years, becomes fissured and furrowed, morphing into a plated pattern over time (Callaham, 2013), the configuration thereof dissipating heat, hence helping to protect the cambium [93] (Jensen and McPherson, 2008). Ponderosa pine trees have been known to outlive fires of such severity as to scorch <50% of their crown (Oliver and Ryker, n.d). A contributing factor to their survival is the comparatively open canopy structure that the species’ stands form, this trait further enabling the dissipation of heat (Jensen and McPherson, 2008). Often growing in even-aged stands, its seeds are relatively large for members of the genus Pinus, thus tend not be carried far by the wind (Jensen and McPherson, 2008). P. Ponderosa can populate sites of which the general conditions, including soil quality and depth, is too poor for most plant biota, and particularly other perennials, to become established. Thus, the species frequently cohabits with graminoids, forbs, and shrubs (National Park Service, n.d). One of the functional traits that both helps this species to withstand fires (Jensen and McPherson, 2008), and to persist in relatively inhospitable terrain is its seedlings’ capacity to sink roots to depths to 50cm in moist porous soil (Larson, 1963), thereon to 2m in maturity (Oliver and Ryker, n.d). Monoecious [94], Ponderosa pine produces cones from 7 years and upwards to 350 years or more, with stands aged 60 – 160 years bearing the most viable seeds (Ibidem). However, both in volume and frequency, seed crops vary across the subspecies (Ibidem): the seed distributions seemingly coupled to the historical fire regime within which the P.ponderosa variants reside. As with elevation, a north to south alignment [95] in the trait’s geographical arrangement is evident. These are but a few of the factors that make Ponderosa pine one of, if not the most, die-hardy of fire-adapted species.

“the pinecone often crowned the mythic Tree of Life, abode of the Great Mother, fecund source and vessel of nurturance, healing and transformation”. Ronnberg and Martin, 2010.

The distribution of these and other endurer, evader, and resister species reveals a landscape’s historical fire regime. Whereupon combinations of the three fire-adapted types are found at one site it speaks to the mixed-severity regime, this being commonplace due to the highly diversified topographic, alluvial, and other environmental characteristics of many ecoregions of the western United States.

4.2.9 Parturition of a Pyroparadigm

“Even if we fully understood all these details, we still would not have the ability to stop large fires, just as we are unable to stop earthquakes, volcanoes, hurricanes, tornadoes, and other natural cataclysms. Nor should we want to. The environmental consequences of disrupting these large-scale processes would be catastrophic”. Jensen and McPherson, 2008.

The finer points of the relationships between the variables that dictate spatiotemporal patterns and processes within fire regimes are becoming ever more illuminated by the day, and so much so that fire ecology is described as being at the “forefront” of scientific discovery (Moritz et al, 2005, p.17912). For example, a study just released by the National Oceanic and Atmospheric Administration (NOAA, 2017) established that, as paleoecological records suggested could be the case, plant growth is accelerating as CO2 levels rise in the atmosphere. The team, led by Elliot Campbell of the University of California, studied a data-set spanning 54,000 years, concluding that the higher the level of atmospheric CO2, the greater the ratio of CO2 plants fix in their tissues. The research builds on earlier studies that indicated a correlation between elevated atmospheric CO2 and increased biomass production (Ainsworth and Long, 2005), and tree branching (Saxe et al, 1998). Systemically, the implications are significant, and especially in the context of wildfires. CO2, which at the time of writing is at over 407 parts per million (ppm) in the atmosphere (Scripps, 2017), and fast rising, is creating a growth spurt in some forest regions (Pan et al, 2013), which in the language of thermodynamics translates to simultaneously increasing fuel loads, and the combustibility thereof, given the higher the CO2 fixation, the higher the quantity of O2 being released. While both fuel load and the combustibility thereof is increasing, so too, in some regions, is the probability of an ignition source (i.e. lightening), given that the greater the increase in average global temperature, the greater the energy circulating in the atmosphere, therein, the greater the possible number and size of some types of weather system, including dry thunder storms.

Today, roughly 100 lightning flashes per second are recorded, which translates to approx. 1.4 billion lightning flashes annually (NOAA, n.d) [Fig. 30]. The sum thereof constitutes a mix of lightning types, of which some are more likely than others to start a wildfire. Whereupon lightning occurs in the presence of precipitation its hue will be red if rain, blue if hail, both of which are indicative of a reduced probability of a cloud-to-ground strike igniting a fire. However, white lightning signals an absence of atmospheric moisture, and is thus more likely to start a fire whereupon wider environmental conditions (i.e. biomass state) are conducive thereto (Schneider and Breedlove, n.d). By means of giving some measure of the potency of nature’s foremost fire starter, air’s resistance to the movement of the electrical charge known as lightning can generate temperatures of 27,760°C, that being five times hotter than the Sun’s surface (NOAA, n.d). Unevenly distributed, both spatially and temporally, lightning flashes increase in the summer months when thunderstorms are more frequent, and are several times more likely in the tropics, where 70% of lightning activity occurs (Ibidem). The fact that lightning activity coincides with the period when regional fuel loads are generally tinder-dry, means that even an incremental increase in the number of summer storms, therein lightning strikes, in the western United States may cause more, not less wildfires in the future.

As will be discussed later in this thesis, a significant body of empirical and theoretical data makes clear that while we know not the specifics of the future climate, suffice is the evidence as to assert that Earth Systems interact in such a way as creates planetary-scale chain reactions, not all of which we yet fully understand. However, the matter thereof makes the research field all the more compelling, for as information and insights fast-emerge, that in turn develop our understanding of the way in which fast and slow variables within the planetary-scale schemata interplay [Fig. 31], so too emerges a new paradigm, and not in a superficial sense, but as fulfils the criteria [96] as was specified in the seminal publication that first popularized the term (Khun, 1962).

>Continue to Chapter 4 [part III] here.

Footnotes

[56] Broadly corresponding with a climatic region, “biome” describes the largest biogeographical region (Allaby, 2012).

[57] Covering 22.9 million square kilometres, the Neoarctic is a biogeographic region extending from North America to Central Mexico (Escalante et al, 2010) that was first delimited in the 19th Century by Sclater (1858), and Wallace (1876). The latter delineated the region into four subdivisions, of which he stated, though “pretty clearly indicated by physical features and peculiarities of climate and vegetation”, zoologically, “while the species of several sub-regions are in most cases different”, at the taxonomic level of genus, “even the vast range of the Rocky Mountains has not been an effectual barrier against this wide dispersal of the same forms of life”. Following in Wallace’s footsteps, the WWF also divides the Neoarctic into four bioregions having applied a similar methodology (Ricketts et al, 1999).

[58] ‘Biogeographic realm’ is the broadest delineation of taxonomic composition. First introduced by Miklos Udvardy in a paper for UNESCO’s Man and the Biosphere Programme (Udvardy, 1975), the classification system was adopted by WWF in its Global 200 scheme in the 1990s, employing the same general methodological approach (Olson and Dinerstein, 1998). Udvardy’s system defined 8 categories of biogeographic province, with a further 193 subcategories characterized by one of 14 biome types. Whereas, WWF’s system is more detailed, specifying 8 forest types and 867 ecoregions (Olson et al, 2001).

[59] An ‘ecoregion’ designates a “large unit of land or water containing a geographically distinct assemblage of species, natural communities, and environmental conditions” (WWF, n.d).

[60] Some several hundred definitions for ‘forest’ exist worldwide (United Nations, 2010). However, within the research domain of dendrology a ‘forest’ is defined as an area greater than 0.5ha of land featuring trees of 5m or more in height, of which the density is at least suffice to meet a minimum of 10% canopy cover (FOA, 2005).

[61] The term necromass describes “any quantitative estimate” of the sum of the mass of dead organisms p/unit area or volume within a specified time (Lincoln, Boxhall, and Clark, 1998).

[62] Quantified on a scale of 0-100, albedo describes the percentage of the Sun’s energy (radiation) reflected back into space from the Earth’s surface. On average, Earth’s surfaces have an albedo of .31. However, forests have a relatively low albedo, averaging between .08 and .15 (ESSEA, 2017).

[63] In 1584 Elizabeth I chartered Sir Walter Raleigh to explore and colonise “remote, heathen and barbarous lands, countries, and territories, not actually possessed of any Christian Prince or inhabited by Christian People” (Lillian Goldman Law Library, 2008). However, attempts at the colonisation of North America were unsuccessful until 1606, when King James I chartered the London Company to establish colonial settlements, of which Jamestown, Virginia, established in 1607, was the first.

[64] Graminoids refers to plants of the order Poales (Poaceae), more commonly known as the grass family (Allaby, 2012).

[65] Functional traits are described as the “morphological, biochemical, physiological, structural, phenological or behavioural characteristics of organisms that influence performance or fitness” (Nock, Vogt, and Beisner, 2016).

[66] Phylogenetics pertains to the evolutionary history of a taxonomic group of any rank [Domain, Kingdom, Phylum, Class, Order, Family, Genus, or Species] (Lincoln, Boxshall, and Clark, 1998).

[67] The term basal area (BA) describes ‘the cross-sectional area of a tree stem when measured at breast-height’, and it is calculated using the formula BA (m 2) = pi * DBH (cm) 2 / 40000, (Erdle, 2012).

[68] Shrubland is a biome type comprised evergreen sclerophyll shrubs, which comprised hard leaves, are woody plants that grow to less than 10m tall.

[69] Snag refers to standing dead tree, or part thereof (NOAA, 2016).

[70] Whereupon its tissues are damaged by fire, a tree develops physical and chemical boundaries at the injury site: a process that helps to reduce the probability of infection. As in humans, this process is called ‘scarring’, and in the record thereof helps dendrologists to establish information about the timing and possibly intensity of an historical fire (USDA, 2016).

[71] Topkill refers to mortality of aerial biomass, which may, or may not recover by resprouting in the aftermath of a fire).

[72] Course woody debris is described as “dead woody materials in various stages of decomposition, including sound and rotting logs, snags, and large branches” (Enrong, Xihua, and Jianjun, 2006).

[73] Stand refers to a biotic unit comprised a single species that is homogeneous both in composition and age (Lincoln, Boxshall, and Clark, 1998).

[74 /75] The finer the fuel, the higher the surface area to volume ratio, i.e. whereas a blade of grass might have a ratio of 1:3,000, a log might have a ratio of 1:6 (TDA, 2002).

[76] Pyrolysis refers to an irreversible thermochemical decomposition that occurs in carbon-based [organic] matter whereupon exposed to high temperatures in the absence or near absence of oxygen. The process turns organic matter into its gaseous components, leaving a solid residue in the form of carbon, ash, and pyrolytic oil (Boslaugh, 2017).

[77] Systema Naturæ per ignem regnis [Nature System by means of the three kingdoms of fire] is a reference to the title of the 10th edition of Caroli Linnæi’s publication of Systema Naturæ, which published in 1758, had the extended title Systema Naturæ per regna tria naturæ, Secundum Classes, Ordines, Genera, Species, Cum Characteribus, differentiis, synonymis, locis.

[78] Ecologist Richard Vogl thought the concept of successional stages of limited in fire ecology. He considered it more useful to conceive of species through the lens of a 5-part functional group comprised increasers, decreasers, neutrals, invaders, or retreaters (Vogl, 1974).

[79] Serotiny refers to a functional trait, which found in some flora species, enables the retention of seeds in resistant structures, including cones and pods, whereupon dispersal follows a major environmental disturbance (Lincoln, Boxshall and Clark, 2003).

[80] Perennial refers to plants of which the lifespan exceeds 24 months. 81 Ephemerals may be described as flora that lives fleetingly.

[82] Occurring at temperatures of 500 °C and above, flashover refers to the near-simultaneous ignition of organic matter within an enclosed area.

[83] Mesophytes are plants adapted to environments where precipitation is not scarce or excessive.

[84] Autecology is the study of how individual organisms or single species interact with both living and nonliving entities within their environment.

[85] On the Tendency of Species to form Resilience is a reference to a joint presentation made by Alfred Wallace and Charles Darwin at the Linnaen Society of London on July 1st 1858 to announce their theory of evolution by natural selection.

[86] A method of molecular systematics, which resides within field of phylogeny, Molecular phylogenetics involves the analysis of hereditary molecular differences by means of establishing the evolutionary relationships between species (Brown, 2002).

[87] The Bayesian method enables calculation of the probability that an unknown organism belongs to a specified taxonomic lineage (Lincoln, Boxshall and Clark, 2003).

[88] Imbibition is a diffusion process, which in seeds involves the absorption of water.

[89] Meristematic tissue is comprised “unprogrammed” living cells that are yet to be “assigned a role” within a plant, otherwise described as “undifferentiated cells” (Steadham, 2017, online).

[90] Latin motto ‘Vitai Lampada Tradunt’ translates to ‘They hand on the torch of life’. 91 Pioneer species are usually the first species to populate a disturbed site.

[92] Ramet refers to an individual member within a clonal community.

[93] Cambium refers to the layer of cells at the interface of the xylem and phloem, which in woody plants continue to divide throughout the plant’s life, thus creating secondary xylem and phloem, of which the consequence is a thickening of bark.

[94] Monoecious refers to a species that bears both male and female reproductive organ.

[95] P. ponderosa subspecies west of the Sierra Nevada, CA, average medium seed crops every 2-3 years and heavy crops every 8 years. Whereas, subspecies in Montana have been recorded to have just one good seed crop in 23 years (Oliver and Ryker, n.d).

[96] Whereupon, having re-evaluated the premise of a scientific theory, the research community concludes there potential for a more viable alternative, thereon displaces prior fundamental assumptions, and its “imagination” is stimulated suffice to transform a discipline, or collective thereof, a new paradigm will emerge. In ‘The Structure of Scientific Revolutions’ Thomas Khun described the processes thereof through a series of historical precedents, including the events that led to Lavoisier’s oxygen theory of combustion.

The thesis is also available in PDF format, downloadable in several parts on Academia and Researchgate.

Note that figures have been removed from the digital version hosted on this site, but are included in the PDFs available at the links above.

Citation: Sterry, M. L., (2018) Panarchistic Architecture: Building Wildland-Urban Interface Resilience to Wildfire through Design Thinking, Practice and Building Codes Modelled on Ecological Systems Theory. PhD Thesis, Advanced Virtual and Technological Architecture Research [AVATAR] group, University of Greenwich, London.