1 Introduction
1.1 Multiaged management
Ecological forestry which maintains a wide range of ecosystem services while also supplying timber requires a diverse landscape of highly varied forest structures (Aplet, 1994; Nolet et al., 2018; O’Hara, 2001) . These, in turn, require a variety of silvicultural techniques to implement and sustain (O’Hara, 1998; Schütz, 2002). The development of multiaged stand structures has long been of interest to silviculturists as a key alternative to the conceptually and logistically simpler, even-aged management (Schütz, 1999). Multiaged silviculture refers to the retention of trees of distinctively different age classes, growing together within the same stand. These cohorts may co-occur at the tree level, or in small, even-aged patches within the stand. In the latter case, the distinction between even- and multiaged management can become blurred with increasing patch size, but patches are generally much smaller (often less than 1 ha) than the stands they compose. The pursuit of multiaged stand structures has often met with mixed results (O’Hara, 2002) and this has led to the investigation of several different systems for achieving such structures. Research into the efficacy and results associated with these is ongoing (Beese et al., 2019; Nolet et al., 2018). One such system that has gained popularity in recent decades is known as the retention system, which allows for the retention of a range of tree densities, in dispersed or aggregated spatial patterns and can be used to maintain, multiaged stands, or convert from even-aged management (S. J. Mitchell & Beese, 2002).
Redwood forests offer a prime opportunity for multiaged management because coast redwood (Sequoia sempervirens) regenerate reliably via stump sprouting and are relatively shade tolerant. The very high leaf areas observed in these forests, suggest their suitability for a multi-layered forest structure (Berrill & O’Hara, 2007; Van Pelt et al., 2016). Additionally, with their high timber value and productivity, Redwood forests are of keen interest to private timber producers. Despite redwood’s fitness for multiaged stand structures, the successful development of subordinate cohorts depends on adequate access to light, and light deficiency can lead to reduced vigor and mortality in young sprouts and understory trees (Barrett, 1988; Muma et al., 2022; O’Hara et al., 2007; Webb et al., 2012).
Complicating redwood regeneration and sprout development is the fact that the competing hardwood species, tanoak (Notholithocarpus densiflorus), is also shade tolerant and a vigorous resprouter. Tanoak is a keystone species in terms of wildlife habitat and First Nation’s cultural identities, but from a timber production standpoint it is often perceived as a nuisance due to a lack of market development combined with its widespread proliferation following intensive, repeated conifer harvesting (Bowcutt, 2011). While redwood grows more quickly than tanoak in multiaged stands, competition from hardwoods such as tanoak reduce conifer growth and drought resistance (Berrill et al., 2018; Dagley et al., 2023). There has been little scientific investigation into the development of tanaok under shade (Waring & O’Hara, 2008; Wilkinson et al., 1997).
1.2 Management effects on sprouting
The most commonly used metrics for quantifying sprouting response are percent of stumps sprouting following cutting, the sprout density or the number of sprouts arising from a cut stump or within a sprout clump, and sprout development which can include height and/or diameter.
An important consideration in comparing sprout response across studies is the time between treatment and measurement. Sprout growth in most species is initiated by the mobilization of carbohydrates stored in the underground portions of the tree (Del Tredici, 2001), and differences resulting from external conditions may not be realized early in development. Redwood sprout clumps can consist of 100 or more stems the first year after cutting (Neal, 1967), but rapidly self-thin in full light (Boe, 1975). With overstory competition, this loss may proceed even more rapidly, possibly resulting in the mortality of the entire clump (O’Hara & Berrill, 2010). Whereas the thinning of sprout clumps, whether from internal, or external competition may last 20-30 years in eastern hardwoods (Gould et al., 2007), this process may occur over hundreds of years in long-lived redwoods (O’Hara et al., 2017).
Because the metrics of percent sprouting, sprout density, and sprout development capture different characteristics of sprout response, they often vary with factors such as species, site characteristics, overstory density, parent stump age/diameter, and geographic province. Even when these variables are accounted for, unexplained variation may remain between sites (Keyser & Loftis, 2015; Nieves et al., 2022).
1.2.1 Composition
Due to their rapid initial growth, sprouting species may alter the composition of a regenerating stand (Del Tredici, 2001). This can lead to an increase in less desirable species (Keyser & Zarnoch, 2014). In redwood ecosystems, redwood sprouts typically outsize the stump sprouts of tanoak, a common associate, in the first 5 years following partial harvest (Muma et al., 2022). It has yet to be seen how these dynamics might change over time, or what their cumulative effect will be on the regeneration of other species (Berrill et al., 2018). Various interactions between treatments and other disturbance factors could lead to differences in regeneration, such as in the case of deer browsing following the use of fire (Wilkinson et al., 1997), heavier deer browsing closer to watercourses (Schneider et al., 2023), or bears preferentially damaging regenerating conifers (especially redwoods) exhibiting rapid diameter growth (Berrill et al., 2017; Dagley et al., 2018).
1.2.2 Sprout growth
One of the clearest relationships among sprouting species is the positive one between sprout growth and understory light (Berrill et al., 2018; Gardiner & Helmig, 1997; Keyser & Zarnoch, 2014; Knapp et al., 2017). Like most sprouting species, despite redwoods shade tolerance it requires a certain threshold of light to maintain growth (O’Hara & Berrill, 2010). The effect of understory light is weakest very early in development when growth is dominated by stored carbohydrates in the parent stem and root system (Gardiner & Helmig, 1997; Keyser & Loftis, 2015).
Sprout growth is also dependent on stump diameter, with larger stumps producing more rapid growth. This has been observed in redwood and tanoak (Berrill et al., 2018; Harrington et al., 1992) and is common among eastern hardwoods as well Keyser & Loftis (2015), but varies among species (Knapp et al., 2017).
1.2.3 Percent sprouting and number of sprouts
It is common among many sprouting species for percent of stumps sprouting after cutting to decline with increasing tree size or age, but this effect is known to vary by species and may be related to site factors as well. (Johnson, 1977; Nieves et al., 2022). In redwoods, some authors have found evidence of this trend (Neal, 1967; Wiant & Powers, 1967), while others have not (Barrette, 1966; Lindquist, 1979). This may be due to the very wide range of tree sizes and ages possible with redwoods. It has been suggested that percent of stumps sprouting may initially increase with age up to a certain point, and then decrease with trees older than around 200 to 400 years (O’Hara et al., 2007; Powers & Wiant, 1970). Decreasing percent sprouting has been demonstrated for tanoak, among other coastal hardwoods (Harrington et al., 1992).
Residual overstory density may affect the percent sprouting for some species and locales, but detection of this effect has varied across studies and is sensitive to the range of residual basal areas observed in a study (Nieves et al., 2022). Redwood studies have found this phenomenon weak or absent (Barrett, 1988; Lindquist, 1979). The number of sprouts produced by a cut stump for eastern hardwoods is usually not correlated with overstory density (Atwood et al., 2009; Knapp et al., 2017), and this is assumed to be the case in redwood forests as well (Lindquist, 1979; O’Hara & Berrill, 2010).
After cutting, 90-100% of second-growth redwoods (trees smaller than 90 cm dbh) can be expected to sprout (Barrette, 1966; Lindquist, 1979). However, when larger older redwoods are cut, their stumps are less likely to resprout; percent sprouting among larger older trees approaches 50% (Boe, 1975; Neal, 1967). Among cut stumps that do sprout, survival of all the sprouts on a stump is not guaranteed. Entire sprout clumps can die quickly in low light environments (O’Hara & Berrill, 2010). The survival of these new sprouts in each subsequent year is a function of overstory density, especially when approaching closure of the overstory. Percent sprouting has also been found to vary by site and regional factors (Keyser & Loftis, 2015; Nieves et al., 2022). These have not been explored for redwoods, but they represent a possible set of confounding factors in the detection of sprouting trends.
1.3 Forest fuels
Throughout many of the fire-adapted forests of California, fire exclusion combined with timber harvest has led to dense, younger stand—often comprised of suppressed trees—proliferation of more fire-sensitive species, and an accumulation of surface fuels (Safford & Stevens, 2017; Stephens et al., 2009). This situation combined with climate change has led to increased size and frequency of high-severity fires in many regions (Parks & Abatzoglou, 2020; Westerling, 2016), prompting widespread interest in fuel treatments and resilient stand structures.
This interest has seen less momentum in the redwood region, likely due to the perceived safety of these typically moist forests from the threat of large wildfires. Yet redwood litter is among the most flammable of conifer litter types (Fonda et al., 1998), seasonal drought leads to cured fuels, especially during extended breaks in coastal fog (Jacobs et al., 1985), and numerous physiological adaptations suggest that redwood has evolved under fire disturbance pressure (Varner & Jules, 2017). More concretely, there have been at least six large fires in redwood ecosystems since 2003, burning at least 189,000 ha including widespread areas of canopy loss. Scientific consensus places the pre-colonization fire return interval for redwood forests at 6-25 years across their range (Lorimer et al., 2009). It is assumed that much of this activity is attributable to indigenous burning (Varner & Jules, 2017).
TODO: summarize fuel loading in various classes found by these studies
There have been several studies that have quantified various fuel strata in redwood forests. Kittredge (1940) did so for duff and litter in a redwood plantation. Greenlee (1983) studied fuels at Big Basin State Park. Stuart (1985) reported on fuels at Humboldt Redwoods State Park. Finney and Martin (1993a, 1993b) reported on fuels in second-growth redwood forests (aged ~100 years) at Annadel and Humboldt Redwoods State Parks. Graham (2009) reported on fuels in old-growth stands across redwood’s range. Glebocki (2015) studied fuels with and without thinning treatments in young (< 50 years) redwood-Douglas fir stands. No fuel studies, to my knowledge, have been conducted in redwood forests actively managed with multiaged silviculture, but fuel dynamics represent a potentially important decision variable to consider when managing forest stands that may be subjected to intentional or unintentional fire.
1.3.1 Management effects on fuels
Depending on the method used, thinning and harvest treatments may increase, or not affect surface fuel loading. Whole tree removal results in the least fuel accumulation but is more expensive than other options (Han & Han, 2020). Most other treatment methods increase surface fuels (Agee & Skinner, 2005; Stephens et al., 2009). The magnitude of this increase is variable, reflecting factors such as treatment mode, intensity, and pre-existing conditions (Schwilk et al., 2009). Additional research is needed to clarify the effects of these factors on short (Hood et al., 2020; Schwilk et al., 2009; Stephens et al., 2009), and long-term changes to surface fuel load resulting from specific management actions (Hood et al., 2020; Stephens et al., 2012).
The majority of fine dead fuels (< 8 cm) generated by treatment activities typically decompose within 10 years (Burton et al., 2022; Hood et al., 2020; Martinson & Omi, 2013; O’Hara et al., 2017; Stephens et al., 2012). But live woody fuels, which respond vigorously to increased growing space, often persist or increase over time (Keyes & Varner, 2006). The nature of this response depends on eco-type and the amount of growing space created by the treatment which can become dominated by herbaceous plants (Vilà-Vilardell et al., 2023), shrubs (Odland et al., 2021), or small trees (Hood et al., 2020).
Overtime duff and litter loads are frequently lower in more open stands than stands with a more closed canopy. This may result from increased decomposition rates due to greater insolation and increased throughfall, or reduced deposition rates resulting from fewer canopy fuels (Hood et al., 2020; Keane, 2008).
Most fuel reduction thinning research focuses on ponderosa pine (Pinus ponderosa) forests in the United States, with additional studies from other Mediterranean and semi-arid regions (Burton et al., 2022; Schwilk et al., 2009; Vilà-Vilardell et al., 2023). Far fewer studies have been conducted in coastal forests.
1.4 Pyrosilviculture
Forest management for timber and other objectives and prescribed and wild fire are inherently interlinked. This requires research which considers these historically disjunct realms of research in a wholistic way. Fire informed and fire dependent silviculture has been a hallmark of traditional forest stewardship practices for indigenous across northern California since time immemorial (Anderson, 2013). Fire informed silviculture has also been practiced and researched in the Southeastern U.S. for around 100 years (R. J. Mitchell et al., 2009). The American west has been slower to embrace this paradigm shift in thinking now termed “pyrosilviculture” (North et al., 2021). It is my hope that this thesis serve as an example of bridging the gap between the art and science of growing trees to support multiple uses and the thinning and burning practices typically regarded as “fuels management” activities (York et al., 2021). The ability to envision these realms of understanding as integral and essential pieces of a common forest stewardship will lead to new insights and increase our capacity for better land management.