Darker, cooler, wetter: forest understories influence surface fuel moisture

Highlights

• Understorey vegetation alters near-surface microclimate and surface fuel moisture

• Dense understorey forest was cooler and more humid at the near-surface

• Screen height microclimate measurements could underpredict of fuel moisture

• The effects of the understorey should be considered when estimating fuel moisture

Abstract

The moisture content of dead leaves, twigs and bark on the forest floor is a key determinant of fire behaviour. The microclimate inside forests, which drive the moisture content of these dead fuel components, is typically measured at screen height (150 cm). However, in some forest types, the surface fuel at ground level may be subject to additional sheltering from low shrubs, ferns and grasses, which could alter the microclimate near the surface (hereafter near-surface). In such cases, screen height measurements may not adequately represent the near-surface conditions that determine dead fuel moisture contents.

We sought to quantify the effect of understorey vegetation on near-surface microclimate. We measured in-forest temperature, relative humidity and solar radiation in eucalypt forests over two fire seasons at both screen height and the near-surface using weather stations at 25 sites. The sites encompassed wet eucalypt forest (n=18) with a dense, mesic understorey and dry eucalypt forest (n=7) with a sparser, scleromorphic understorey.

Wet forests with dense understorey vegetation had near-surface air temperatures that averaged 1.3°C lower, relative humidities that averaged 13.1% higher and total solar radiation that was 0.84 MJ less per day compared with those measured at screen height. These microclimate differences led to predicted fuel moistures which averaged 4.7% higher at the near-surface compared with screen height – this was statistically significant. In contrast, dry forests with less understorey vegetation, had near-surface air temperatures that averaged 4.2°C higher, and relative humidities that averaged 3.1% lower compared to screen height. These differences were not large enough to translate into statistically significant differences in predicted fine fuel moisture between heights.

Overall, these findings show that understorey vegetation plays an important role in moderating near-surface microclimate in some forest types and this needs to be taken into consideration when predicting fuel moisture.

Introduction

Forest wildfires typically ignite in the surface fuel layer of leaves, twigs and bark and this fuel strata is important for sustaining fire spread (Gould, Lachlan McCaw et al. 2011). Dead Fine Fuel Moisture Content (FFMC – referring to fuel less than 6 mm in diameter) is a key determinant of ignitability (Fernandes, Botelho et al. 2008, Cawson and Duff 2019), fire spread (Rothermel 1972, Burrows 1999) and fuel consumption (Knapp, Keeley et al. 2005, de Groot, Pritchard et al. 2009). In general, if dead FFMC is high (e.g. exceeding 16%), fires are unlikely to ignite and burn (Tolhurst and Cheney 1999). Conversely, for very low dead FFMC (5-7%), fire behaviour can be erratic with spot fires igniting rapidly and crown fires likely (McArthur 1967, Sullivan and Matthews 2013). As such, reliable information about dead FFMC is critical for effective fire management decision making.

Dead FFMC is constantly changing in response to climatic conditions. Seasonal and daily fluctuations in temperature, relative humidity, solar radiation, precipitation and wind, determine evaporative demand and the moisture vapour differential between the fuel and the atmosphere (Viney 1991, Matthews 2014). The microclimatic conditions of the air immediately surrounding the fuel particle are most critical to the moisture content of dead fine fuel as this is the interface at which the exchange of moisture between the fuel and atmosphere occurs (Viney 1991). Yet, rarely do measurements of atmospheric conditions occur close to the fuel surface.

The within-forest climate (hereafter referred to as microclimate) can differ substantially from the regional climate due to vegetation and small scale topographic effects (Chen, Saunders et al. 1999, Zou, Barron-Gafford et al. 2007, Walsh, Nyman et al. 2017). The forest canopy plays a key role by intercepting and buffering the external forest climate (hereafter referred to as macroclimate). For instance, solar radiation is absorbed or scattered by the canopy, which reduces the penetration of light energy into the forest (Martens, Breshears et al. 2000, Zou, Barron-Gafford et al. 2007). This shading results in progressively cooler daytime air temperatures and less direct heating of biomass from the canopy towards the ground (Madigosky and Vatnick 2000, Hardwick, Toumi et al. 2015). Wind is also intercepted and its speeds are progressively weakened by each vegetation layer encountered (Grant and Nickling 1998, Moon, Duff et al. 2013). Turbulent wind allows for mixing of warmer air downwards; in the absence of this, within-forest daytime temperatures tend to be cooler near to the ground than at or above the canopy (Chen, Franklin et al. 1993, Hardwick, Toumi et al. 2015). As relative humidity inversely follows air temperatures, daytime within-forest relative humidity also tends to stratify vertically – being lowest in the canopy and increasing downwards (Madigosky and Vatnick 2000, Ray, Nepstad et al. 2005). As canopy density increases, it creates a subcanopy microclimate that is increasingly distinguished from the macroclimate (Ray, Nepstad et al. 2005, Ma, Concilio et al. 2010). As such, there are strong relationships between canopy density and dead fine fuel moisture in forests (Tanskanen, Venalainen et al. 2005, Cawson, Duff et al. 2017).

Forest microclimate and fuel moisture research typically involves in-forest weather observations made at screen height (150 cm from ground) (Wotton, Stocks et al. 2005, Burton, Cawson et al. 2019). This is because screen height measurements conform with open-air meteorological station standards (Canterford 1997). However, measurements at these heights are unable to capture the effects of understorey vegetation, which is often present between the litter bed and screen height. This undergrowth can vary substantially in composition, structure and density between forest types (as reviewed by Barbier, Gosselin et al. 2008, Hong, Qi et al. 2014). It is likely that understorey vegetation adds an extra layer of vegetation interception (Geiger, Aron et al. 1995), creating a close to ground (near-surface) microclimate that varies from that measured at screen height, though this has not been widely studied.

Most research concerning near-surface forest microclimate focuses on the effect of changes to the tree canopy due to human disturbances like timber harvesting (Potter, Teclaw et al. 2001, Devine and Harrington 2007) or from natural forest processes like single tree-fall gap and the succession of one forest type to another (Porte, Huard et al. 2004, Fletcher, Pickett et al. 2007). There is less focus on how understorey vegetation may further moderate the near-surface microclimate. Of the few studies that do consider understorey vegetation, it has been shown that when coupled with canopy cover, there is a reduced amplitude of fluctuations in near-surface microclimate (as compared to the solitary effects of the canopy) but that this microclimate also varies as a function of understorey vegetation density (Fetcher, Oberbauer et al. 1985, Brooks and Kyker-Snowman 2008). Similar studies are lacking for Australian eucalypt forests, yet understanding the drivers of near-surface microclimate and fuel moisture in these forests is critical for predicting fire behaviour.

Eucalypt forests are a key fuel type in south-eastern Australia, contributing to large and intense wildfires (Cruz, Sullivan et al. 2012, Murphy, Bradstock et al. 2013). There are a range of different eucalypt forests that vary in their structural characteristics, species composition and flammability (Florence 1996). One important division is between wet and dry eucalypt forests, which differ markedly on several measures including understorey structure and composition (Department of Environment Land Water and Planning 2016). Wet forests typically have a dense array of broad-leaved mesic species in the understorey, while dry forests have a sparser mix of scleromorphic species. Accurate predictions of dead fuel moisture across contrasting forest types are needed to inform fire management decisions. For instance, fuel moisture thresholds are used as part of the decision-making process for determining when prescribed burning can be carried out safely and effectively (Slijepcevic, Anderson et al. 2015). Currently, existing moisture models tend to perform poorly when predicting surface moisture in wet eucalypt forest (Slijepcevic, Anderson et al. 2013, Slijepcevic, Anderson et al. 2015). Understanding the impact that understorey vegetation has on the near-surface microclimate may help explain why this is the case. Importantly, it has the potential to deepen our understanding of the flammability of these differing forests.

Our study sought to quantify differences in microclimate between screen height and near to the forest floor (near-surface) in contrasting wet and dry eucalypt forest types, and in doing so consider the potential effect of understorey vegetation on fuel moisture. Specifically, we asked:

• To what extent does understorey vegetation influence near-surface microclimatic conditions?

• Is understorey vegetation likely to have a substantial influence on surface dead fine fuel moisture contents?