Eco-hydrological controls on microclimate and surface fuel evaporation in complex terrain

https://doi.org/10.1016/j.agrformet.2017.12.255Get rights and content

Highlights

  • Microclimate and evaporation from surface fuels was measured across complex terrain.

  • Variable evaporation was attributed to aridity-related heterogeneity in vegetation.

  • Shortwave radiation is the dominant control on evaporation in water limited systems.

  • Relative contribution of longwave radiation increases as water limitation decreases.

Abstract

The micrometeorological factors driving variation in litter moisture (GWClit) at the landscape scale are poorly understood, particularly in areas with heterogeneous vegetation and complex terrain. In this research we seek to quantify how climate and eco-hydrology contribute to variation in litter moisture and potential evaporation at the forest floor. Research sites were established at 12 locations in southeast Australia with variable precipitation, solar exposure, and drainage position. We measured solar radiation, air temperature, relative humidity, litter moisture, and litter temperature. Spatial patterns of GWClit, examined during two drying phases, show that moisture in the litter and its rate of evaporation are closely linked to vegetation density, which is largely a function of aridity. This creates a pattern whereby aridity-related differences in vegetation structure controls the spatial variation in GWClit, with regional effects being driven by precipitation while local effects are caused by variation in solar exposure and drainage position. By parametrising a model of daily potential evaporation (Ep) at the forest floor we explore how vegetation and topography influence evaporative demand above the litter layer. The model shows that Ep is driven primarily by net radiation and that the role of vapour pressure deficit is almost negligible due to high moisture content within the sub-canopy air mass and high aerodynamic resistance. In dry forests the net radiation is directly related to shortwave radiation and Ep remains high despite low temperatures. In the tall wet forests, commonly found at high elevations, along drainage lines and on slopes with polar-facing aspects, the long-wave radiation (i.e. temperature) was just as important for Ep as the shortwave radiation. Low energy inputs to the forest floor in these tall forests means that a significant rainfall event results in surface fuels that remain wet for much longer than fuels in the dry open forest. This leads to a spatial pattern of flammability whereby surface fuels in densely vegetated areas are more often less likely to burn than those in the open forests. These wet compartments limit landscape-scale fire activity by creating dis-connectedness in the available fuel. Heatwaves and the duration of dry spells determine the degree with which these wet compartments persist as barriers to fuel connectivity through a fire season. Elsewhere in the landscape, however, the large inputs of shortwave radiation to the forest floor means that surface fuels reach flammable conditions within several hours or days after rainfall and are therefore flammable for much of the fire season.

Introduction

The litter layer (otherwise known as fine surface fuel) is an important component of fire regimes in forests (Catchpole, 2002; Viegas et al., 1992; Wotton, 2009). This layer of dead biomass can comprise a large component of the overall fuel load, and its amount and moisture status can determine fire ignition probabilities and fire behaviour (Duff et al., 2013; Gould et al., 2011; McArthur, 1967). Furthermore, this fuel source can often be managed effectively with prescribed burning (Birk and Bridges, 1989), so there has been much effort in developing tools for predicting its accumulation rates and moisture dynamics. After fire, the accumulation of fine fuel is a function of litter input from live biomass and decomposition, both varying with climate but typically following a negative exponential accumulation curve (Olson, 1963; Ossola and Nyman, 2017; Raison et al., 1983; Thomas et al., 2014). In temperate eucalypt forests of Australia, the accumulation of surface fuel after fire approaches an asymptote at about 5–10 years after burning, stabilising at loads in the range of 10–20 t ha−1 (Duff et al., 2013; Gould et al., 2011; Thomas et al., 2014), with values a function of system productivity.

The availability of surface fuels to burning depends on their moisture content (Plucinski and Anderson, 2008; Possell and Bell, 2013), which is determined by the hydrological properties of the litter, precipitation and the microclimate above and within the litter layer (Matthews, 2006). In wet conditions when gravimetric water content of litter (GWClit) is above the fibre saturation point (> ∼ 0.35 kg kg−1), the temporal changes in moisture content are caused by drainage and evaporation of free water. In dry conditions when GWClit < ∼ 0.35 kg kg−1 the temporal changes in moisture content is caused by desorption and absorption of moisture as the litter approaches some equilibrium moisture content (EMC) with the surrounding atmosphere. The bulk of research on moisture contents of surface fuels has focused on representing changes in moisture of dry fuels when the assumptions underlying EMC models are applicable (Catchpole et al., 2001; Matthews et al., 2010; Resco de Dios et al., 2015; Sharples and McRae, 2011; Slijepcevic et al., 2013). Resolving variation in moisture contents in this low range of GWClit is important for predicting fire behaviour and ignition probabilities in a dry and flammable landscape when fuels are within the ignitable range.

However, in landscapes with complex topography and variable vegetation, the range of GWClit at any particular point in time may be very large, resulting in fine surface fuels that are in different stages of drying depending on the attributes of the specific location (Cawson et al., 2017; Gibos, 2010; Nyman et al., 2015; Slijepcevic et al. (in press)). A wide range of GWClit can therefore co-exist in the landscape over relatively small spatial scales. For instance, in a study of topographic effects on litter moisture Nyman et al. (2015) found that the difference in moisture content as a result of slope orientation (aspect and slope gradient) meant that ∼50% of the time during summer the assumptions associated with EMC models would apply to the equatorial-facing slope but not the polar-facing slope, limiting the predictive capacity of these models to only a fraction of the landscape. More shaded conditions on the polar-facing slopes meant that the dominant hydrological process controlling flammability of the litter was related to evaporation of free water and not desorption/adsorption processes associated with GWClit changes in dry fuel beds. In landscapes where a wide range of moisture contents co-exit, the rate at which water evaporates from fuels in wet and shaded parts of the landscape (when GWClit > 0.35 kg kg−1) may be more important for ignition and fire spread across the landscape than the exchange of moisture within fuels that are already dry (GWClit < 0.35 kg kg−1). High moisture contents in wet and shaded parts of the landscape promote patchiness in flammable surface fuel, thus reducing the likelihood of ignition sources (e.g. lightning) translating to active fire, and limiting fire spread in the event of a fire having started in a dry patch (Sharples, 2009). These factors related to spatial variation in GWClit, patchiness and connectivity of fuels are particularly important during benign fire weather (e.g planned burning) when fire behaviour and burn outcomes are most sensitive to flammability differentials (Bradstock, 2010; Bradstock et al., 2010; McCaw, 2013; McRae et al., 1979).

Existing process-based models for predicting GWClit take into account the full spectrum of hydrological processes underlying moisture dynamics in fuels in both wet and dry conditions (Matthews, 2006; Nelson Jr., 2000). However, the capacity to implement these models at the landscape scale is limited because we currently lack sufficient insight into the interactive effects of topography, climate and vegetation on the driving variables that control fuel moisture at the forest floor. Variation in vegetation structure with slope orientation and drainage position can be large (Gutiérrez-Jurado and Vivoni, 2013; Kirpatrick and Nunez, 1980; Nyman et al., 2015; Zhou et al., 2013), which in turn affects the local energy and water budget. And variation in vegetation structure with topography depends on climate. For instance, in a location with very high precipitation where is water not limiting, the vegetation structure is less dependent on aspect and drainage position than in a location where water is limiting. So changes in vegetation structure and the partitioning of energy into transpiration and evaporation across topographically complex landscape are sensitive to the amount of water that is available for plants to growth (Huxman et al., 2005).

Most of the recent efforts to measure and model the effects of vegetation on below-canopy microclimate have been aimed at improving predictions of snowmelt, which is function of the same variables that drive evaporation from fuels (Hardy et al., 2004; Moeser et al., 2014; Musselman et al., 2012; Reid et al., 2014; Seyednasrollah and Kumar, 2014). These studies and those from the fire literature (Schiks et al., 2015; Thompson et al., 2015; Walsh et al., 2017) have shown that microclimate on the forest floor is highly sensitive to forest structure. However, there are few studies that quantify the impacts of vegetation on microclimate more broadly across a landscape with complex terrain where eco-hydrological feedbacks result in complex patterns of spatial variability in vegetation, and where microclimate is subject to combined effects of terrain and vegetation. The paucity of research in this area limits our capacity to understand and model how surface fuel in forested landscapes dry out and become available to burn. How do the interactive effects of vegetation and topography play out in terms in spatial patterns of fuel moisture, microclimate and forest floor evaporation? What are the implication of these effects for predicting fuel moisture variation across complex landscapes? These questions are particularly pertinent in water limited environments where small changes in radiation and rainfall has large implications for eco-hydrology and fire regimes (Boer et al., 2016; Wang et al., 2014; Zhang et al., 2017a).

In this study we therefore aim to quantify the interactive effects of topographic position, aridity and vegetation cover on litter moisture, microclimate and potential evaporation above the forest floor. The work is set in context of fuel moisture dynamics and fire regimes but has application to hydrological processes and partitioning of evaporation and transpiration in these systems more generally. The aims are achieved by;

  • 1

    Monitoring litter moisture in different topographic positions across a rainfall gradient and examining the landscape-scale implications of the observed spatial variation.

  • 2

    Parameterising models of net radiation and potential evaporation below forest canopies in complex terrain using microclimate data from instrumented plots.

  • 3

    Using the models to identify major factors underlying variation in evaporation from the litter layer.

Section snippets

Overview

Data on microclimate and litter moisture were collected at 4 sites with different mean annual precipitation (MAP). At each of the 4 sites, experimental plots (20 × 20 m) were established in different positions with contrasting aspect and contributing drainage area. There are a total of 12 experimental plots. Data from these plots were used to i) quantify variation in litter moisture across the domain, and ii) parameterise a model of net radiation and potential evaporation (Ep) above the forest

Spatial-temporal patterns of litter moisture

Gravimetric moisture in the litter packs (GWClit) was highly variable within and between sites during the summer and autumn drying periods (Fig. 4; Table 2, Table 3). In summer the GWClit on the last day of the drying period ranged from 0.11 to 1.36 kg kg−1 with Christmas Hills North and The Triangle South being the driest and wettest respectively (Fig. 4 a and c; Table 2). In autumn the GWClit ranged from 0.19 to 2.06 kg kg−1 with Christmas Hills North (The Triangle North) being the driest

Discussion

Moisture content (GWClit) in the litter packs was highly variable at both local (0.1–1 km) and regional (1–10 s km) scales. At the regional scale there is a systematic increase in litter moisture content with increasing annual precipitation (MAP). However, this patterns is driven mainly by vegetation which increases in density with increased MAP. Increased vegetation density means less shortwave radiation reaching the litter layer. Temperature, which decreases at higher elevations, also tends

Conclusions

Mountainous landscapes are subject to large variations in eco-hydrological processes due to spatial variation in the availability of water and energy at the earth surface. The orientation of the terrain, as determined by aspect and slope, affects the amount of incoming radiation, while elevation affects temperature due to convection and adiabatic expansion. These terrain-related effects on air temperature and radiation interact with biota and precipitation regimes to produce complex patterns of

Acknowledgements

Research funding was provided by Natural Disaster Resilience Grants Scheme (NDRGS), Melbourne Water and the Victorian Department of Environment, Land, Water and Planning (DELWP). We are grateful for being able to collect field measurements on land managed by Melbourne Water and Parks Victoria. Wim Bovill, Daniel Metzen, Philip Noske and Mats Nyman helped with setting up the instrumented plots. We thank two anonymous reviewers who helped improve on a previous version of the manuscript.

References (72)

  • V. Resco de Dios

    A semi-mechanistic model for predicting the moisture content of fine litter

    Agric. Forest Meteorol.

    (2015)
  • A. Slijepcevic et al.

    Testing existing models for predicting hourly variation in fine fuel moisture in eucalypt forests

    For. Ecol. Manage.

    (2013)
  • K.B. Wilson et al.

    Factors controlling evaporation and energy partitioning beneath a deciduous forest over an annual cycle

    Agric. For. Meteorol.

    (2000)
  • T.B. Wilson

    Leaf litter water content and soil surface CO2 fluxes in a deciduous forest

    Agric. For. Meteorol.

    (2014)
  • S.L. Berry et al.

    Co-Evolution of Climate, Soil and Vegetation, Encyclopedia of Hydrological Sciences

    (2006)
  • M.M. Boer

    Future changes in climatic water balance determine potential for transformational shifts in Australian fire regimes

    Environ. Res. Lett.

    (2016)
  • BoM

    Gridded Climate Data

    (2014)
  • R. Bradstock et al.

    Divergent responses of fire to recent warming and drying across south-eastern Australia

    Glob. Change Biol.

    (2014)
  • R.A. Bradstock

    A biogeographic model of fire regimes in Australia: current and future implications

    Global Ecology Biogeography

    (2010)
  • R.A. Bradstock et al.

    Effects of weather, fuel and terrain on fire severity in topographically diverse landscapes of south-eastern Australia

    Landsc. Ecol.

    (2010)
  • M.I. Budyko

    Climate and Life

    (1974)
  • N. Burrows

    Fire behaviour in jarrah forest fuels: 1. Laboratory experiments

    Calmscience-Como-

    (1999)
  • G. Caccamo et al.

    Using remotely-sensed fuel connectivity patterns as a tool for fire danger monitoring

    Geophys. Res. Lett.

    (2012)
  • E. Catchpole et al.

    Estimating fuel response time and predicting fuel moisture content from field data

    International J. Wildland Fire

    (2001)
  • W. Catchpole

    Fire properties and burn patterns in heterogenous landscapes

  • D.C. Cheal

    Growth Stages and Tolerable Fire Intervals for Victoria's Native Vegetation Data Sets. 1742425895

    (2010)
  • T.J. Duff et al.

    Predicting continuous variation in forest fuel load using biophysical models: a case study in south-eastern Australia

    Int. J. Wildland Fire

    (2013)
  • T.J. Duff et al.

    Revisiting wildland fire fuel quantification methods: The challenge of understanding a dynamic, biotic entity

    Forests

    (2017)
  • P.M. Fernandes et al.

    Using fuel and weather variables to predict the sustainability of surface fire spread in maritime pine stands

    Can. J. For. Res.

    (2008)
  • K.E. Gibos

    Effect of Slope and Aspect on Litter Layer Moisture Content of Lodgepole Pine Stands in the Eastern Slopes of the Rocky Mountains of Alberta

    (2010)
  • H.A. Gutiérrez-Jurado et al.

    Ecogeomorphic expressions of an aspect-controlled semiarid basin: II. Topographic and vegetation controls on solar irradiance

    Ecohydrology

    (2013)
  • T.E. Huxman

    Ecohydrological implications of woody plant encroachment

    Ecology

    (2005)
  • D.A. Jones et al.

    High-quality spatial climate data-sets for Australia

    Aust. Meteorol. Oceanogr. J.

    (2009)
  • S. Kang et al.

    The effects of spatial aggregation of complex topography on hydroecological process simulations within a rugged forest landscape: development and application of a satellite-based topoclimatic model

    Can. J. For. Res.

    (2004)
  • A. Kennedy et al.

    Ecological fire management in North East Victoria

  • J. Kirpatrick et al.

    Vegetation-radiation relationships in mountainous terrain: Eucalypt-dominated vegetation in the Risdon Hills, Tasmania

    J. Biogeogr.

    (1980)
  • Cited by (0)

    View full text