Ecohydrological energy inputs in semiarid coniferous gradients: Responses to management- and drought-induced tree reductions

Abstract

Large-scale, rapid reductions in forest and woodland tree cover caused by fire, drought-induced die-off, or wildfire-mitigating thinning prescriptions, all three of which differentially affect canopy structure, are increasingly altering coniferous-dominated landscapes across extensive regions such as western USA. These types of reductions in canopy cover can result in substantial increases in near-ground solar radiation, which in turn drive numerous ecological processes. However, existing relationships for how reductions in canopy cover translate into changes in incoming near-ground solar radiation do not account for the ways in which fire, die-off, and thinning differentially alter either or both the foliar and woody components of canopy architecture and the degree to which such alterations depend on foliar density. We systematically quantified trends in near-ground solar radiation for a broad range of canopy cover for two of the most extensive semiarid coniferous forest and woodland vegetation types in the western USA: those dominated by a combination of piñon and juniper (using Pinus edulis and Juniperus monosperma as representative species) or by ponderosa pine (Pinus ponderosa). We used hemispherical photography to account for how canopy architecture affected mean and variance in near-ground solar radiation over a broad range of canopy cover (from as low as ∼5% to as high as ∼85%). For both vegetation types, we evaluated four disturbance types: undisturbed, controlled burns, drought- and beetle-induced die-off, and prescriptive thinning treatments. We also assessed near-ground solar radiation for undisturbed vegetation spanning an elevation continuum that included both piñon-juniper and ponderosa pine vegetation types. Our results quantify how near-ground solar radiation varies substantially and systematically among forest gradient types and as a function of forest disturbance type. Trends in near-ground solar radiation differed among gradients associated with fire, die-off, or thinning, dependent on how each affects the foliar and/or woody components of canopy architecture. Deviations from undisturbed conditions for remaining disturbed tree cover were greatest for burned, intermediate for die-off and least for thinned. The differences in microclimate quantified here and how they vary with type of tree reduction are relevant for assessing vegetation responses following reductions in tree cover. In addition, the differences are large enough to require consideration in evaluating land surface interactions of forests and woodlands with the atmosphere (e.g., increases of >40 W m⁻² relative to undisturbed conditions). Our results provide a means to enable managers to rapidly relate readily-obtainable field estimates of canopy cover to approximate estimates of near-ground solar radiation.

Introduction

A key characteristic of much of the terrestrial biosphere and its forested landscapes is the proportion of woody plant cover. Landscape gradients of woody cover fundamentally differentiate ecosystems ranging from sparse savannas and parklands to dense woodlands and forests with closed canopies, particularly for semiarid gradients (Belsky and Canham, 1994, Breshears, 2006). The amount of cover and spatial heterogeneity of woody vegetation influences processes at ecosystem and landscape scales (Breshears, 2006), perhaps most directly via canopy cover effects on the amount and variation of radiation that penetrates the overstory and arrives near the land surface. Below-canopy solar radiation has broad ecohydrological relevance that includes effects on soil evaporation and other aspects of the water budget; light limitations and phenological responses for understory herbaceous species and forest and woodland seedlings; and heat loads that can drive nurse plant effects on other plants or behavioral ecology and habitat utilization for animals (Kittredge, 1948, Forseth et al., 2001, Guthery et al., 2005, Huxman et al., 2005, Suarez and Kitzberger, 2008).

Near-ground solar radiation depends on the overall amount of canopy cover at a site. Although near-ground solar radiation generally increases with reductions in forest or woodland canopy cover, specific characteristics of such trends that may be potentially important in an ecohydrological context have not been directly assessed, including the rate of attenuation with increases in canopy cover, the nature the relationship (e.g., linear, curvilinear, or polynomial), and the spatial variation in near-ground solar radiation. Collectively these characteristics could result in substantial differences in landscape-scale estimates of near-ground solar radiation that are important for forest and woodland dynamics, as well as atmospheric and land surface interactions (Gash and Nobre, 1997, Kurc and Small, 2004, Notaro et al., 2006, Chapin et al., 2008). Initial estimates of such relationships have been modeled previously by assuming homogenous canopy architecture (Martens et al., 2000), but these relationships do not account for the role of canopy heterogeneity associated with variation in the foliar and woody components of the canopy. Consequently, more direct assessments of how near-ground solar radiation changes systematically with forest and woodland canopy cover and canopy architecture are needed.

The importance of assessing landscape-scale trends in near-ground solar radiation is particularly relevant given a suite of ways in which forest and woodland tree cover can be reduced by both management and drought. Large-scale, rapid reductions in tree cover caused by fire, drought- and beetle-induced die-off, or wildfire-mitigating thinning prescriptions are increasingly altering coniferous vegetation across extensive regions such as western USA. Piñon-juniper (Pinus and Juniperus species) woodlands and ponderosa pine (Pinus ponderosa P. & C. Lawson) forests are the predominant semiarid coniferous vegetation types in the western USA (McPherson, 1997). In some cases, such reductions could return forest structure to less dense, pre-fire suppression conditions, which is indeed the target of some management activities. Other types of disturbances such as windfall can also reduce tree canopy cover, and the dynamics of semiarid coniferous vegetation can be driven by many factors that can be interrelated. Nonetheless thinning, fire, and die-off are three particularly prevalent, extensive and rapid ways in which tree cover is reduced and warrant particular focus. Forest management in these semiarid coniferous systems is undertaking substantial thinning with specified prescriptions that might change trends in the mean amount of near-ground solar radiation along gradients of tree cover and/or the associated spatial variance (Naumburg and DeWald, 1999). Even more dramatic are extensive reductions in tree cover that can result from fire (Allen et al., 2002, Romme et al., 2009, Hurteau and North, 2009), or from drought- and beetle-induced tree die-off, as recently observed across much of the western USA (Breshears et al., 2005, Gitlin et al., 2006, Floyd et al., 2009, Negron et al., 2009) and across the globe (Allen et al., 2010). Due to climate change, more such large-scale disturbances are likely (Nicholls, 1995, Allen et al., 2002, Westerling et al., 2006, Adams et al., 2009, Adams et al., 2010). Notably, thinning, fire, and drought-induced die-off differ with respect to their effects on the foliar and woody components of the tree: thinning removes both the foliar and woody components systematically, whereas die-off only initially removes the foliar components, and the effects of fire can be complex, affecting either only the foliar component or both the foliar and woody components. Importantly, the way in which each of these three types of tree reduction could alter near-ground solar radiation depends not only on the effect of each on the foliar and woody components but are also expected to do so in a density-dependent manner. These three major types of tree reductions driven by management or drought are occurring across expansive landscapes, but effects of such reductions on near-ground radiation are complicated and less well understood than might be initially thought because they have not been systematically evaluated both across types of reduction and across a gradient of tree density. Such estimates could help managers relate readily-obtainable field estimates of canopy cover to approximate estimates of near-ground solar radiation, thereby allowing more direct consideration of how changes in tree cover on near-ground solar radiation may influence properties of interest such as soil water content and understory vegetation.

In short, although solar radiation penetration through canopy overstory should increase as canopy cover decreases, surprisingly how such response functions change with vegetation type, canopy cover, and type of canopy reduction remains unclear. We addressed this knowledge gap by systematically assessing near-ground solar regimes along gradients of canopy cover associated with piñon-juniper woodland and ponderosa pine forest, using example study sites from northern Arizona. We assess how site-specific tree cover and type of canopy reduction relate to mean conditions of solar radiation and corresponding spatial variance in direct radiation penetration through overstory canopy. Our first objective was to quantify near-ground radiation as related to canopy cover (from as low as ∼5% to as much as ∼85%) for three undisturbed (reference) coniferous canopy cover gradients. These reference gradients included one for piñon-juniper woodlands, one for ponderosa pine forests, and one for an elevation continuum that spanned both vegetation types. Our second objective was then to quantify near-ground radiation as related to canopy cover for locations impacted by fire, die-off, and thinning for both piñon-juniper woodlands and for ponderosa pine forests. We also conducted supplemental analyses to assess the robustness of the results. We discuss our results in the context of their relevance for ecosystem management, including assessing land surface–atmosphere interactions in climate change projection.