Nutrition and mycorrhizae affect interspecific patterns of seedling growth on coarse wood and mineral soil substrates JOHN L. WILLIS
1,
AND MICHAEL B. WALTERS2
1
Department of Forestry, Mississippi State University, 361 Thompson Hall, 775 Stone Blvd., Mississippi State, Mississippi 39762 USA 2 Department of Forestry, Michigan State University, 124 Natural Resources Building, 480 Wilson Road, East Lansing, Michigan 48824 USA Citation: Willis, J. L., and M. B. Walters. 2018. Nutrition and mycorrhizae affect interspecific patterns of seedling growth on coarse wood and mineral soil substrates. Ecosphere 9(7):e02350. 10.1002/ecs2.2350
Abstract. Smaller-seeded tree species often require mineral soil and coarse woody debris (CWD) to establish. Limited data suggest that species-based variation in tree seedling–CWD relations exists; however, the generalizability of this pattern and its mechanistic basis remain unknown. We investigated interspecific substrate–seedling relations and the potential that differences in mycorrhizal status and substrate nutrition underlie these patterns with a potted plant experiment. Eight temperate tree species were grown in a lath shade house on six species of CWD and mineral soil that had either been sterilized with gamma irradiation or left untreated to separate the effects of mycorrhizae from substrate nutrition. We assessed the impact of substrate identity on height growth and investigated possible mechanisms via relating growth to substrate inorganic N concentration, seedling foliar N (%), and mycorrhizal colonization. Seedling height varied across substrates consistently among species (i.e., substrate effects were strong, but seedling species x substrate interactions were weak). Seedlings were generally tallest on mineral soil followed by Betula papyrifera and Thuja occidentalis CWD, and shortest on B. alleghaniensis or Acer saccharum CWD. Sterilization main effects on height were not significant; however, the response of species to sterilization varied, as Acer rubrum and T. occidentalis seedlings were significantly shorter on sterilized substrate. For all seedling species, growth correlated positively with substrate [N] and foliar N. Among the four species examined for mycorrhizae, root colonization density was highest on mineral soil and lowest on A. saccharum and B. alleghaniensis CWD). For these species, mean colonization density was more strongly associated with mean seedling height than substrate inorganic [N]. Beneficial mycorrhizal effects were suggested by positive height–mycorrhizal root density relationships for B. alleghaniensis, A. rubrum, and T. occidentalis. Collectively, for the group of northern hardwood tree species examined in this study, these results demonstrate that in a controlled environment, most species grow best on mineral soil, B. alleghaniensis and Picea glauca attained maximum growth on CWD, species-specific growth responses occur on CWD, and that substrate nutrient availability and mycorrhizal fungi contribute to the variation observed in seedling growth response across substrates. Key words: coarse wood; mycorrhizae; regeneration; seedlings; substrate. Received 11 December 2017; revised 30 May 2018; accepted 4 June 2018. Corresponding Editor: Karen A. Haubensak. Copyright: © 2018 The Authors. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. E-mail:
[email protected]
INTRODUCTION
seedlings, acquiring adequate resources may be particularly challenging, as they lack the root system and leaf area needed to exploit resources beyond their immediate proximity. As such, a
Accessing resources is critical for seedling survival and development. For recently germinated
❖ www.esajournals.org
1
July 2018
❖ Volume 9(7) ❖ Article e02350
WILLIS AND WALTERS
Limited evidence also suggests that CWD is not a generic substrate. Rather, the ability of CWD to support developing tree seedlings may differ among phyletic groups (e.g., conifers vs. angiosperms) or individual tree species (Cornett et al. 2001, Marx and Walters 2006, 2008, Orman et al. 2016). For instance, evidence suggests that CWD from conifer species is a more favorable substrate for smaller-seeded species seedling establishment and survival than CWD from angiosperm species (Cornett et al. 2001, Bolton and D’Amato 2011). While no study has explicitly tested this notion for broader conifer vs. angiosperm groups, results reported by Marx and Walters (2006) showed that yellow birch seedlings grew larger on eastern hemlock CWD compared to sugar maple (Acer saccharum Marsh.) CWD. Moreover, in a natural experiment, Marx and Walters (2008) reported that eastern hemlock CWD also supported more and older eastern hemlock and yellow birch seedlings than sugar maple CWD, suggesting that eastern hemlock CWD may provide a better environment for seedling survival than sugar maple CWD. What remains unknown, however, is whether or not there are generalizable trends in substrate availability across CWD substrates (e.g., conifer vs. angiosperm), why certain species of CWD are more favorable than others for smaller-seeded species seedling survival, and whether larger-seeded species seedling survival varies across CWD substrate. Variation in resource availability may explain why certain substrates are more favorable for seedling development than others. Among nutrients, nitrogen (N) is generally considered the most limiting nutrient in temperate forests (Cole and Rapp 1981), and it has been shown to be less available on CWD than mineral soil (Harmon et al. 1986, Baier et al. 2006). Variation in N availability also exists among species of CWD (Takahashi et al. 2000, Laiho and Prescott 2004) and has been shown to positively correlate with yellow birch seedling survival and growth (Marx and Walters 2006, 2008). Collectively, these patterns suggest that variation in nutritional quality, and specifically N availability, may partly underlie differences among species of CWD in their ability to support developing tree seedlings. Given fragmentary evidence of growth-limiting and variable N availability among species of
seedling’s germination substrate may have a strong bearing on its early growth and survival (Caspersen and Saprunoff 2005). Seedlings are often found over a range of substrates in the forest understory. In managed northern hardwood forest in the Lake States region, hardwood litter is the most common seedling establishment substrate (Willis et al. 2015). Compared to other substrate types, hardwood litter may present a more imposing physical barrier for root radicle penetration, and resource availability may be more variable and limited as the most stable supply of moisture and mineral nutrients is contained in the underlying humus and mineral soil layers (Houle 1992, Prescott et al. 2000). As such, recently germinated seedlings must exert energy growing their roots through the litter layer in order to access the underlying resources. Given their limited supply of seed-stored carbohydrates, smaller-seeded species, such as birch (Betula spp.), northern white cedar (Thuja occidentalis L.), and eastern hemlock (Tsuga canadensis L.), have less reserve energy available to allocate toward root development (Moles and Westoby 2004). Consequently, in northern hardwood forests, smaller-seeded species seedling establishment is often confined to mineral soil and heavily decayed, large pieces of coarse woody debris (CWD), where resources are more easily accessed and less variable (Cornett et al. 2001, Shields et al. 2007, Marx and Walters 2008, Bolton and D’Amato 2011, Willis et al. 2015). Nevertheless, mineral soil and CWD should not be considered equivalent seedling establishment substrates. Evidence to date suggests that mineral soil provides a higher quality nutrient resource environment than CWD (Cornett et al. 1997, Mori et al. 2004). However, due to their small initial stature, smaller-seeded seedlings may be more likely to face several hazards on mineral soil than on CWD, including greater risk of high water tables, pathogens, competing vegetation, and leaf smothering (Harmon and Franklin 1989, Packer and Clay 2000, Simard et al. 2003, O’Hanlon-Manners and Kotanen 2004, Walters et al. 2016). Smaller-seeded seedlings established on CWD may also be at lower risk to drought stress, as CWD substrates are known for their high moisture-holding capacity (Cornett et al. 1997, Mori et al. 2004). ❖ www.esajournals.org
2
July 2018
❖ Volume 9(7) ❖ Article e02350
WILLIS AND WALTERS
species established on seven common forest substrates (six species of CWD and mineral soil) that was either sterilized or unsterilized. Seeking mechanistic explanations to the patterns observed, we examined seedling growth and leaf N as functions of mycorrhizal colonization (for two AMF and two EMF species), and substrate [N]. Specifically, we predicted the following: (1) Seedling growth and survival will vary among substrates and seedling responses will be species specific; (2) seedling survival, growth, and foliar N will be positively associated with substrate identity (CWD species and mineral soil) N availability, and this effect will be strongest for sterilized substrates where N availability is unmodified by mycorrhizal fungi; (3) species of CWD will differ in their ability to supply mycorrhizal inoculum; (4) for unsterilized substrates, survival and growth will be positively related to mycorrhizal density; and (5) for AMF-associated species, mycorrhizal colonization and growth will be highest for seedlings growing on unsterilized mineral soil, and will be lower than that of EMF-associated species on unsterilized CWD. Collectively, our findings will provide functional information on species-specific relationships with individual substrate types in northern hardwood forests and can help inform silvicultural efforts aimed at maintaining biodiversity.
CWD and lower availability than mineral soil, differences among species of seedlings in their ability to access nutrients via mycorrhizal fungi and differences among species of CWD in their ability to provide mycorrhizal inoculum could be important components of success for seedlings on CWD. Mycorrhizal fungi can be found in mineral soil and in CWD, suggesting that seedlings growing on either type of substrate have the opportunity to become colonized (Harvey et al. 1987, Smith and Read 2008, Tedersoo et al. 2008). Several studies have reported positive seedling growth effects of mycorrhizal colonization on mineral soil (Smith and Read 2008), and Marx and Walters (2006) reported a positive association between colonization density and seedling growth for yellow birch and hemlock seedlings on three types of CWD substrate. Nevertheless, the odds of a seedling becoming colonized may not be the same across all substrate types, as Marx and Walters (2006) also reported that yellow birch seedlings were colonized less frequently when grown on sugar maple CWD compared to yellow birch and eastern hemlock CWD. Therefore, it is possible that differences in mycorrhizal colonization could account for the strong associations reported between individual seedling species and particular substrate types (Caspersen and Saprunoff 2005, Marx and Walters 2006, 2008). One factor that may affect the odds of a seedling successfully establishing on CWD is the type of fungal symbiont it associates with. Species associating with ectomycorrhizal fungi (EMF) may have a higher probability of becoming colonized on CWD than species associating with arbuscular mycorrhizal fungi (AMF), as EMF are capable of existing saprotrophically on CWD (Lindahl et al. 1999) and AMF are not. Furthermore, EMF can access nitrogen and phosphorous from a variety of pools that are inaccessible to AMF (Turner 2008, Courty et al. 2010, Phillips et al. 2013). Consequently, EMF-associated species may better able to survive and grow on nutrient-deficient substrates, such as CWD, than AMF-associated species. To more holistically examine the influence of individual substrate types on northern hardwood regeneration, we conducted an outdoor lath shade house potted plant experiment that tracked the growth and survival of eight tree ❖ www.esajournals.org
METHODS Field methods Throughout the summer of 2012, we located CWD samples of sugar maple, yellow birch, paper birch (Betula papyrifera Marsh), eastern hemlock, northern white cedar, and balsam fir (Abies balsamea (L.) Mill.) in five locations in the northern Lower and Upper Peninsulas of Michigan. Specifically, samples were collected from The Huron Mountain Club, a privately owned eastern hemlock–northern hardwood-dominated forest preserve, and also two conifer swamps and two upland hardwood stands managed by the Michigan Department of Natural Resources (MDNR). All CWD located were in decay stage 3 (bole could support itself but strong signs of decay in the heartwood and sapwood) or 4 (boles could no longer support itself with missing sapwood and crumbled heartwood; Graham and Cromack 1982) and were identified to species by 3
July 2018
❖ Volume 9(7) ❖ Article e02350
WILLIS AND WALTERS
addition to the seeded/planted plots, we left three pots of each substrate-sterilization treatment combination unplanted for inorganic nitrogen extractions (described in later section). We conducted the experiment in three open hoop houses nested within an outdoor opensided lath-roofed house (~50% shade). Each hoop house was covered with clear plastic to exclude N additions from precipitation with ends and bottoms on sides open, allowing air to move freely through the houses. To quantify the light environment, we measured photosynthetically active radiation (PAR) within each house with a quantum sensor (Apogee Instruments, Logan, Utah, USA). Photosynthetically active radiation ranged from 254 to 344 lmolm 2s 1 (~15–20% of open-sky quanta) across houses under cloudless mid-morning conditions. Potential issues with varying PAR levels over the lath house were mitigated via random assignment of individual pots across the houses and periodic rotations of seedlings within the house during the experiment. In addition, all pots were kept at least five meters from the edge of each house to minimize edge effects. Throughout the experiment, we monitored randomly selected seedlings for substrate moisture within the rooting zone. Dryness within the rooting zone of any seedling triggered watering for all seedlings. Each seedling was watered with deionized water until the pots became saturated. We used seedling height as our index of growth. Beginning one month after seedling establishment in pots (early July 2013), we took an initial height measurement (cm) for every germinated/planted seedling. Seedling height growth and survival were then assessed at the end of September. At this time, four replicates of yellow birch and paper birch, and three replicates of all other species were randomly selected from each unique treatment substrate combination and destructively harvested to examine mycorrhizal colonization and foliar N content. This was done to preserve our ability to associate mycorrhizal colonization and foliar N content with the observed height growth patterns in the event of low-overwinter survival. Harvest intensity was greater for paper birch and yellow birch because these species had already differentiated in height, and thus provided the most promising data to explore. Fewer replicates of the other
bark and branching pattern. We collected at least 16 samples of CWD for each species except northern white cedar (11). In addition to CWD, we collected mineral soil–humus from each site from multiple soil cores (to 20 cm depth). All samples were bagged, placed in an ice-filled cooler, and transported to the Tree Research Center at Michigan State University for temporary storage (4°C).
Shade house methods In late May 2013, CWD, by species, and soil samples were homogenized across sites and divided into two equally sized populations with one randomly selected for a gamma irradiation sterilization treatment (30–60 kGy; Sterigenics, Schaumburg, Illinois, USA; McNamara et al. 2003). Sterilized and unsterilized substrates were loaded into pots (10.2 width 9 24.1 cm height) (Stuewe and Sons, Corvallis, Oregon, USA) already one-third filled with sterilized (72 h at 110°C) coarse silica sand (Best Sand, Chardon, Ohio, USA). Each pot was seeded with at least five sterilized pre-stratified seeds of one species with the goal of obtaining one germinant per pot. Pots containing more than one germinant were thinned to one seedling prior to the start of the experiment. Species seeded included red maple (Acer rubrum L.), northern white cedar, eastern hemlock, yellow birch, paper birch, eastern white pine (Pinus strobus L.), and white spruce (Picea glauca (Moench)). With the exception of red maple and northern white cedar, all species associate primarily with EMF. Seeds were sterilized with a water–10% bleach mixture. All seeds were obtained from the Michigan Department of Natural Resources (MDNR) or Sheffield’s Seed (Locke, New York, USA; Plant Hardiness Zones 4b and 5b, United States Department of Agriculture). With the exception of red maple seedlings, which germinated poorly across substrates, seed germination allowed for at least eight replicates on almost all unique treatment combinations. Substrates with less than eight initial replicates were excluded from the study. To compensate for the lack of red maple germination, we transplanted newly germinated wildlings collected from a stand near the Tree Research Center into pots. The root system of each wildling was sterilized in a water– 10% bleach mixture just prior to planting. In ❖ www.esajournals.org
4
July 2018
❖ Volume 9(7) ❖ Article e02350
WILLIS AND WALTERS
species were harvested in order to preserve our ability to detect future (2nd year) changes in growth and survival, as seedlings had yet to differentiate by the end of the first growing season. The remaining seedlings were overwintered in the hoop houses and then assessed for final growth and survival and harvested approximately two months into the second growing season (i.e., end of June 2014). Limited resources prevented us from quantifying mycorrhizal colonization for all species, so we quantified it for two AMF species (northern white cedar, red maple) and two EMF species (yellow birch, eastern white pine) from each unsterilized substrate (procedure described further). Although these species are thought to primarily associate with either EMF or AMF, each species was examined for both EMF and AMF colonization to ensure the accuracy of our colonization assessment. For northern white cedar and eastern white pine, four seedlings were randomly selected for mycorrhizal examination for each unsterilized substrate identity. Limited seedling germination and survival on certain substrates restricted our sampling for yellow birch and red maple on certain substrates. For these species, between two and five seedlings were sampled on each substrate type. To confirm the sterilizing effects of the gamma irradiation treatment, we also examined mycorrhizal colonization on one seedling grown on each sterilized substrate from the abovedescribed subset of four species. No mycorrhizal colonization was detected for seedlings grown on sterilized substrate.
Each seedling selected for mycorrhizal inspection was examined for both EMF and AMF colonization using the following procedure. Residual substrate from harvested roots was gently removed by floating the roots in water. Fine roots ( F
