Matthew J. Rinella

Predicting plant responses to mycorrhizae: integrating evolutionary history and plant traits.

We assessed whether (1) arbuscular mycorrhizal colonization of roots (RC) and/or plant responses to arbuscular mycorrhizae (MR) vary with plant phylogeny and (2) MR and RC can be more accurately predicted with a phylogenetic predictor relative to a null model and models with plant trait and taxonomic predictors. In a previous study, MR and RC of 95 grassland species were measured. We constructed a phylogeny for these species and found it explained variation in MR and RC. Next, we used multiple regressions to identify the models that most accurately predicted plant MR. Models including either phylogenetic or phenotypic and taxonomic information similarly improved our ability to predict MR relative to a null model. Our study illustrates the complex evolutionary associations among species and constraints of using phylogenetic information, relative to plant traits, to predict how a plant species will interact with AMF.

Managing soil nitrogen to restore annual grass‐infested plant communities: effective strategy or incomplete framework?

Theoretical and empirical work has established a positive relationship between resource availability and habitat invasibility. For nonnative invasive annual grasses, similar to other invasive species, invader success has been tied most often to increased nitrogen (N) availability. These observations have led to the logical assumption that managing soils for low N availability will facilitate restoration of invasive plant-dominated systems. Although invasive annual grasses pose a serious threat to a number of perennial-dominated ecosystems worldwide, there has been no quantitative synthesis evaluating the degree to which soil N management may facilitate restoration efforts. We used meta-analysis to evaluate the degree to which soil N management impacts growth and competitive ability of annual and perennial grass seedlings. We then link our analysis to current theories of plant ecological strategies and community assembly to improve our ability to understand how soil N management may be used to restore annual grass-dominated communities. Across studies, annual grasses maintained higher growth rates and greater biomass and tiller production than perennials under low and high N availability. We found no evidence that lowering N availability fundamentally alters competitive interactions between annual and perennial grass seedlings. Competitive effects of annual neighbors on perennial targets were similar under low and high N availability. Moreover, in most cases perennials grown under competition in high-N soils produced more biomass than perennials grown under competition in low-N soils. While these findings counter current restoration and soil N management assumptions, these results are consistent with current plant ecological strategy and community assembly theory. Based on our results and these theories we argue that, in restoration scenarios in which the native plant community is being reassembled from seed, soil N management will have no direct positive effect on native plant establishment unless invasive plant propagule pools and priority effects are controlled the first growing season.

Primary Productivity and Precipitation-Use Efficiency in Mixed-Grass Prairie: A Comparison of Northern and Southern US Sites

Abstract Precipitation-use efficiency (PUE) is a key determinant of aboveground net primary production (ANPP). We used long-term datasets to contrast ANPP and PUE estimates between northern (southeast Montana) and southern (north Texas) mixed-grass prairies. Effects of varying amounts and temporal distribution of precipitation on PUE were examined at the Montana site, using a rainout shelter and irrigation. Results show that 1) ANPP was 21% less in Montana than Texas (188 g · m−2 vs. 237 g · m−2); 2) plant function type (PFT) composition varied between the two study locations, with cool-season perennial grasses (CSPG) dominating in Montana (52%) and warm-season perennial grasses (WSPG) dominating in Texas (47%); 3) production dynamics varied between the two sites with 90% of ANPP completed by 1 July in Montana as compared to 31 August in Texas; 4) average PUE estimates were greater in Montana (0.56 g dry matter · m−2 · mm−1 of precipitation) than Texas (0.40 g · m−2 · mm−1); and 5) contributions to PUE estimates varied among PFT and location, with CSPG estimates being greater in Montana than Texas (52% vs. 31%) and WSPG estimates being greater in Texas than Montana (47% vs. 27%). Seasonal droughts and supplemental irrigations at the Montana site substantially altered ANPP, PFT biomass composition, and PUE. Results show PUE was responsive to PFT composition relative to amount and seasonal distribution of precipitation. Therefore, one should expect changes in ANPP and PUE to occur with shifts in precipitation patterns until PFT composition becomes adjusted to the regime.

A common soil handling technique can generate incorrect estimates of soil biota effects on plants

An active area of research seeks to understand how soil biota effects on plants vary across experimental factors (i.e. regions, treatments). The study biotas are obtained by gathering soil sample(s) from randomly selected location(s) within each experimental unit, with an experimental unit being a site within a study region or a field plot of a manipulative experiment. Then, plant growth is measured in glasshouse containers housing soil and biota from the various sites or plots. Results of these glasshouse bioassays are sensitive to a common soil handling decision. In particular, it is common to either: (1) fill each container with soil from one experimental unit (e.g. Callaway et al., 2004; Hood et al., 2004; Kardol et al., 2006; Wardle et al., 2012), or (2) fill each container with a mixture of soils from multiple experimental units (i.e. all sites within a region, all plots that received the same treatment) (e.g. Van der Putten et al., 1993; Nijjer et al., 2007; Felker-Quinn et al., 2011; Pendergast et al., 2013; Rodr ıguezEcheverr ıa et al., 2013; Yang et al., 2013; Gundale et al., 2014; Pizano et al., 2014; Hilbig & Allen, 2015; Larios & Suding, 2015) (Fig. 1). We define samples generated from these two approaches as ‘individual soil samples’ (ISS) and ‘mixed soil samples’ (MSS). The term ‘individual soil sample’ is slightly misleading, as ISS are often formed by mixing multiple samples gathered from the same experimental unit (i.e. pooling subsamples). Combining subsamples from individual experimental units is a perfectly acceptable approach. Conversely, the express purpose of this paper is to illustrate that, without exception, the MSS approach of mixing together soils from multiple experimental units (Fig. 1) is fatally flawed. Hypotheses regarding differences among regions or treatments cannot be legitimately tested by the MSS approach of mixing together soils from multiple sites within regions or multiple plots receiving the same treatment. The importance of this point is clearly underappreciated: We estimate, 52% of published studies use MSS in place of the correct ISS methodology (of 76 evaluated studies using ISS or MSS, 40 used MSS) (K. O. Reinhart & M. J. Rinella, unpublished, 2015). Estimating treatment (e.g. region, nutrient) differences entails computing residual variance. Residual variance describes variation not explained by treatments, and it is needed to compute relevant statistics (i.e. P-values, confidence intervals). In experiments considered here, there are two contributors to residual variance in plant growth: (1) spatial variation in soil biotas (i.e. site-to-site variation not explained by region, plot-to-plot variation not explained by treatment) and (2) glasshouse variation owing to environmental gradients (e.g. temperature) and plant genetics. With the ISS approach (Fig. 1), having two contributors to residual variance poses no unique analytic challenges, and standard regression and analysis of variance (ANOVA) approaches give correct inferences. With the MSS approach, all information regarding residual variation in soil mutualists and pathogens is lost, and if this variation is nonzero, MSS and ISS are guaranteed to give different inferences, with the MSS inferences being incorrect. More specifically, if residual variation in mutualists and/or pathogens is nonzero, statistical estimates from MSS will be falsely precise and evidence for differences among treatments (e.g. regions, nutrients) will be weaker than reported. Assuming only factors being studied cause soil mutualists/ pathogens to vary spatially is highly unrealistic, particularly given that plant disease expression (e.g. Martin & Loper, 1999) and soil microbe compositions (e.g. Ettema & Wardle, 2002; Ritz et al., 2004) are known to vary widely across even small spatial scales (i. e. < 1.0 m).

Fire Alters Emergence of Invasive Plant Species from Soil Surface-Deposited Seeds

Abstract Restoration of historic fire regimes is complicated by concerns about exotic plant invasions, yet little is known of how the two may interact. Seeds of Japanese brome, spotted knapweed, Russian knapweed, and leafy spurge were subjected to fire at six fuel loads (100 to 700 g m−2) and a nonburned control. Fires were simulated with field-cured grass and time–temperature profiles were developed from thermocouples at the soil surface. Emergence was determined by species and fuel load in growth chambers. Fuel load explained 98% of variation in mean heat dosage and emergence decreased with increasing fuel load across species. Emergence was reduced 79 to 88% relative to nonburned treatment with 100 g m−2 of fuel and at least 97% with 200 g m−2 of fuel. Emergence probabilities were less than 0.01 for all species but spotted knapweed with a 300 g m−2 fuel load. Results indicate high potential for fire to disrupt the life cycle of invasive species through direct seed mortality. The relationship between fuel load and seedling emergence provides good predictability of fire effects on surface-deposited seeds. A single fire is unlikely to eradicate many invasive species because they often produce abundant seeds and some will undoubtedly escape fire. However, abrupt reductions in seedling emergence with relatively light fuel loads indicate that fire may be an effective tool for increasing mortality of invasive plant seed across a broad range of habitats. Nomenclature: Japanese brome, Bromus japonicus Thunb. ex Murr.; spotted knapweed, Centaurea maculosa Lam.; Russian knapweed, Acroptilon repens (L.) DC; leafy spurge, Euphorbia esula L.

Spotted knapweed response to season and frequency of mowing

Spotted knapweed (Centaurea maculosa Lam.) is a non-indigenous weed that has invaded millions of hectares of rangeland in the United States. Mowing may be useful for reducing this weed. Our objective was to investigate the response of spotted knapweed and grasses to season and frequency of mowing. Response of grass and spotted knapweed to 16 mowing treatments applied annually for 3 years was studied at 2 sites. Treatments consisted of combinations of spring, summer, and fall mowing. Treatments were arranged in a randomized-complete-block design with 4 replications (16 treatments; 4 replications; 2 sites = 128 plots). After repeating mowing treatments for 3 years, a single fall mowing when spotted knapweed was in the flowering or seed producing stage reduced its cover and adult density as much as any treatment consisting of repeated mowing. Fall mowing decreased adult density 85 and 83% below that of the control at Sites 1 and 2, respectively. Treatments reduced seedling density at Site 2, but the response was not consistent between years or among treatments. Spotted knapweed cover was decreased by several mowing treatments at each site (10-36%), while grass cover was only decreased by 3 mowing treatments (18-23%) at Site 1 in 1998. We recommend a single annual mowing, applied at the flowering or seed producing stage, for the partial control of spotted knapweed. DOI:10.2458/azu_jrm_v54i1_rinella

GRASSLAND INVADER RESPONSES TO REALISTIC CHANGES IN NATIVE SPECIES RICHNESS

The importance of species richness for repelling exotic plant invasions varies from ecosystem to ecosystem. Thus, in order to prioritize conservation objectives, it is critical to identify those ecosystems where decreasing richness will most greatly magnify invasion risks. Our goal was to determine if invasion risks greatly increase in response to common reductions in grassland species richness. We imposed treatments that mimic management-induced reductions in grassland species richness (i.e., removal of shallow- and/or deep-rooted forbs and/or grasses and/or cryptogam layers). Then we introduced and monitored the performance of a notorious invasive species (i.e., Centaurea maculosa). We found that, on a per-gram-of-biomass basis, each resident plant group similarly suppressed invader growth. Hence, with respect to preventing C. maculosa invasions, maintaining overall productivity is probably more important than maintaining the productivity of particular plant groups or species. But at the sites we studied, all plant groups may be needed to maintain overall productivity because removing forbs decreased overall productivity in two of three years. Alternatively, removing forbs increased productivity in another year, and this led us to posit that removing forbs may inflate the temporal productivity variance as opposed to greatly affecting time-averaged productivity. In either case, overall productivity responses to single plant group removals were inconsistent and fairly modest, and only when all plant groups were removed did C. maculosa growth increase substantially over a no-removal treatment. As such, it seems that intense disturbances (e.g., prolonged drought, overgrazing) that deplete multiple plant groups may often be a prerequisite for C. maculosa invasion.

Growth Regulator Herbicides Prevent Invasive Annual Grass Seed Production

Abstract Auxinic herbicides, such as 2,4-D and dicamba, that act as plant growth regulators are commonly used for broadleaf weed control in cereal crops (e.g., wheat, barley), grasslands, and noncroplands. If applied at late growth stages, while cereals are developing reproductive parts, the herbicides can reduce seed production. We tested whether growth regulators have this same effect on the invasive annual grass Japanese brome. The herbicides 2,4-D, dicamba, and picloram were applied at typical field use rates to Japanese brome at various growth stages in a greenhouse. Picloram reduced seed production nearly 100% when applied at the internode elongation, boot, or heading stages of growth, whereas dicamba appeared to be slightly less effective and 2,4-D was much less effective. Our results indicate it may be possible to control Japanese brome by using growth regulator herbicides to reduce its seed production, thereby depleting its short-lived seed bank. Nomenclature: 2,4-D; dicamba; picloram; Japanese brome, Bromus japonicus Thunb.; barley, Hordeum vulgare L.; wheat, Triticum aestivum L.

Long-term population dynamics of seeded plants in invaded grasslands

In recent decades, dozens of studies have involved attempts to introduce native and desirable nonnative plant species into grasslands dominated by invasive weeds. The newly introduced plants have proved capable of establishing, but because they are rarely monitored for more than four years, it is unknown if they have a high likelihood of persisting and suppressing invaders for the long-term. Beyond invaded grasslands, this lack of long-term monitoring is a general problem plaguing efforts to reintroduce a range of taxa into a range of ecosystems. We introduced species from seed and then periodically measured plant abundances for nine years at one site and 15 years at a second site. To our knowledge, our 15-year data are the longest to date from a seeding experiment in invaded, never-cultivated grassland. At one site, three seeded grasses maintained high densities for three or more years, but then all or nearly all individuals died. At the second site, one grass performed similarly, but two other grasses proliferated and at least one greatly suppressed the dominant invader (Centaurea maculosa). In one study, our point estimate suggests that the seeded grass Thinopyrum intermedium reduced C. maculosa biomass by 93% 15 years after seeding. In some cases, data from three and fewer years after seeding falsely suggested that seeded species were capable of persisting within the invaded grassland. In other cases, data from as late as nine years after seeding falsely suggested seeded populations would not become large enough to suppress the invader. These results show that seeded species sometimes persist and suppress invaders for long periods, but short-term data cannot predict if, when, or where this will occur. Because short-term data are not predictive of long-term seeded species performances, additional long-term data are needed to identify effective practices, traits, and species for revegetating invaded grasslands.

Grass Seedling Demography and Sagebrush Steppe Restoration

Abstract Seeding is a key management tool for arid rangeland. In these systems, however, seeded species often fail to establish. A recent study in Wyoming big sagebrush steppe suggested that over 90% of seeded native grass individuals die before seedlings emerged. This current study examines the timing and rate of seed germination, seedling emergence, and seedling death related to this demographic bottleneck. We seeded monocultures of two native perennial bunchgrasses, Pseudoroegenaria spicata (Pursh) Á. Löve and Elymus elymoides (Raf.) Swezey, and one introduced bunchgrass, Agropyron desertorum (Fisch. ex Link) Schult., in 2007, 2008, and 2009 and tracked sown seed and seedling fate. Across the study years and species we found that germination was rapid and high, with species obtaining 50% germination by December, less than 2 mo after planting. Emergence of germinated seed did not occur until late February for A. desertorum and March for the two native grasses. In 2007 the majority of emergence and death was constrained to several weeks, whereas in 2008 and 2009 emergence and death was distributed across several months. The timing of seedling emergence did not influence survival probability or midday plant water potential (probability of exceedance < 0.05). Survival probabilities once seedlings emerged were greater for native species (0.71) than A. desertorum (0.51) in 2 of the 3 study yr (probability of exceedance > 0.98). The early germination of grasses following fall seeding, and the long 2- to 3-mo period that germinated grass seed remain in the soil before emerging, support the hypothesis that seedling recruitment might be limited largely by ecological processes and conditions during winter or early spring (such as soil freeze–thaw events, seed pathogens, or physical crusts). Delaying seeding to early winter or spring and other management tools that mitigate these factors driving this bottleneck might greatly improve restoration outcomes in these systems. Resumen Las resiembras son una herramienta clave de manejo para pastizales áridos. En estos sistemas, sin embargo, las especies sembradas a menudo no se establecen. En un estudio reciente en un pastizal de Artemisia en Wyoming se sugiere que más del 90% de los individuos sembrados de pastos nativos mueren antes que la plántula germine. Este estudio examina la época y tasa de germinación de las semillas, la aparición de la plántula, y la muerte de plántula relacionadas con el cuello de botella demográfico. Se sembraron monocultivos de dos especies nativas de pastos amacollados, Pseudoroegenaria spicata (Pursh) Á. Löve y Elymus elymoides (Raf.) Swezey, y también un pasto amacollado introducido, Agropyron desertorum (Fisch. ex Link) Schult., durante 2007, 2008, y 2009 y se le dio seguimiento a las semillas sembradas así como el destino de las plántulas. A través de los años de estudio y especies se vio que la germinación fue rápida y alta, con la obtención de la germinación del 50% en diciembre, menos de 2 meses después de la siembra de especies. La aparición de semillas germinadas no ocurrió hasta finales de febrero para A. desertorum y en marzo para las dos especies de pastos nativos. En 2007 la mayoría de aparición y muerte estaba limitada a varias semanas mientras que en 2008 y 2009 el surgimiento y la muerte se distribuyeron a través de varios meses. El tiempo de aparición de las plántulas no influyó en la probabilidad de la supervivencia o al potencial de agua de la planta al mediodía o (probabilidad de superación < 0.05). Las probabilidades de supervivencia una vez que surgieron las plántulas fueron mayores para las especies nativas (0.71) que A. desertorum (0.51) en dos de los tres años de estudio (probabilidad de superación > 0.98). La germinación temprana de gramíneas después de que cae la semilla y el periodo tardo de dos a tres meses la semilla germinada permanece en el suelo antes de emerger y apoya la hipótesis que el reclutamiento de plántulas puede estar altamente limitado por el proceso ecológico y las condiciones durante el invierno o el inicio de la primavera tales (como la descongelación del suelo, los patógenos de la semilla, o las costras físicas). Retrasando la siembra a principios del invierno o primavera y usando otras herramientas de manejo que mitiguen los factores que impulsan este cuello de botella se puede mejorar considerablemente los resultados de la restauración en estos sistemas