Belowground response of prairie restoration and resiliency to drought

Agriculture, Ecosystems and Environment 266 (2018) 122–132

Contents lists available at ScienceDirect

Agriculture, Ecosystems and Environment journal homepage: www.elsevier.com/locate/agee

Belowground response of prairie restoration and resiliency to drought a

b

Racheal N. Upton , Elizabeth M. Bach , Kirsten S. Hofmockel a b c

a,c,⁎

T

Department of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, IA, 50011, United States School of Global Environmental Sustainability, Department of Biology, Colorado State University, Fort Collins, CO, 80523, United States Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, Richland, WA, 99354, United States

A R T I C LE I N FO

A B S T R A C T

Keywords: bacteria fungi microbial ecology prairie restoration soil tallgrass prairie drought resiliency

Agricultural land use is a major threat to biodiversity and ecosystem functions in tallgrass prairies. However, there are proposed bioenergy systems that can use biomass harvested from restored tallgrass prairie, creating a potential free market incentive for landowners to restore prairies. These alternative management practices may alter associated soil microbial communities and their ecosystem services. We examined changes in soil microbial community structure, function, and resiliency to drought following two prairie restorations from row-crop agriculture and through subsequent succession in a fertilized and unfertilized tallgrass prairie. The soil microbial community structure was assessed through amplicon (16S and ITS) sequencing, function through potential extracellular enzyme activity, and resiliency indices were calculated for both microbial diversity measures and extracellular enzyme activity. We hypothesized that 1) distinct soil microbial communities in each management system will continue to develop over time reflecting the extent of divergence between the plant communities, due to the strong selective forces plant communities have on the soil microbiome. 2) Microbial extracellular enzymatic function will continue to diverge between the management systems across sampling years. 3) We will see increased resiliency to drought in the prairies potentially due to the greater diversity in this management system for the microbial and plant community, creating a possible enhancement in functional redundancy. Our experiment demonstrates that soil microbial communities continue to diverge from row-crop agriculture as prairie restoration progresses. Planted prairie bioenergy systems with higher plant diversity supported greater microbial diversity than corn systems. Corn monocultures were less resistant to drought stress, as evidenced by decreased microbial activity and richness. Prairies with increased microbial diversity exhibited increased functional resiliency than corn systems, as measured by cellulose-degrading enzyme activity. Prairies that received nitrogen fertilization maintained high microbial diversity and activity, even under drought. Our study demonstrates that diverse cropping systems may benefit from nitrogen fertilization to confer resiliency to disturbance events. Increasing resiliency, while maintaining productivity, is key to managing alternative crops that are sustainable systems for biofuel uses. Our multi-year study reveals the benefits of long-term experiments for capturing the dynamic range of microbial mediation of soil carbon and nutrients and the importance of resiliency in both developing sustainable management systems and modeling predictive biogeochemical models.

1. Introduction Agricultural land use is a major threat to biodiversity and ecosystem functions such as regulating climate and maintaining water quality (Tilman et al., 2002). In the tallgrass prairie region of North America, conversion to row-crop agriculture has reduced the tallgrass prairie ecosystem by 86% (Samson and Knopf, 1996; Samson et al., 2004), making tallgrass prairie among the most “in crisis” ecosystems in the world (Hoekstra et al., 2005; Wright and Wimberly, 2013). In the past decade, there has been increased pressure to produce bioenergy sources from agricultural systems, continuing to threaten the remaining ⁎

tallgrass prairies (Börjesson and Tufvesson, 2011; Fargione et al., 2008; Searchinger et al., 2008; Wright and Wimberly, 2013). However, there are proposed bioenergy systems that can use biomass harvested from restored tallgrass prairie, creating a potential free market incentive for land owners to restore prairie, particularly on marginal lands (Gelfand et al., 2013; Jarchow and Liebman, 2012; Tilman et al., 2006). These alternative management system practices are aimed to maximize aboveground productivity, while reducing the detrimental impacts of agriculture on the ecosystem as a whole (Tilman et al., 2002; Tscharntke et al., 2005; Turner et al., 2007). Investigation into how to successfully implement agricultural conversion to systems that support

Corresponding author at: Department of Ecology, Evolution, and Organismal Biology, Iowa State University, Ames, IA, 50011, United States. E-mail address: [email protected] (K.S. Hofmockel).

https://doi.org/10.1016/j.agee.2018.07.021 Received 1 April 2018; Received in revised form 13 July 2018; Accepted 25 July 2018 0167-8809/ © 2018 Published by Elsevier B.V.

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Putten, 2010). In ecosystems managed for multiple ecosystem services, such as bioenergy feedstock systems, there is a tremendous need to evaluate how different management practices impact microbial resilience to current and predicted climatic events. Assessment of management practices like N fertilization on fungal and bacterial communities as well as their enzymatic activities is crucial to determine the long-term sustainability of these systems. Using a microbial-focused approach, this study examined the longterm impact of N fertilization and management system (corn, prairie, and N fertilized prairie) on the microbial community structure and enzymatic function of restored grasslands used for biofuel production. Additionally, due to a natural drought occurrence, we were able to assess potential resiliency of these management systems within the microbial community and their enzymatic activity as well as the natural recovery afterwards. The Comparison of Biofuel Systems (COBS) field site is aimed at investigating potential biofuel cropping systems to increase plant diversity, sustainability, and maintain productivity. For this study, we evaluated three differing management systems: corn monoculture and restored native prairie with and without N fertilization. We seek to understand how management practices impacted fungal and bacterial community compositions, function, extracellular enzymes, and resiliency to drought across time in response to both the shifting plant communities and abiotic factors. We hypothesize that 1) distinct soil microbial communities in each management system will continue to develop over time reflecting the extent of divergence between the plant communities, due to the strong selective forces plant communities have on the soil microbiome. 2) Microbial extracellular enzymatic function will continue to diverge between the management systems across sampling years. 3) We will see increased resiliency to drought in the prairies due to the potentially greater diversity in this management system for the microbial and plant community, creating a potential enhancement in functional redundancy. To further inform sustainability efforts, the potential main drivers, plant community and N fertilization, and environmental cues of microbial community changes and function were studied across a 4-year sampling period.

more biodiversity, increase carbon (C) storage, and promote nutrient retention is vital to learning how to manage biofuel agroecosystems for increased sustainability (Jarchow and Liebman, 2012; Kim et al., 2012). In addition, understanding the biogeochemistry occurring within biofuel agroecosystems could advance opportunities to support biofuel production on marginal lands; which traditionally have low C and nutrient storage and therefore could provide multiple benefits if converted to reconstructed prairie and appropriately managed to maximize ecosystem services (Gelfand et al., 2013; Isbell et al., 2015; Schulte et al., 2017). Diverse prairies are being reconstructed on traditional agricultural fields to expand upon potential sustainable biofuel systems (Borsari et al., 2009; Jarchow and Liebman, 2012). During restoration from rowcrop agriculture to diverse prairie communities, the abundance and chemistry of plant inputs to soil will change, reflecting plant community composition, phenology, and response to environmental cues (Anderson-Teixeira et al., 2009; De Deyn et al., 2008). For example, compared to diversified systems, traditional continuous corn agroecosystems have less abundant C inputs into the soil, reflected in lower root biomass and less diverse rhizodeposition (Allmaras et al., 2004; Collins et al., 1999; Dietzel et al., 2015; Dietzel et al., 2017). These changes in plant community ecology may affect soil microbial community composition, with potential cascading effects on C and nitrogen (N) cycling, which have important implications for the long-term fertility and productivity of corn biofuel systems (Barber et al., 2017; Klopf et al., 2017; Landis et al., 2008; McBride et al., 2011; Wieder et al., 2013). Tallgrass prairie restoration for biofuel services increases root biomass, soil C pools, and microbial biomass (Baer et al., 2010; Klopf et al., 2017); however, it is unclear how plant-microbe community interactions will change across time in restorations managed for biofuel. Soil microbial communities are highly dynamic in membership (Buckley and Schmidt, 2003; Kuzyakov and Blagodatskaya, 2015). Therefore, understanding the soil microbiome response to prairie restoration requires multi-year evaluations relative to traditional agricultural systems. Changes in microbial community can lead to fluctuations in microbial C and N cycling activity and shape the local environment. Specifically, community level changes in extracellular enzymatic activity can lead to a feedback loop between the soil’s physical properties, plant community, and microbial community (Braissant et al., 2003; Burns et al., 2013; Sasse et al., 2017). How rapidly microbial communities and their extracellular enzymes respond to this changing plant community, as well as the implications for coupled C-N cycling, needs to be further investigated to implement biofuel production that maximize ecological benefits (Averett et al., 2004). Shifts in the plant communities in response to N fertilization can impact microbial community structure and diversity, with implications for microbial activity and function. The short-term effect of N fertilization is an increase in plant diversity and productivity, although this can change across time (Jarchow and Liebman, 2013). The long-term effect of N fertilization is a decrease in native plant diversity, as fastgrowing plants, both exotic and native, arise in the community (FloresMoreno et al., 2016; Harpole and Stevens, 2016; Morgan et al., 2016). The impacts of N fertilization on microbial communities are highly context dependent and often mediated by the response of the aboveground community to the N inputs (Leff et al., 2015; Kneller et al., 2018). Nitrogen fertilization can also increase the sensitivity of the plant community to disturbance events, generating a compounded effect to potentially alter the plant community (Collins et al., 2017; Tognetti and Chaneton, 2015). Response to disturbance events or the resiliency of the community is often evaluated based on plant communities without consideration of how microbial community composition and function respond (Lau and Lennon, 2012; Sheik et al., 2011). Resiliency within the microbial community under natural disturbances has been investigated, and yet we know little of how microbial community changes affect belowground C and N cycling (Hawkes et al., 2005; van der

2. Methods 2.1. Field Site Description and Soil Sampling Process All samples were collected at the Comparison of Biofuel Systems (COBS) field study site (Boone County, IA). The COBS research site was established in 2008; prior, the site was used for corn-soybean rotations (Jarchow and Liebman, 2013). The COBS research site is a randomized complete block design comparing six bioenergy systems; plots measured 27 × 61 m2. This study investigated three of these systems: continuous corn (maize; grain and 50% silage harvest for biomass), planted tallgrass prairie (∼75% aboveground biomass harvest), and fertilized tall-grass prairie (∼75% aboveground biomass harvest). Both tallgrass prairie systems were seeded with a mixture of 31 native genotype prairie species, in May 2008 (Jarchow and Liebman, 2013). The first fertilization of the fertilized prairie management system occurred on April 17th, 2009 and was applied as ammonium nitrate; following all annual nitrogen fertilization for all management systems was applied as urea-ammonium nitrate. Annually, the fertilized prairies received 84 kg N ha-1yr-1 in late March to mid April, and continuous corn systems received N inputs in accordance with spring nitrate tests (∼168 kg N ha-1 yr-1) applied immediately prior to corn planting (Jarchow and Liebman, 2012). The unfertilized prairie (referred to as the ‘prairie’ management in this study) did not receive any fertilizer inputs of any kind. The aboveground biomass is harvested to a height of 15-20 cm across all management systems annually in early November. Detailed COBS research site descriptions can be found in Jarchow and Liebman (2013). Soil samples were collected during peak growing season midAugust in 2011, 2012, and 2014. Soil cores were collected from the top 10 cm of soil using a 5 cm diameter slide-hammer coring device 123

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

chimeric OTUs using USEARCH (Quast et al., 2012). All bacterial samples were rarefied to 20,715 reads per sample.

(Giddings Machine Company, Windsor, CO). Four cores were collected sterilely from each plot (4 plots per treatment) from the continuous corn, prairie, and fertilized prairie systems and stored at 4 °C during transportation to the laboratory. Soil cores were placed through an 8 mm sieve by breaking soil along natural points of weakness; whole soil subsamples were stored at −20 °C and −80 °C.

2.5. Carbon and Nitrogen Enzyme Assays Microbial soil bulk enzyme potential activity was measured using a fluorometric enzyme assay (DeForest, 2009; German et al., 2011; SaiyaCork et al., 2002; Turner, 2010) modified for high-throughput and high repeatability (Hargreaves and Hofmockel, 2015). For each sample, 1 g of frozen soil (−20 °C) was resuspended in 125 mL of 50 mM sodium acetate buffer, pH 5. Soil slurry samples were placed into 5 mL tubes with labeled substrates and incubated for 2 hours in the dark at 25 °C on 140 rpm shaker. Substrates used in this study, all with a final concentration of 400 mM, included N-acetyl-glucosaminide to measure Nacetyl- β-D-glucosaminidase (NAG), β-D-glucopyranoside to measure βglucosidase (BG), β-D-xylopyranoside to measure β-xylosidase (BX), βD-cellobioside to measure cellobiohydrolase (CB), and phosphate to measure acid phosphatase (AP; Table S1) all labeled with 4-methylumbelliferyl (MUB) (Bach and Hofmockel, 2015; Hargreaves and Hofmockel, 2015). Following incubation, soil slurry with labeled substrate was transferred to 96 deep well black micro-plates for optical density determination. The optical density was read at 450 nm on a microplate reader (BioTek, Winooski, VT). For all samples and substrates the absolute potential enzyme activity (nmol h-1 g-1 dry soil) was calculated using the equations from German et al. (2011).

2.2. Climatic Data Collection and Soil Water Content Climatic data was collected from National Center for Environmental Information, (National Oceanic and Atmospheric Administration) including average maximum and mean daily temperature, total growing degree days, total heating degree days, total cooling degree days, and total precipitation for each sampling growing season for each individual year, which was defined as May 1st to August 20th for the purposes of this study. Soil gravimetric water content was determined for each management system by subsampling 10 g of fresh, wet soil and drying at 100 °C overnight, and followed by using the difference in dry and wet weight for determining water content. 2.3. DNA Extraction DNA was extracted using 0.25 g of sub-sample soil using the PowerSoil DNA Isolation Kit (Mo-Bio, Carlsbad, CA). Extracted DNA was quantified through nanodrop and PicoGreen fluorometry and subsamples were sent off for amplicon based sequencing (ITS and 16S) analysis at Argonne National Laboratory (Lemont, IL) on dry ice. Staff at ANL-NGS core facility performed all downstream processes, including amplification, library preparation, and sequencing. Fungal internal transcribed spacers (ITS) region 1 was amplified using modified ITS1F (5’ CTTGGTCATTTAGAGGAAGTAA 3’) and ITS2 (5’GCTGCGTT CTTCATCGATGC 3’) primer set (Smith and Peay, 2014). Bacterial 16S gene, region V4, was amplified using 515f/806 r standard primer set. Both bacterial and fungal amplicons for ITS1 (pair-ended: 250 bp X 250 bp) and 16S (pair-ended: 150bp x 150bp) were sequenced using Illumina MiSeq 500-cycle kit at the Next Generation Sequencing Core with the Illumina MiSeq sequencing system on separate runs.

2.6. Statistical Analysis All statistical analyses were performed in R Studio version 1.0.136 (R Core Team 2016). Using ‘vegan’ package, fisher’s alpha diversity index was used to determine richness from the relative species abundance of the fungal and bacterial communities (Oksanen et al., 2007; Oksanen et al., 2013). Main effects tested on fungal and bacterial richness, evenness, and Shannon’s diversity index including management system (prairie, fertilized prairie, and corn) and sampling year (2011, 2012, and 2014) using a mixed model ANOVA (α = 0.05), block was used as random factor. A Tukey’s Honest Significant Difference (Tukey’s HSD) post-hoc test was performed on the same mixed model ANOVA to determine specific difference between treatment groups (Miller, 1981; Yandell, 1997). All main effects have been tested for significant interactions. A PERMANOVA analysis was performed for OTU abundance using a two-way factorial design (management system, sampling year, and management system and sampling year interaction) with 999 permutations using the ‘ADONIS’ function in the ‘Vegan’ package to evaluate to determine if sampling year and management system were main effects for the fungal and bacterial communities separately. Additionally, to determine if there were significant relationships between community composition and environmental factors, the Eigenvalues created from OTU abundance variability for each treatment were correlated to each environmental factor individually: soil gravimetric water content (Table S3), average maximum temperature, mean temperature, growing degree days, heating degree days, cooling degree days, precipitation (Table S2), root biomass, and mean soil carbon (Dietzel et al. 2015, 2017) using the ‘envfit’ function. Factors that were significant on fungal and bacterial community diversity were analyzed for species indicator analysis using the ‘Indicspecies’ package to determine which groups of the fungal and bacterial community were distinct under each circumstance (De Cáceres and Jansen, 2012;De Cáceres, 2013). To evaluate extracellular enzyme activity, all enzyme activity measurements were log2 transformed to meet normality restrictions. A mixed model ANOVA (α = 0.05) was used between management system and sampling year on each individual enzyme separately. Individual sampling years were analyzed to evaluate trends across management system. A Tukey’s HSD was performed on the same mixed

2.4. ITS and 16S Sequencing Analysis Sequencing for both bacteria and fungi were analyzed using methods from Caporaso et al. (2010) using the open source pipeline ‘hundo’ (https://github.com/pnnl/hundo). Raw ITS (fungal) and 16S (bacterial) sequence reads were de-multiplexed with using EA-Utils (Aronesty, 2013) with zero mismatches allowed in the barcode sequence. Reads were quality filtered with BBDuk2 (Bushnell, 2014) to remove adapter sequences and PhiX with matching kmer length of 31 bp at a hamming distance of 1. Reads shorter than 51 bp were discarded. Reads were merged using USEARCH (Edgar, 2010) with a minimum length threshold of 175 bp and maximum error rate of 1%. Sequences were de-replicated (minimum sequence abundance of 2) and clustered using the distance-based, greedy clustering method of USEARCH (Edgar, 2013) at 97% pairwise sequence identity among operational taxonomic unit (OTU) member sequences. De novo prediction of chimeric sequences was performed using USEARCH during clustering. Fungal taxonomy was assigned to OTU sequences using BLAST (Camacho et al., 2009) alignments followed by least common ancestor assignments across UNITE version 7 database (Kõljalg et al., 2013). OTU seed sequences were filtered against UNITE version 7 database (Kõljalg et al., 2013) to identify chimeric OTUs using USEARCH. Following sequencing pipeline, all fungal samples were rarefied to 12,949 reads per sample. Bacterial taxonomy was assigned to OTU sequences using BLAST alignments followed by least common ancestor assignments across SILVA database version 123 clustered at 99% (Camacho et al., 2009; Quast et al., 2012). OTU seed sequences were filtered against SILVA database version 123 clustered at 99% to identify 124

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

sampling year accounted for approximately 31.7% of the total variance in bacterial community structure (Fig. 2). Similarly to the fungal community, only root biomass and soil carbon were found to have a significant relationship with bacterial OTU abundance (roots: P < 0.001, R2 = 0.57; soil C: P < 0.001, R2 = 0.67; Fig. 2).

ANOVA model to determine the specific differences between sampling years and management systems. To assess if there were significant correlative dynamics between enzymatic activity and environmental factors (soil gravimetric water content (Table S3), average maximum temperature, mean temperature, growing degree days, heating degree days, cooling degree days, precipitation (Table S2), root biomass, and mean soil carbon (Dietzel et al. 2015, 2017), a Pearson’s correlation was performed.

3.3. Management System Impacts

Mean precipitation was 434 mm in 2011, 258 mm in 2012, and 540 mm in 2014. The precipitation in 2012 was less than 50% of the 60yr mean annual precipitation. Additionally, the average daily mean and maximum temperature was higher than either the 2011 or 2014 growing seasons (Table S2). Soil gravimetric water content was lower in 2012 during the drought then either 2011 or 2014, with only 11-13% water content comparatively to 15-19% (Table S3). This drought allowed us to examine management effects on microbial resilience, under conditions that incorporate the natural duration and intensiveness of drought, including all other compounding factors, such as increased temperature and loss of productivity.

Our results show management system is a main effect for fungal community structure and diversity, with both prairies diverging from the corn to be more diverse (Table S5a). Management system was a significant main effect on fungal richness (F2,33 = 8.4; P = 0.001) and Shannon’s diversity index (F2,32 = 4.1; P = 0.03), but not on evenness (Fig. 3, Table S5a). Fungal richness was over 30% greater in the two prairie ecosystems than in the continuous corn (Fig. 3). Fungal community structure was significantly changed under different management systems (Table S5b). Both the relative abundance and normalized abundance of Ascomycota phylum was significantly affected by management system (F2,34 = 12.5; P = 3.8e-6) with no interaction with sampling year (Table S5b). Ascomycota phylum was most dominant in the continuous corn across years, and was lowest across sampling years in the unfertilized prairie. Glomeromycota were almost absent in the corn plots across all years, and had a significant increase in both prairie systems (F2,34 = 21.3; P < 0.0001; Table S5b), but note ITS primers are often biased against Glomeromycota (Lindahl et al., 2013). Basidiomycota and Zygomycota abundance was not significantly impacted by management system (Table S5b). Consistent with the fungal response, bacterial communities in prairies also diverged from the corn community to have greater diversity; however, the fertilized prairie was intermediate between the corn and unfertilized prairie management systems. Shannon’s diversity index was the only diversity measurement for the bacterial community that was moderately affected by management system (F2,31 = 3.1; P = 0.06; Fig. 4). In a direct two-way ANOVA by management system only prairie and continuous corn differed based on bacterial richness and Shannon’s diversity index (F1,20 = 5.5; P = 0.03, F1,20 = 4.8; P = 0.04, Table S5c). Individual bacterial phyla abundance did not generally show a strong relationship to management system. The exception is Acidobacteria, which was most abundant in the corn, followed by the unfertilized prairie then the fertilized prairie (F = 5.1; P = 0.01; Table S5d).

3.2. Microbial Community Composition

3.4. Time Effects

ITS1 sequencing provided 7,950 unique fungal OTUs. Fungal communities across all samples were dominated by the Ascomycota phylum, comprising over 40% of identified reads, followed by the Zygomycota, as defined by Benny et al. (2014), 33%, and Basidiomycota, 21% (Table S4). Our study demonstrates that management system and sampling year significantly impact fungal communities. Total fungal OTU community structure is significantly impacted by both management system and sampling year with no interaction (PERMANOVA, F2,34 = 4.6; P = 0.001, F2,34 = 1.5; P = 0.03). Bray-Curtis dissimilarity matrix resulted in approximately 30.9% of the total variance in the fungal community explained by management system and sampling year (Fig. 1). Environmental and ecological factors were tested to determine if they had a strong correlative relationship between the fungal OTU abundance; only root biomass and soil carbon significantly affected fungal community composition (roots: P < 0.001, R2 = 0.67; soil C: P < 0.001, R2 = 0.63; Fig. 1). 16S V4 sequencing provided 16,872 unique bacterial OTUs. Acidobacteria, Actinobacteria, Proteobacteria, Bacteroidetes, and Verrucomicrobia were the dominant bacterial phyla (Table S4). Bacterial community structure was significantly impacted by both management system and sampling year, with no interaction (PERMANOVA, F2,34 = 2; P = 0.001, F2,34 = 3.2; P = 0.01). Management system and

Time impacted fungal OTU abundance and diversity to a lesser extent than management system. The natural drought in 2012 reduced fungal richness and Shannon’s diversity relative to both other sampling years (F2,20 = 5.7; P = 0.008, F2,20 = 3.1; P = 0.06), and evenness did not differ (Table S5a). Sampling year significantly impacted the most abundant fungal phyla, with system-specific responses. In both prairie ecosystems relative and normalized abundance of Basidiomycota increased across sampling years concurrent with shifts in plant communities and independent of drought; however, in the corn system the relative abundance remained consistent across all sampling years (Table S5b). Individual orders of the Basidiomycota phylum were only found in the later sampling years of the study in the prairies, such as Russulales, Corticiales, and Phallales. Glomeromycota were almost absent in corn across all years. In both prairies, the relative abundance of Glomeromycota increased across sampling years, starting at 2.7 to 3.9% in 2011 and increasing to 5.7 to 6.4% in 2014 (Table S5b). Normalized abundance of Glomeromycota followed the same trends as relative abundance across management systems and sampling years. Bacterial diversity measurements were more strongly impacted by sampling year than management system, due to the strong effects of the drought, 2012 (Table S5c). Sampling year had a significant impact on bacterial richness, evenness, and Shannon’s diversity index (F2,31 = 7.2;

2.7. Resiliency Indices Resiliency was calculated in response to the 2012 drought for extracellular enzyme activity and bacterial and fungal richness. Resiliency was determined using the De Vries et al. formula (2012) and Orwin and Wardle (2004): Resiliency Index (tx) = (2|D0|)/(|D0|+|Dx|)-1, where D0 = Prior to the disturbance event and Dx = the absolute difference between the initial measurement and after the disturbance (De Vries and Shade, 2013; Orwin and Wardle, 2004). Resiliency Index values near 1 would indicate a near exact return to pre-disturbance value. A one-way ANOVA (α = 0.05) was used to determine if management system was a main effect on the resiliency indices with log2 transformation when necessary to meet normality requirements. 3. Results 3.1. Climatic Variables and Soil Water Content

125

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Fig. 1. Non-metric multidimensional scaling (NMDS) of fungal community OTU abundance using a Bray-Curtis dissimilarity matrix. A) Environmental factors root biomass and soil C were determined to be significantly correlated to fungal community composition (roots: P < 0.001, R2 = 0.67; soil C: P < 0.001, R2 = 0.63). (A) Management system (CC = continuous corn; FP = fertilized prairie; P = prairie) and sampling year (B) were both determined significant on fungal community structure using a PERMANOVA analysis (P < 0.05).

Fig. 2. Non-metric multidimensional scaling (NMDS) of bacterial community OTU abundance using a Bray-Curtis dissimilarity matrix (CC = continuous corn; FP = fertilized prairie; P = prairie). A) Environmental factors root biomass and soil C were determined to be significantly correlated to bacterial community composition (roots: P < 0.001, R2 = 0.57; soil C: P < 0.001, R2 = 0.67). (A) Management system (CC = continuous corn; FP = fertilized prairie; P = prairie) and sampling year (B) were both determined significant on fungal community structure using a PERMANOVA analysis (P < 0.05).

2014 (2011: P = 4.9e-4, P < 0.0001, 2014: P < 0.0001, P = 0.005). Investigating individual management systems showed sampling year had a significant impact on bacterial richness in continuous corn (F2,9 = 5.1; P = 0.02) and a moderately significant impact in the unfertilized prairie (F2,9 = 2.6; P = 0.09). Bacterial communities were even more sensitive to inter-annual variation than fungal communities. Significant decreases occurred during the 2012 drought in 4 of 5 most abundant phyla: Acidobacteria, Actinobacteria, Proteobacteria, and Verrucomicrobia (Table S5d). Among the most abundant bacterial phyla, only Bacteroidetes increased in

P = 0.003, F2,31 = 10; P = 0.0004, F2,31 = 13.2; P < 0.0001; Fig. 4; Table S5c). Independent of management system, water availability strongly shaped the bacterial community response. Relative to 2011, bacterial richness decreased on average 7% under drought (2012) and increased 6% with increased precipitation (2014) (P = 0.003) across all management systems. All management systems responded in the same direction for bacterial richness in each sampling year, however, the magnitude of the response is dependent on the management practices (Fig. 4). Shannon’s diversity index and evenness were significantly lower for drought year 2012 compared to sampling years 2011 and 126

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Fig. 3. Mean fungal community richness, evenness, and Shannon’s Diversity Index across management systems (CC = continuous corn; FP = fertilized prairie; P = prairie) and sampling year treatments (error bars are +/- 1SE).

3.5. Interaction of Time and Management System on Indicator Family Analyses

abundance across all management systems during drought (F2,34 = 8.2; P = 0.002), and Acidobacteria increased in relative abundance in the fertilized prairie. The relative abundance of Proteobacteria was significantly impacted by sampling year (F2,34 = 18.6; P < 0.0001) and had a significant interaction with management system (F4,32 = 3.5; P = 0.02; Table S5d). All management systems had a significant decrease in relative abundance of Proteobacteria in 2012, with the largest decrease occurring in the corn system. Acidobacteria relative abundance was significantly affected by year (F2,34 = 4.3; P = 0.02) and moderately by the interaction with management system (F4,32 = 2.3; P = 0.07; Table S5d). Acidobacteria peaked in 2012 during drought and decreased in 2014 in both the corn and unfertilized prairie, but the fertilized prairie community was stable despite environmental changes. Across all management systems relative abundance of Verrucomicrobia decreased with time and was significantly different under wet conditions in 2014 compared to previous sampling years (F2,34 = 15.6; P = 4.5e-5; Table S5d). Actinobacteria decreased in relative abundance in 2012 across all management systems. Only sampling year was significant for impacting the relative abundance of Actinobacteria (F2,34 = 14.6; P < 0.0001; Table S5d).

Indicator taxa analysis reveals which microbial taxa, in this case families, are distinct of specific treatment conditions, including significant interactions between management systems and sampling year. The number of indicator families showed a statistical interaction between management systems and sampling years. The corn plantings generally had the fewest distinct fungal families, with the exception of 2012 during the drought, where it had the highest number of indicator fungal families. The unfertilized prairie had the largest change in indicator fungal families during the drought year, with marked reductions resulting in no distinct fungal families being found in 2012. The fertilized prairie had a moderate number of indicator fungal families in all sampling years (Fig. 5). Bacterial family indicator analysis showed a high number of repeated, but less diverse, bacterial families in the corn across years. Generally, the fertilized prairie had the highest number of indicative bacterial families, the exception being 2012 during the drought, when the unfertilized prairie had the highest number of indicative bacterial families (Fig. 5).

Fig. 4. Mean bacterial community richness, evenness, and Shannon’s Diversity Index across management systems (CC = continuous corn; FP = fertilized prairie; P = prairie) and sampling year treatments (error bars are +/- 1 SE). 127

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Fig. 5. Indicator fungal and bacterial families by sampling year and management system (a) corn, (b) fertilized prairie, and (c) prairie (α = 0.05).

3.7. Resiliency Measurements

3.6. Interaction of Time and Management System on Extracellular Enzyme Activity

Resiliency indices show the amount of change in either microbial richness or extracellular enzyme activity during the drought event, in each management system. Analyzing resiliency indices by management system reveal the capacity of the aboveground practices to increase microbial resilience to drought, both directly and indirectly. In general, resiliency indices did not show any difference when investigating richness measurements, but did show differences in enzymatic activity between the corn and prairie management systems. Bacterial and fungal richness resiliency indices were unaffected by management system, despite varying degrees of change in the fungal and bacterial richness measurements during the drought event (Table 2). Unfertilized prairie supported greater resiliency indices of BX activity over corn (F2,8 = 4.4; P = 0.05), indicating stronger return to pre-disturbance levels. Fertilized prairie had moderately greater CB resiliency indices then all other management systems (F2,9 = 3.5; P = 0.08). AP activity was greater in the drought year than in the pre-drought measure for all systems (Table 1), and continued to increase in the post-disturbance measure for both prairie systems, yielding negative resilience indices. AP activity in corn systems decreases post-disturbance, yielding a positive resiliency index. (F2,9 = 3.3; P = 0.08; Table 2).

Generally, extracellular enzyme activity showed an increase across sampling years, suggesting an increase in certain decomposition functions in the microbial community across time (Table S5e). C-cycling enzymes, BG and CB varied significantly by sampling year, with decreased activity in 2012 during the drought (F2,30 = 3.9; P = 0.03; F2,30 = 16.8; P = 1.3e-5), but activity was not affected by management system. BX activity in the corn reached peak activity during the wettest year in 2014 (F2,9 = 16; P = 0.001). Sampling year did not impact the fertilized prairie mean potential activity and was slightly higher than corn mean activity (P = 0.08). All 3 C-cycling enzymes activities significantly correlated with growing (GDD) and heating degree days (HDD); with decreased enzymatic activity with greater GDD and HDD (Table S6). BX and CB were also significantly negatively correlated with mean and maximum temperature, and were positively correlated with precipitation (Table S6). CB was the only C-cycling enzyme to have a significantly positive relationship with gravimetric water content. For C-N-cycling NAG, 2012 had a decrease in activity from both 2011 and 2014 in both the corn and unfertilized prairie (F2,29 = 5.2; P = 0.012) with sustained enzymatic potential in the fertilized prairie, despite differences in precipitation regimes. NAG was negatively correlated with all temperature measurements, mean temperature, maximum temperature, GDD, HDD, and positively correlated with precipitation (Table S6). Phosphatase activity, AP, also followed the similar patterns of activity as other enzymes, with increased activity across the experimental sampling time points (F2,30 = 30.1; P < 0.0001), the only difference being that all management systems had an increase in activity in each progressive sampling year (Table 1). AP was negatively correlated to gravimetric water content, GDD, and HDD (Table S6). During drought in 2012, fertilized prairie had significantly higher activity than any other management system for all enzymes (Table 1).

4. Discussion Our study found that planted tallgrass prairies can increase microbial community diversity and support more resilient ecosystem functions like cellulose degradation. Our results show management system had a strong influence on soil microbial community structure, with both diverse prairie plantings supporting higher fungal diversity compared to the corn monoculture and unfertilized prairie supporting the greatest bacterial diversity. Increased microbial diversity corresponded with increased plant diversity, root inputs, and soil carbon. Both fungal and bacterial communities are central to biogeochemical cycling and our 128

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Table 1 Mean potential extracellular enzyme activity (nmol h-1 g-1 dry soil) in each management systems by sampling years (+1SE). Enzymes measured included carbon cycling enzymes β-glucosidase (BG), β-xylosidase (BX), and cellobiohydrolase (CB), carbon and nitrogen cycling enzyme N-acetyl-glucosaminidase (NAG), and phosphate cycling enzyme acid phosphatase (AP). Enzyme

Management System

2011

2012

2014

BG

Corn Fertilized Prairie Corn Fertilized Prairie Corn Fertilized Prairie Corn Fertilized Prairie Corn Fertilized Prairie

331.39 ± 64.96 504.61 ± 208.35 318.77 ± 93.71 58.12 ± 13.03 115.55 ± 57.74 61.75 ± 21.96 89.84 ± 31.71 126.57 ± 60.27 95.23 ± 51.56 164.43 ± 101.99 126.44 ± 42.37 109.35 ± 30.61 1007.78 ± 64.37 904.18 ± 388.06 692.37 ± 167.84

305.36 ± 77.71 989.53 ± 323.61 357.2 ± 120.8 31.45 ± 8.37 139.59 ± 41.61 57.88 ± 21.51 13.07 ± 4.23 56.56 ± 15.62 17.15 ± 5.5 48.47 ± 10.78 159.32 ± 36.39 104.8 ± 32.02 1835.19 ± 376.33 4819.79 ± 1294.15 2667.75 ± 1012.18

894.83 ± 255.93 538.81 ± 106.56 845.43 ± 371.13 182.85 ± 43.6 116.16 ± 22.11 152.17 ± 66.97 137.52 ± 28.54 58.65 ± 6.76 88.76 ± 38.13 313.24 ± 119.51 130.36 210.11 ± 65.98 7003.7 ± 1294.01 3126.03 ± 469.33 4847.54 ± 1651.52

BX

CB

NAG

AP

Prairie

Prairie

Prairie

Prairie

Prairie

from the diverse plant communities. The prairies also had a significantly higher abundance of Glomeromycota, arbuscular mycorrhizal fungi, suggesting these beneficial interactions are occurring more frequently within the diversified prairies than within the corn plots (Allan, 2017; Van der Heijden et al., 1998). In addition to changes seen in major fungal phyla, we also see trends with indicator microbial families, with the prairies having generally higher numbers of indicative microbial families except for during the drought (2012). During drought, the fertilized prairie and unfertilized prairie diverged from each other for indicative microbial families, with the fertilized prairie greatly reduced in both bacterial and fungal families and the unfertilized prairie being solely defined by a large number of bacterial families. This demonstrates the dynamic interactions of aboveground management practices and environmental factors, precipitation, temperature, etc. that shape the soil microbial community simultaneously. Studying these interactions between management and environmental conditions over time will allow for greater understanding of these dynamic interactions and inform future management decisions in preparation of disturbance events, i.e. drought, etc. Findings in the two prairies are demonstrating that these feedstock systems potentially support more diverse and dynamic microbial communities. Bacteria were more sensitive to drought and environmental changes across sampling years than fungal communities, causing bacterial communities to be dominantly shaped by inter-annual variability. All dominant bacterial phyla decreased in abundance during the drought with the exception of Bacteroidetes. Previous studies investigating bacterial response to drought have shown that Bacteroidetes can increase or decrease in the microbial population depending on the system or soil type (Acosta-Martinez et al., 2014; Chodak et al., 2015). This increase of Bacteroidetes across management systems may signify a drought resistant bacterial group shaped strongly by the soil environment. Indicator taxa analysis also showed changes in response to drought conditions, with the unfertilized prairie being solely defined by bacterial communities. Yet, similar to the fungal taxa analysis, the two prairie systems had a greater amount of diversity in their indicator taxa, but they were highly dynamic, compared to the more consistent, but less diverse corn signature microbial community. Differences in microbial community composition and diversity, overall and in response to drought, corresponded with differences in ecosystem functions such as cellulose decomposition. Generally, potential enzyme activity did not differ between management systems in non-drought years, although there was a trend toward increased EEA in the prairie systems. Enzyme activity responded similarly across time in all systems, suggesting the local edaphic and environmental factors were driving extracellular enzyme pools in soil. EEAs had decreased activity with increased growing and heating degree days, an

Table 2 Mean resiliency indices ( ± 1SE) for bacterial and fungal richness and extracellular enzyme activity. Enzymes measured were carbon cycling enzymes βglucosidase (BG), β-xylosidase (BX), and cellobiohydrolase (CB), carbon and nitrogen cycling enzyme N-acetyl-glucosaminidase (NAG), and phosphate cycling acid phosphatase (AP). Statistically different means among management systems are denoted with different letter superscripts (α=0.05). Measurement

Corn

Fertilized Prairie

Prairie

Fungal Richness Bacterial Richness BG Activity BX Activity CB Activity NAG Activity AP Activity

0.8 ± 0.05 0.8 ± .06 0.6 ± 0.1 0.4 ± 0.1B 0.08 ± 0.02B 0.4 ± 0.2 0.2 ± 0.2A

0.6 ± 0.05 0.8 ± 0.05 0.2 ± 0.5 0.4 ± 0.3AB 0.5 ± 0.2A 0.4 ± 0.5 -0.5 ± 0.2B

0.6 ± 0.1 0.8 ± 0.08 0.6 ± 0.09 0.8 ± 0.07A 0.1 ± 0.03B 0.5 ± 0.1 -0.3 ± 0.3AB

work demonstrates that systems with increased microbial diversity also exhibits increased microbial functional resiliency, as measured by extracellular enzyme activity before, during, and after a severe drought. The ability for bioenergy systems to support biodiversity and ecosystem functioning above and beyond biofuel feedstock production is central to developing sustainable biofuel management systems. Furthermore, it is important for these systems to sustain or recover ecosystem functioning in response to stresses like drought, which may increase in the immediate future. Overall, tallgrass prairie restoration for biofuel feedstock purposes met these additional goals better than corn systems. Fungal communities were strongly shaped by management practices. Ascomycota was the most abundant fungal phylum across all management systems, therefore, representing an important aspect of fungal diversity within the site. However, they did not differ between the monoculture cropping system and the diversified prairies, so the phyla maybe difficult to use as a potential metric of prairie restoration success. Other fungal phyla, particularly the Basidiomycota and Glomeromycota, which increased across sampling years within only the prairies, could serve as future metric of the success of a prairie restoration on the soil fungal community. Basidiomycota increased in both prairies as the time from restoration increased. Members of Basidiomycota have highly diverse lifestyles, including ectomycorrhizal fungi, plant pathogens, and are highly studied decomposers of lignin and carbon (Kellner and Vandenbol, 2010). Their potential role in degrading cellulose within soil outside of woodland ecosystems is still understudied, but they are proposed to be highly important for degradation of C within all soil environments (Kjøller and Rosendahl, 2014); within the two prairie types Basidiomycota could play an important role in the degradation of the increased root biomass and litter

129

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

capture the full microbial community response to disturbance events. After diminished activity in response to drought, microbial communities across all management systems returned to a pre-drought state in 2014, but enzyme activities rebounded more strongly in prairie systems, which supported overall greater microbial diversity than in corn. This evidence suggests connections between above- and belowground biodiversity and functional resiliency of ecosystems. Further studies are needed to identify the mechanisms behind these observed relationships. Inclusion of belowground communities and activities in long-term studies may play an important role in developing and adopting alternative agricultural management systems to assure long-term sustainability and maintenance of soil for the complete ecosystem.

agricultural perspective on temperature, which also reflected the harsh conditions of the drought year in 2012. During the drought in 2012, enzyme activity was impacted by management system, with the fertilized prairie supporting higher potential enzymatic activity than unfertilized prairie and continuous corn. Potentially, a shift to a community more distinguished by bacteria than fungi contributed to a decrease in overall microbial contributions to C and N cycling in unfertilized prairie during drought, which has ecosystem scale consequences for soil health, fertility, and environmental services. Interestingly, the highest extracellular enzyme activity across all ecosystems occurred in 2014, coinciding with the greatest rainfall and similar pre-drought soil gravimetric water content during the growing season. All EEAs demonstrated a positive correlative relationship with precipitation, showing the strong effect of rainfall on microbial extracellular enzyme production across management systems. Increased inputs from root biomass in prairies, and stover build-up in the no-till corn system (Dietzel et al., 2015) may have coupled with optimal environmental conditions to support increased extracellular enzyme activity cycling activity in 2014. Management aimed at increasing plant diversity has been shown to have many benefits, ranging from increased pollinator abundance and decreased herbivore and disease damage (Schulte et al., 2017). Here we show that adoption of diverse prairie communities as a biofuel cropping system can confer the additional benefit of increased microbial resiliency to drought. Our experiment demonstrates increased resiliency in the prairies’ extracellular enzyme activity during a natural drought disturbance event and the natural recovery afterwards compared to the corn system. The fertilized prairie has half the N inputs as the corn system does, but over 9 times as many root inputs (Dietzel et al., 2015, 2017; Jarchow and Liebman, 2012), indicating that both the abundant and diverse C inputs as well as the N application create a more productive microbial environment. Our study showed a strong correlative relationship for both fungal and bacterial OTU abundance and root biomass and soil C (Fig. 1,2), demonstrating yet another trophic level heavily shaped by aboveground management and root abundance. Fungal and bacterial communities did not decrease in diversity as dynamically in the fertilized prairies compared to the decreased microbial diversity measurements in the unfertilized prairie and continuous corn during drought. The application of N in a tallgrass prairie showed potential increased in microbial diversity, and a less dynamic reduction in microbial richness during the drought. In the fertilized prairie, there may be a higher incidence of functional redundancy, due to the higher microbial diversity occurring in the microbial community, allowing for resilience to disturbance events such as drought. When multiple microorganisms overlap functionally within a management system, there are many levels of sensitivity amongst these organisms, allowing for continued functioning of the whole community when a perturbation occurs. Inclusion of these overlapping, but functionally maintaining attributes of the fertilized prairie microbial community should be considered when assessing bioenergy cropping management strategies (Griffiths and Philippot, 2013; Shade et al., 2012). Our study demonstrated that microbial communities in restored prairies supported higher microbial diversity and greater functional resiliency than in corn systems. Both high C inputs, through increased root biomass (Dietzel et al., 2015, 2017; Jarchow and Liebman, 2012), and N application created a more active and diverse microbial community. Such insight is important for potential prairie restoration management approaches that desire maximum aboveground production; including harvest for biofuel feedstock or livestock feed hay. With changing global environment, including altered precipitation patterns, identifying resilient aboveground practices and their soil microbial communities is key to determining best management practices to build productive systems that can withstand and adapt to changing environments. Additionally, our study demonstrated discrepancy between the influence of aboveground management and the fungal and bacterial community, highlighting the need to undertake long term studies to

Author contributions RU, EB, KS all contributed equally to the design of the research and editing the manuscript. RU performed the experiment, analyzed the data, and wrote the paper with input from EB and KS. Declaration of Interests Declarations of interest: none, for all authors. Acknowledgments This research was supported by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research program under award number DESC0010775. We also thank Matt Woods and the late David Sundberg for maintaining the COBS research plots. We thank Sheryl Bell for all her help and advise with this project, as well as undergraduate assistants Montana Smith, Christina Davis, Jacinta Msira, Jack Nielsen and Maia Clipsham. Thank you to our reviewers and editor for their thoughtful input. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.agee.2018.07.021. References Acosta-Martinez, V., Cotton, J., Gardner, T., Moore-Kucera, J., Zak, J., Wester, D., Cox, S., 2014. Predominant bacterial and fungal assemblages in agricultural soils during a record drought/heat wave and linkages to enzyme activities of biogeochemical cycling. Applied Soil Ecology 84, 69–82. Allan, S.N., 2017. Disturbance and the Community Composition of Arbuscular Mycorrhizal Fungi in Ontario Tallgrass Prairies. Electronic Thesis and Dissertation Repository. 4835. . Allmaras, R.R., Linden, D.R., Clapp, C., 2004. Corn-residue transformations into root and soil carbon as related to nitrogen, tillage, and stover management. Soil Science Society of America Journal 684, 1366–1375. Anderson-Teixeira, K.J., Davis, S.C., Masters, M.D., Delucia, E.H., 2009. Changes in soil organic carbon under biofuel crops. Gcb Bioenergy 11, 75–96. Aronesty, E., 2013. Comparison of sequencing utility programs. The Open Bioinformatics Journal 7. Averett, J.M., Klips, R.A., Nave, L.E., Frey, S.D., Curtis, P.S., 2004. Effects of soil carbon amendment on nitrogen availability and plant growth in an experimental tallgrass prairie restoration. Restoration Ecology 124, 568–574. Bach, E.M., Hofmockel, K.S., 2015. Coupled carbon and nitrogen inputs increase microbial biomass and activity in prairie bioenergy systems. Ecosystems 183, 417–427. Baer, S.G., Meyer, C.K., Bach, E.M., Klopf, R.P., Six, J., 2010. Contrasting ecosystem recovery on two soil textures: implications for carbon mitigation and grassland conservation. Ecosphere 1, 1–22. Barber, N.A., Chantos-Davidson, K.M., Amel Peralta, R., Sherwood, J.P., Swingley, W.D., 2017. Soil microbial community composition in tallgrass prairie restorations converge with remnants across a 27‐year chronosequence. Environmental Microbiology 19, 3118–3131. Benny, G.L., Humber, R.A., Voigt, K., 2014. 8 Zygomycetous Fungi: Phylum Entomophthoromycota and Subphyla Kickxellomycotina, Mortierellomycotina, Mucoromycotina, and Zoopagomycotina, Systematics and evolution. Spring Publishing, Inc., pp. 209–250. Börjesson, P., Tufvesson, L.M., 2011. Agricultural crop-based biofuels–resource efficiency and environmental performance including direct land use changes. Journal of

130

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Kim, S., Dale, B.E., Ong, R.G., 2012. An alternative approach to indirect land use change: Allocating greenhouse gas effects among different uses of land. Biomass and Bioenergy 46, 447–452. Kjøller, R., Rosendahl, S., 2014. Cultivated and fallow fields harbor distinct communities of Basidiomycota. Fungal Ecology 9, 43–51. Klopf, R.P., Baer, S.G., Bach, E.M., Six, J., 2017. Restoration and management for plant diversity enhances the rate of belowground ecosystem recovery. Ecological Applications 27, 355–362. Kneller, T., Harris, R.J., Bateman, A., Muñoz-Rojas, M., 2018. Native-plant amendments and topsoil addition enhance soil function in post-mining arid grasslands. Science of The Total Environment 621, 744–752. Kuzyakov, Y., Blagodatskaya, E., 2015. Microbial hotspots and hot moments in soil: concept & review. Soil Biology and Biochemistry 83, 184–199. Landis, D.A., Gardiner, M.M., van der Werf, W., Swinton, S.M., 2008. Increasing corn for biofuel production reduces biocontrol services in agricultural landscapes. Proceedings of the National Academy of Sciences 105, 20552–20557. Lau, J.A., Lennon, J.T., 2012. Rapid responses of soil microorganisms improve plant fitness in novel environments. Proceedings of the National Academy of Sciences 109, 14058–14062. Leff, J.W., Jones, S.E., Prober, S.M., Barberán, A., Borer, E.T., Firn, J.L., Harpole, W.S., Hobbie, S.E., Hofmockel, K.S., Knops, J.M., McCulley, R.L., 2015. Consistent responses of soil microbial communities to elevated nutrient inputs in grasslands across the globe. Proceedings of the National Academy of Sciences 112, 10967–10972. Lindahl, B.D., Nilsson, R.H., Tedersoo, L., Abarenkov, K., Carlsen, T., Kjøller, R., Kõljalg, U., Pennanen, T., Rosendahl, S., Stenlid, J., Kauserud, H., 2013. Fungal community analysis by high‐throughput sequencing of amplified markers–a user’s guide. New Phytologist 199 (1), 288–299. McBride, A.C., Dale, V.H., Baskaran, L.M., Downing, M.E., Eaton, L.M., Efroymson, R.A., Mulholland, P.J., et al., 2011. Indicators to support environmental sustainability of bioenergy systems. Ecological Indicators 11, 1277–1289. Miller, R.G., 1981. Springer series in statistics. Simultaneous statistical inference. Morgan, J.W., Dwyer, J.M., Price, J.N., Prober, S.M., Power, S.A., Firn, J., Borer, E.T., et al., 2016. Species origin affects the rate of response to inter‐annual growing season precipitation and nutrient addition in four Australian native grasslands. Journal of Vegetation Science 27, 1164–1176. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O’hara, R., Wagner, H., et al., 2013. Package ‘vegan’. Community ecology package, version, 2 (9). Oksanen, J., Kindt, R., Legendre, P., O’Hara, B., Stevens, M.H.H., Oksanen, M.J., Suggests, M., 2007. The vegan package. Community Ecology Package 10, 631–637. Orwin, K.H., Wardle, D.A., 2004. New indices for quantifying the resistance and resilience of soil biota to exogenous disturbances. Soil Biology and Biochemistry 36, 1907–1912. Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Glöckner, F.O., et al., 2012. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Research gks1219. Saiya-Cork, K., Sinsabaugh, R., Zak, D., 2002. The effects of long term nitrogen deposition on extracellular enzyme activity in an Acer saccharum forest soil. Soil Biology and Biochemistry 34, 1309–1315. Samson, F.B., Knopf, F.L., 1996. Prairie conservation: preserving North America’s most endangered ecosystem. Island Press. Samson, F.B., Knopf, F.L., Ostlie, W.R., 2004. Great Plains ecosystems: past, present, and future. Wildlife Society Bulletin 32, 6–15. Sasse, J., Martinoia, E., Northen, T., 2017. Feed Your Friends: Do Plant Exudates Shape the Root Microbiome? Trends in Plant Science. Schulte, L.A., Niemi, J., Helmers, M.J., Liebman, M., Arbuckle, J.G., James, D.E., Asbjornsen, H., et al., 2017. Prairie strips improve biodiversity and the delivery of multiple ecosystem services from corn–soybean croplands. Proceedings of the National Academy of Sciences 114, 11247–11252. Searchinger, T., Heimlich, R., Houghton, R.A., Dong, F., Elobeid, A., Fabiosa, J., Yu, T.-H., et al., 2008. Use of US croplands for biofuels increases greenhouse gases through emissions from land-use change. Science 319, 1238–1240. Shade, A., Peter, H., Allison, S.D., Baho, D.L., Berga, M., Bürgmann, H., Martiny, J.B., et al., 2012. Fundamentals of microbial community resistance and resilience. Frontiers in Microbiology 3. Sheik, C.S., Beasley, W.H., Elshahed, M.S., Zhou, X., Luo, Y., Krumholz, L.R., 2011. Effect of warming and drought on grassland microbial communities. The ISME Journal 5, 1692–1700. Smith, D.P., Peay, K.G., 2014. Sequence depth, not PCR replication, improves ecological inference from next generation DNA sequencing. PLoS One 9, e90234. Tilman, D., Cassman, K.G., Matson, P.A., Naylor, R., Polasky, S., 2002. Agricultural sustainability and intensive production practices. Nature 418, 671. Tilman, D., Hill, J., Lehman, C., 2006. Carbon-negative biofuels from low-input highdiversity grassland biomass. Science 314, 1598–1600. Tognetti, P.M., Chaneton, E.J., 2015. Community disassembly and invasion of remnant native grasslands under fluctuating resource supply. Journal of Applied Ecology 52, 119–128. Tscharntke, T., Klein, A.M., Kruess, A., Steffan-Dewenter, I., Thies, C., 2005. Landscape perspectives on agricultural intensification and biodiversity–ecosystem service management. Ecology letters 8, 857–874. Turner, B.L., 2010. Variation in pH optima of hydrolytic enzyme activities in tropical rain forest soils. Applied and Environmental Microbiology 76, 6485–6493. Turner, B.L., Lambin, E.F., Reenberg, A., 2007. The emergence of land change science for global environmental change and sustainability. Proceedings of the National Academy of Sciences 104, 20666–20671. Van der Heijden, M.G., Klironomos, J.N., Ursic, M., Moutoglis, P., 1998. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and

Cleaner Production 19, 108–120. Borsari, B., Onwueme, I., Kreidermacher, E., Terril, T., 2009. Renewable energy from restored prairie plots in southeastern Minnesota, USA. WIT Transactions on Ecology and the Environment 121. Braissant, O., Cailleau, G., Dupraz, C., Verrecchia, E.P., 2003. Bacterially induced mineralization of calcium carbonate in terrestrial environments: the role of exopolysaccharides and amino acids. Journal of Sedimentary Research 73, 485–490. Buckley, D.H., Schmidt, T.M., 2003. Diversity and dynamics of microbial communities in soils from agro‐ecosystems. Environmental Microbiology 5, 441–452. Burns, R.G., DeForest, J.L., Marxsen, J., Sinsabaugh, R.L., Stromberger, M.E., Wallenstein, M.D., Zoppini, A., et al., 2013. Soil enzymes in a changing environment: current knowledge and future directions. Soil Biology and Biochemistry 58, 216–234. Bushnell, B., 2014. BBMap: A Fast, Accurate, Splice-Aware Aligner. Ernest Orlando Lawrence Berkeley National Laboratory, Berkeley, CA (US). Camacho, C., Coulouris, G., Avagyan, V., Ma, N., Papadopoulos, J., Bealer, K., Madden, T.L., 2009. BLAST+: architecture and applications. BMC bioinformatics 10, 421. Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello, E.K., Gordon, J.I., et al., 2010. QIIME allows analysis of high-throughput community sequencing data. Nature methods 7, 335–336. Chodak, M., Gołębiewski, M., Morawska-Płoskonka, J., Kuduk, K., Niklińska, M., 2015. Soil chemical properties affect the reaction of forest soil bacteria to drought and rewetting stress. Annals of microbiology 65, 1627–1637. Collins, H., Christenson, D., Blevins, R., Bundy, L., Dick, W., Huggins, D., Paul, E., 1999. Soil carbon dynamics in corn-based agroecosystems: results from carbon-13 natural abundance. Soil Science Society of America Journal 63, 584–591. Collins, S.L., Ladwig, L.M., Petrie, M.D., Jones, S.K., Mulhouse, J.M., Thibault, J.R., Pockman, W.T., 2017. Press–pulse interactions: effects of warming, N deposition, altered winter precipitation, and fire on desert grassland community structure and dynamics. Global Change Biology 23, 1095–1108. De Cáceres, M., 2013. How to use the indicspecies package (ver. 1.7. 1). Centre Tecnològic Forestal de Catalunya, Catalonia. De Cáceres, M., Jansen, F., 2012. Indicspecies-package. Functions to assess the strength and significance of relationships of species site group associations. Version, 1 (5). De Deyn, G.B., Cornelissen, J.H., Bardgett, R.D., 2008. Plant functional traits and soil carbon sequestration in contrasting biomes. Ecology letters 11, 516–531. De Vries, F.T., Liiri, M.E., Bjørnlund, L., Bowker, M.A., Christensen, S., Setälä, H.M., Bardgett, R.D., 2012. Land use alters the resistance and resilience of soil food webs to drought. Nature Climate Change 2, 276. De Vries, F.T., Shade, A., 2013. Controls on soil microbial community stability under climate change. Frontiers in microbiology 4. DeForest, J.L., 2009. The influence of time, storage temperature, and substrate age on potential soil enzyme activity in acidic forest soils using MUB-linked substrates and LDOPA. Soil Biology and Biochemistry 41, 1180–1186. Dietzel, R., Jarchow, M.E., Liebman, M., 2015. Above-and belowground growth, biomass, and nitrogen use in maize and reconstructed prairie cropping systems. Crop Science 55, 910–923. Dietzel, R., Liebman, M., Archontoulis, S., 2017. A deeper look at the relationship between root carbon pools and the vertical distribution of the soil carbon pool. Soil 3, 139. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460–2461. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature methods 10, 996–998. Fargione, J., Hill, J., Tilman, D., Polasky, S., Hawthorne, P., 2008. Land clearing and the biofuel carbon debt. Science 319, 1235–1238. Flores-Moreno, H., Reich, P.B., Lind, E.M., Sullivan, L.L., Seabloom, E.W., Yahdjian, L., Báez, S., et al., 2016. Climate modifies response of non-native and native species richness to nutrient enrichment. Phil. Trans. R. Soc. B 371, 20150273. Gelfand, I., Sahajpal, R., Zhang, X., Izaurralde, R.C., Gross, K.L., Robertson, G.P., 2013. Sustainable bioenergy production from marginal lands in the US Midwest. Nature 493, 514. German, D.P., Weintraub, M.N., Grandy, A.S., Lauber, C.L., Rinkes, Z.L., Allison, S.D., 2011. Optimization of hydrolytic and oxidative enzyme methods for ecosystem studies. Soil Biology and Biochemistry 43, 1387–1397. Griffiths, B.S., Philippot, L., 2013. Insights into the resistance and resilience of the soil microbial community. FEMS Microbiology Reviews 37, 112–129. Hargreaves, S.K., Hofmockel, K.S., 2015. A modified incubation method reduces analytical variation of soil hydrolase assays. European Journal of Soil Biology 67, 1–4. Harpole, W.S., Stevens, C.J., 2016. Addition of multiple limiting resources reduces grassland diversity. Nature 537, 93–96. Hawkes, C.V., Wren, I.F., Herman, D.J., Firestone, M.K., 2005. Plant invasion alters nitrogen cycling by modifying the soil nitrifying community. Ecology Letters 8, 976–985. Hoekstra, J.M., Boucher, T.M., Ricketts, T.H., Roberts, C., 2005. Confronting a biome crisis: global disparities of habitat loss and protection. Ecology letters 8, 23–29. Isbell, F., Tilman, D., Polasky, S., Loreau, M., 2015. The biodiversity‐dependent ecosystem service debt. Ecology letters 18, 119–134. Jarchow, M.E., Liebman, M., 2012. Tradeoffs in biomass and nutrient allocation in prairies and corn managed for bioenergy production. Crop Science 52, 1330–1342. Jarchow, M.E., Liebman, M., 2013. Nitrogen fertilization increases diversity and productivity of prairie communities used for bioenergy. Gcb Bioenergy 5, 281–289. Kõljalg, U., Nilsson, R.H., Abarenkov, K., Tedersoo, L., Taylor, A.F., Bahram, M., Callaghan, T.M., et al., 2013. Towards a unified paradigm for sequence‐based identification of fungi. Molecular Ecology 22, 5271–5277. Kellner, H., Vandenbol, M., 2010. Fungi unearthed: transcripts encoding lignocellulolytic and chitinolytic enzymes in forest soil. PLoS One 5, e10971.

131

Agriculture, Ecosystems and Environment 266 (2018) 122–132

R.N. Upton et al.

Wright, C.K., Wimberly, M.C., 2013. Recent land use change in the Western Corn Belt threatens grasslands and wetlands. Proceedings of the National Academy of Sciences 110, 4134–4139. Yandell, B.S., 1997. Practical data analysis for designed experiments, vol. 39 Crc Press.

productivity. Nature 396, 69. van der Putten, W.H., 2010. Impacts of soil microbial communities on exotic plant invasions. Trends in Ecology & Evolution 25, 512–519. Wieder, W.R., Bonan, G.B., Allison, S.D., 2013. Global soil carbon projections are improved by modeling microbial processes. Nature Climate Change 3, 909–912.

132