The authors of this manuscript have read the journal’s policy and have the following competing interest: DH is an Academic Editor for PLOS ONE.
Nitrogen (N) fertilization affects bioenergy crop growth and productivity and consequently carbon (C) and N contents in soil, it however remains unclear whether N fertilization and crop type individually or interactively influence soil organic carbon (SOC) and total N (TN). In a three-year long fertilization experiment in switchgrass (SG:
Perennial switchgrass (SG:
Past studies showed no consistent pattern of N fertilization effect on SOC and TN contents. N fertilizations can enhance SOC and TN contents by 9–45% in SG croplands [
The large variations of SOC and TN in response to N fertilization are typically attributed to the perennial nature of bioenergy crops and their deep-rooted growth form [
Knowing the abundance of humic-like or protein-like compounds will offer information of chemical recalcitrance [
Given the different nature of SG and GG roots (i.e., chemistry and morphology), we first hypothesize that there is a significant interaction of N fertilization and bioenergy crop type on SOC, TN and plant aboveground biomass such that N fertilization-enhanced SOC, TN and plant aboveground biomass was more pronounced in GG than SG. Alternatively, there is only significant N fertilization effect. In that scenario, we establish the second hypothesize that the N fertilization effect will be significant only under a high fertilization rate because the low fertilization effect can be masked due to large variations in field measurements. Based on a mesocosm experiment examining the two bioenergy crop (SG and GG) seedlings’ characteristics, we set up the third hypothesize that the root leachable dissolved organic matter (DOM) is more structurally complex and less easily decomposed for GG than SG because GG root is larger based on the published data synthesis of root morphology in the same mesocosm study. Although we lack fertilization treatment in the mesocosm study, the root morphology and chemistry of the two bioenergy crops are compared and linked to SOC and TN changes in response to N fertilization.
Initially established in 2011, the bioenergy crop field fertilization experiment is located at the Tennessee State University (TSU) Main Campus Agriculture Research and Education Center (AREC) in Nashville, TN, USA (Lat. 36.12° N, Long. 36.98° W, elevation 127.6 m above sea level). Prior to the establishment of switchgrass and gamagrass croplands, the land use type was the mowed grassland for several decades. No fertilizers were applied during the prior land use. Climate in the region is a warm humid temperate climate with an average annual temperature of 15.1°C, and total annual precipitation of 1200 mm [
In the fertilization experiment, soil samples (0–15 cm) were collected from 12 plots (2 crops × 3 N inputs × 2 replicated plots) on June 6, 2015. Within each plot, 24 cores were randomly collected using a spatially explicit sampling design [
Harvesting of SG and GG aboveground (ABG) biomass was conducted twice in four replicated plots under each of three fertilization treatments during June to October in 2014 and 2015. This resulted in 24 samples in each year (2 crops × 3 N inputs × 4 replicated plots). At each harvest, plants were cut 7 inches above the ground using a Carter Mfg. Co plot harvester with flail cutters and mounted module capable collecting biomass fresh weights in the field. In each plot, subsamples of fresh biomass per unit area were dried to constant weight at 70°C using an Oven King industrial capacity dryer (Washington Industrial Corp. Seattle, WA, USA) to determine dry biomass. The unit of biomass was expressed as Mg ha-1. To analyze biomass C and N concentrations, subsamples of dry biomass in 2014 and 2015 were selected and one composited sample was obtained by equal weight of sample for each crop under each fertilization treatment (i.e., NN, LN, and HN). This generated 6 samples (3 fertilization × 2 crop). Plant samples were analyzed for C and N concentrations using a Costech 4010 elemental analyzer (
The root materials of SG and GG were obtained from the historical archived samples collected from a mesocosm experiment [
For this study, eight replicated root samples for both crops were selected to analyze the abundance and components of dissolved organic matter (DOM) leached from root samples. This generated 16 samples. This analysis was conducted at the Molecular Eco-Geochemistry laboratory of University of Alabama. For DOM leaching, root powders were mixed with carbon-free ultrapure water at a ratio of around 1:8 by mass for most samples. If slurry-like mixture appeared at this ratio, extra water was added until a clear liquid layer appeared. The mixtures were constantly agitated for 42 hours on an orbital shaker at 300 rpm, followed by centrifugation at 4,000 rpm for 20 minutes. The upper liquid layer was carefully transferred to a new vial using a pipette and the leachable DOM in these samples was further analyzed for absorbance and fluorescence properties (i.e., Excitation-Emission Matrix coupled with Parallel Factor Analysis), following the analytical methods described in detail in former publications [
Here we briefly described the procedures on how to conduct the DOM absorbance and fluorescence property analysis. The absorbance of DOM was analyzed using a UV-1800 Shimadzu spectrophotometer, and the spectra from the wavelength of 190 to 670 nm at a 1 nm interval were collected. Three-dimensional fluorescence excitation-emission matrices (EEM) were analyzed using a Horiba Jobin-Yvon Fluoromax-3 spectrofluorometer, with the reading collected at excitation wavelengths from 240 to 500 nm at 5 nm intervals and emission wavelengths from 280 to 538 nm at 3 nm intervals. The EEM spectra were corrected for blanks, the inner filter effect, and the manufacturer’s correction factors and subsequently normalized relative to the area under the water Raman peak [
The parallel factor analysis (PARAFAC) was conducted in MATLAB using the DOMFluor toolbox described in detail by [
Two-way analysis of variance (ANOVA) was used to test the main and interactive effects of N fertilization and crop type on SOC, TN, and C: N, and plant ABG biomass in 2014 and 2015. Tukey HSD
There was no significant interactive effect of fertilization and crop type on SOC, or TN (
Variable | N fertilization | Crop | Crop×N fertilization |
---|---|---|---|
SOC | 0.878 | ||
TN | 0.429 | ||
C: N | 0.401 | 0.163 | |
ABG (2014) | 0.434 | 0.821 | |
ABG (2015) | 0.144 | 0.463 |
Bold numbers denote P < 0.1.
Crop | N Fertilization | SOC | TN | C: N | |||
---|---|---|---|---|---|---|---|
Mean±SE | CV | Mean±SE | CV | Mean±SE | CV | ||
% % | % % | % | |||||
SG | NN | 1.48±0.005a | 0.45 | 0.13±0.002a | 1.86 | 11.05±0.09a | 1.13 |
LN | 1.56±0.09a | 7.91 | 0.15±0.01a | 11.23 | 10.57±0.27a | 3.61 | |
HN | 1.72±0.03b | 2.32 | 0.17±0.0003b | 0.26 | 10.27±0.18a | 2.45 | |
GG | NN | 1.66±0.11a | 9.30 | 0.16±0.007a | 6.08 | 10.10±0.23a | 3.15 |
LN | 1.66±0.14a | 11.52 | 0.16±0.02a | 14.04 | 10.62±0.22a | 2.95 | |
HN | 1.89±0.07b | 5.48 | 0.18±0.004b | 3.51 | 10.40±0.13a | 1.73 |
SG: switchgrass; GG: gamagrass; NN: No N input; LN: Low N fertilizer input (84 kg N ha-1 yr-1)
HN: High N fertilizer input (168 kg N ha-1 yr-1)
There was no significant fertilization effect or interaction of fertilization and crop type on ABG, but there was significant effect of crop type on ABG in both collections in 2014 and 2015 (
There was only significant crop type effect on ABG biomass in each collection year (
Plant ABG biomass was referred to the collection in 2015 only. NN: no N input; LN: low N input (84 kg N ha-1 in urea); HN: high N input (168 kg N ha-1 in urea); ABG: aboveground.
Both SR and FI values were significantly higher for SG than GG samples (
For each panel, the different lowercase letters denote significant difference between SG and GG (
Based on our results, we rejected the first hypothesis that the fertilization and crop type interactively influenced SOC and TN. However, we found that N fertilization significantly increased SOC and TN in both SG and GG croplands. This was likely due to the minimal management and mechanical disturbance in our plots, which minimized soil decomposition due to less exposure of below surface soil to air, consequently diminished soil C and N losses in favor of soil C and N accumulations in perennial bioenergy feedstock grasslands [
The fertilizer-elevated aboveground biomass yield and the belowground rhizodeposits may also have contributed to the SOC and TN sequestrations by supplying additional amounts of C and N to the soil [
Results from this study supported our second hypothesis that relative to no fertilizer input, fertilization resulted in substantial SOC and TN enrichments only at the relatively high N application rates (168 kg N ha-1 yr-1) and less likely so at the low fertilization rates (84 kg N ha-1 yr-1). This finding contrasts with other studies that have demonstrated negative effects of relatively high fertilization rates on soil C and N storage [
Our analysis of leachable DOM from root supported the third hypothesis that GG root contained higher molecular weight and more structurally complex compound than SG root. This result indicates that GG root would be less readily decomposed compared to SG root. SG is known to have a lower specific root length (i.e., root length per unit root biomass) [
On the other hand, stronger linear relationships of SOC and TN with aboveground plant biomass was identified for SG and less so for GG. Given the significantly greater aboveground plant biomass of SG than GG, these results indicate that the contributions of aboveground plant biomass to belowground soil C and N stocks via litterfall input and turnover were stronger in SG than GG. Considering the aforementioned relationship of GG root with soil C and N storage, our results revealed that the plant traits that contributed to the soil C and N sequestrations varied with bioenergy crop species. It was the aboveground plant biomass of SG and the root of GG that have showed likely associations with their respective soil C and N sequestrations. Despite the long known beneficial role of bioenergy crops on soil C, this study highlighted the need to further elucidate the role of different plant traits (e.g., aboveground vs. belowground) in regulating soil C and N sequestration [
This study demonstrated that relative to no fertilizer input, intensive N fertilization (e.g., HN) could significantly increase SOC and TN in bioenergy cropland surface soils (0-15cm). Meanwhile, GG showed significantly higher SOC and TN and significantly lower aboveground biomass than SG. There were strong positive linear relationships of SOC and TN with aboveground biomass in SG, and structurally more complex and less readily decomposed root DOM in GG. This suggested that the intensive N fertilization induced C and N accumulations in soil may be more likely mediated by the aboveground biomass in SG and root chemistry and morphology in GG. Future studies should examine the root characteristics in different bioenergy croplands under the field fertilization experiment.
The modeling method was described in
(DOCX)
(XLSX)
This study was supported by funding awarded to JL from a US National Science Foundation (NSF) HBCU-EiR (No. 1900885), US Department of Agriculture (USDA) Agricultural Research Service (ARS) 1890s Faculty Research Sabbatical Program (No. 58-3098-9-005), and USDA Evans-Allen Grant (No. 1017802). Funding for MAM and GW was provided by the US Department of Energy (DOE) Office of Biological and Environmental Research through the Terrestrial Ecosystem Science Scientific Focus Area at Oak Ridge National Laboratory (ORNL). ORNL is managed by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US DOE. We thank staff members at the TSU’s Main Campus AREC in Nashville, Tennessee for their assistance. We appreciate the anonymous reviewers for their constructive comments and suggestions.