Final Report for GS13-119
A comparison of multiple winter rotational crops with summer sorghum under moderate (100 Kg N ha-1) and low (100 Kg N ha-1) nitrogen fertilization determined that some rotation crops like clover and sugar beet tops provided a fertilizer N credit of approximately 30 kg N ha-1 to sorghum grown in North Central Florida. Our results also showed that growers need to sow subsequent crops quickly (ideally within 2 weeks) after incorporation of rotation crop residues, as much of the nitrogen from the residues was released within 4-6 weeks after incorporation. Rye reduced nitrate-nitrogen availability in the soil after incorporation.
High levels of chemical nitrogen fertilization are a staple of modern agricultural practices in the Unites States, with 11.6 million metric tons of nitrogen applied to agricultural lands in 2011 (USDA ERS, 2014). These inputs, combined with other factors, have significantly increased agricultural productivity over the last 50 years, but also have negative effects on agroecosystems, including decreasing soil fertility and increasing erosion and leaching losses (Matson et al., 1997; Tilman et al., 2002). Chemical nitrogen fertilizer application has also been indicted as a major driver of increasing N2O concentrations in the atmosphere, an important greenhouse gas and cause of concern with regard to climate change (Park et al., 2012). Recently, growing concern over the impacts of chemical fertilizers on agroecosystems and the broader environment has spurred research on means to reduce nitrogen inputs and mitigate the detrimental effects of fertilization, while maintaining yields (Grant et al., 2002; Smith et al., 1997). Based on prior research, Grant et al. (2002) concluded that continuous rotated cropping systems can provide multiple benefits to subsequent crops and the agroecosystem, particularly in no-till systems, relative to fallow systems, provided proper management strategies are followed.
Rotational cropping and residue return have been advocated to decrease the need for external nitrogen inputs, but crop selection and return practices must be properly managed. Vigil and Kissel (1991) determined that return of crop residues with a C:N ratio greater than 40, which typically corresponds to tissue N concentrations of or lower than 10 g N kg-1, will generally result in net soil nitrogen immobilization, and not a return of N to a subsequent crop. Residues with higher tissue N concentrations have the potential to return significant quantities of nitrogen to the soil and a subsequent crop. Yamoah et al. (1998) showed that grain sorghum grown following soybean had higher average yields over an 18 year period relative to continuous sorghum (5130-7120 kg ha-1 versus 4050-6260 kg ha-1 relatively) in Nebraska, and that soybean could contribute up to 83 kg N ha-1 yr-1 to sorghum depending on climactic factors. However, Havlin et al. (1990) demonstrated that continuous sorghum cropping provided the greatest increase in soil nitrogen when compared with sorghum-soybean or continuous soybean systems, and that the effects of fertilizer application on soil organic nitrogen were minimal. These results indicate that the residue fertilizer nitrogen credit to a subsequent crop is strongly affected by environmental conditions as well as crop selection (Cameron et al., 2013). Lacking from these studies though is a quantified understanding of the temporal dynamics of nitrogen in the rooting zone of the subsequent crop, as these studies integrated the cumulative effects of years.
Prior research on grain sorghum following a crimson clover crop has shown the potential for legumes to offset fertilizer requirements. However, even when clover contained 202 to 216 kg N ha-1, only an estimated 128 kg N ha-1 of fertilizer N was replaced by the clover supplied N, or an N recovery rate of 59-63% (Hargrove, 1986). McVay et al. (1989) reported similar effects, with crimson clover preceding grain sorghum replacing 21-80 kg fertilizer N ha-1 based on yield relative to an unfertilized sorghum system with cool season fallow. In contrast, Singh et al. (2012) reported nitrogen uptake rates of 133 to 139 kg N ha-1 by two sweet sorghum cultivars grown in the southeastern U.S. for two years at two sites when fertilized with chemical N at 135 kg ha-1, a recovery rate of 99-103%, which is considerably higher than found with the legume residue studies. These differences in nitrogen uptake are attributable to multiple factors, including cover crop decomposition rate and microbial lockup of nitrogen, but must be considered when selecting rotational crops and calculating nitrogen offsets. In order to accurately determine how much nitrogen may be supplied to a subsequent crop, soil nitrogen mineralization and availability should be investigated for various rotations and environmental conditions.
Nitrogen mineralization and availability in the soil can be measured in numerous ways (Bai et al., 2012; DiStefano and Gholz, 1986). Soil incubation in buried plastic bags or columns is a common measure of soil mineralization, but may produce an unnaturally constrained environment for nitrogen mineralization, including excessive moisture, even in the sandy soils of North Florida (DiStefano and Gholz, 1986; He et al., 2000). Ion exchange resins have the ability to accumulate cations or anions from water moving through the soil and have been widely used and reviewed as a method for in situ comparisons of nitrogen dynamics as they mimic the effects of plant roots and do not allow ion concentrations to build in the surrounding soil (Ziadi et al., 2006). Additionally, the use of ion exchange methods has been shown to be sensitive to smaller perturbations in soil ion availability than other methods while maintaining a more natural, and potentially less disturbed, environment after insertion.
- Ascertain yields and tissue nitrogen concentrations of sorghum grown under high and low fertility in rotation with winter cover crops.
- Monitor soil nitrogen pools and organic matter content, and quantify monthly availability of nitrogen in the rooting zone of sorghum and winter cover crops, and correlate with crop yields and tissue nitrogen concentrations.
A replicated field experiment was conducted in North Florida at the University of Florida Plant Science Research and Education Unit (29°24’N 82°10’W) in Citra, Florida. The soil at the site is a relatively well-drained Arredondo fine sand (loamy, siliceous, semiactive, hyperthermic Grossarenic Paleudults). The previous crop was bahiagrass pasture followed by cool season fallow. The trial was designed with sweet sorghum (‘M81-E’) nitrogen fertilizer rate as the main plot in a completely randomized design with cool season rotation crop as the split plot in a completely randomized design within each main plot. Treatments consisted of nitrogen fertilization rate as main plot effects (low, 20 kg N ha-1 sorghum crop-1; or moderate, 100 kg N ha-1 sorghum crop-1) and cool season rotation crops as sub-plots (fallow, rye [Secale cereale ‘FL-401’], camelina [Camelina sativa (L.) Crantz ‘311’], sugar beet [Beta vulgaris ‘EN-413’ (Betaseed, Inc.)], and red clover [Trifolium pratense ‘Southern Belle’]). Main plots were 30 m long and 6 m wide and sub plots were 6 m by 6 m. There were 4 replicates of each main plot, for a total of eight 30 m by 6 m plots.
Plot areas were rototilled, cultivated and packed prior to sweet sorghum and cool season crop planting. Planting was accomplished using a 4 row John Deere MaxEmerge planter with John Deere 7620 tractor with 0.76 m between row spacings, in row spacing of 6-8 cm and an approximate planting depth of 2.5 cm in mid-May of each year for sweet sorghum. Liquid fertilizer (11-37-0) was applied at sorghum planting at a rate of 20 kg N ha-1, in conjunction with a systemic insecticide-nematicide and herbicides to control weeds. For moderate nitrogen rate treatments, the remaining 80 kg N-1 ha-1 was side-dressed as ammonium nitrate (34-0-0) in two split applications of 40 kg each, approximately three and six weeks after planting.
After final sampling and sorghum harvest each year, remaining sorghum was cut at a stubble height of 10 cm using a two-row forage harvester and removed from the field. All plots were rototilled to incorporate sorghum stubble to a depth of 50cm and cultivated. Cool season rotation crops were planted in mid-October of each year. Camelina, rye, and clover were established using a grain drill with between row spacing of 0.19 m, and sugar beets were planted by push planter with 0.76 m between row spacing and 6-8 cm in row spacing. Rye and camelina were fertilized with granular fertilizer (15-0-15) to a total of 50 kg N ha-1 as two equally split broadcast applications approximately four and eight weeks after planting in each year. On the same dates, clover received broadcast applications of muriate of potash (0-0-60) for a total of 50 kg K2O ha-1. Sugar beets were fertilized in three split side-dressed applications to a yearly total of 100 kg N ha-1 and 50 kg K2O ha-1. Overhead irrigation was provided to all plots at establishment, and as necessary at signs of visual drought stress. Weeds were controlled mechanically in all plots, including fallow. After yield sampling, camelina plots were threshed for grain yield and all beet roots were hand pulled and removed from the field. Remaining biomass, including threshed camelina stem, beet tops, clover and rye were returned to the plots incorporated by rototilling.
Soil nitrate-nitrogen availability was monitored with ion exchange resin bags installed approximately every 30 days in each plot, beginning in January of the first year of rotational crop planting, and lasting until after harvest of sweet sorghum in the 3rd year. Bags were constructed monthly by packing 5 g wet weight of anion exchange resin (Amberlite IRA-400(Cl), Alfa Aesar, Ward Hill, MA) in a synthetic mesh bag and then the bags were sealed with zip-ties. Prior to soil incubation, bags were leached with 0.5 M HCl for at least 30 minutes, and then soaked in sequential DI water baths until the resultant pH was neutral. Bags were installed in the field, one per plot, on a monthly basis by removing an intact soil core to a depth of 20 cm using a soil core sampler, which was replaced intact over each bag. Bags were located within 5 cm of one of the two inner rows of sorghum over the summer, and beet over the winter, and were between rows in the middle of each plot over the winter for all other rotational crops. Bags were removed at the end of each month by means of a string tied to each bag at installation and left on the surface.
After removal, loose soil was brushed from the surface of each bag and any roots which had grown into a bag were removed by forceps. Cleaned bags were stored in individual 50 mL conical centrifuge tubes (Corning, Inc., Corning, NY) at 4°C until extraction. Each bag was extracted with a known volume of 0.1 M HCl and 2.0 M NaCl (ACS Reagent Grade, Thermo Fisher Scientific, Waltham, MA) in capped tubes while shaking for at least 12 hours at ambient temperature. Extracts were filtered through coarse porosity filter paper to remove residual soil particles, and then analyzed for nitrate-N according to EPA method 353.2. Blank correction bags were prepared as above on a monthly basis and stored in sealed containers at 4°C until extracted with the soil incubated bags. Extraction solutions and water were tested monthly for nitrate-nitrogen.
Cumulative nitrate-nitrogen availability during the sorghum growing seasons (2013 and 2014) was calculated as the sum of adsorbed nitrate-nitrogen per plot for each month when sorghum was present in the field scaled based on resin bag size, and assumed uniform incorporation of the cover crops to a depth of 40 cm (the measured depth of the rototiller used for cover crop incorporation).
Sorghum dry biomass yields in 2012 were not significantly affected by nitrogen fertility, and averaged 15.2 Mg ha-1. This is consistent with prior literature for one year nitrogen fertility studies on sandy soils, which have shown minimal effects of nitrogen fertilizer addition on sorghum dry biomass yields after one season (Erickson et al., 2012). There was no significant N x cover effect in both 2013 and 2014, but dry biomass yields were affected by N fertility with 55 and 60% higher yields, respectively, under moderate versus low fertility. Additionally, sorghum biomass yields under low fertility averaged across cover crops over the 3 years showed a linear decrease in yield. This data suggests that long-term, low input sorghum production with high yields is not feasible.
Cover crop affected dry biomass production in 2013, with sorghum producing more biomass when preceded by clover or beets than rye. However, cover crop did not affect sorghum dry biomass yields in 2014. This is likely due to a delayed planting of sorghum in 2014, which occurred 49 days after cover crop incorporation, versus 14 days after incorporation in 2013. The majority of additional nitrogen which may be supplied by an incorporated cover crop is lost within the first 30 days after incorporation. The cover crop by fertility interaction was not significant in either year.
Total nitrogen removal by aboveground sorghum biomass was higher in all years for moderate versus low fertility sorghum, and differences were driven primarily by differences in total biomass yields. Tissue N concentrations were not affected by application rate in 2012 or 2014 (6.4 and 4.8 mg N g dry biomass-1 respectively), but were affected in 2013 and were higher in moderate than low fertility applications. Cover crop affected tissue N concentration in 2013 with clover resulting in higher tissue N concentrations than fallow systems. Additionally, the reported time to mineralize half of the available nitrogen from an incorporated cover crop ranged from 18 to 28 days; thus, planting 14 days following cover crop incorporation in 2013 resulted in the majority of cover crop N remaining available during sorghum growth, while in 2014 the delayed planting likely resulted in the majority of the nitrogen having been mineralized and lost prior to planting.
Consistent with the strong linear decline in sorghum biomass yields under low input conditions discussed above, both sorghum tissue nitrogen concentration and total sorghum nitrogen removal showed strong negative linear trends across years. These results further support the inability to sustainably produce sorghum under continuous low input conditions. However, further research to optimize cover crop management in rotation with sorghum planting and maximize nitrogen contribution from the rotation crop may allow for the apparent credit of up to 30 kg N ha-1 to sorghum from red clover to be combined with reduced nitrogen fertility to reach the predicted nitrogen fertilization requirement for yield maintenance with reduced fertilizer inputs.
Soil Available Nitrogen Dynamics
Soil nitrate-nitrogen availability was significantly affected by cover crop during 11 of 17 monitored time periods, and significantly or marginally interacted with sorghum fertility rate over half of those times (6 of 11 periods). Following cover crop incorporation, all plots experienced a spike in soil available nitrate-nitrogen in May in both 2013 and 2014 (Fig. 1). This effect was likely due at least in part to tillage, which introduces oxygen deeper into the soil profile, disrupts soil aggregates, and stimulates organic matter breakdown (St. Luce et al., 2011), but the effects were variable across cover crops. In the first month after incorporation in 2013, clover plots showed nitrate-nitrogen levels almost three times higher than in fallow plots, and twice as high as observed for beet tops. These results are consistent with prior research on clover decomposition kinetics, which have shown that approximately 50% of nitrogen mineralized from an incorporated clover over 20 weeks occurs, and thus may be lost, in the first 4 weeks (Frankenberger and Abdelmagid, 1985). However, rye and camelina both show relatively lower soil available nitrate-nitrogen than under a fallow system, though the effect was marginal (p=0.11). This was likely due to nitrogen immobilization by microbial communities breaking down the higher C:N residue. Vigil and Kissel (1991) identified a C:N ratio greater than 40 as necessary for net nitrogen immobilization following cover crop incorporation, and later research by Kuo and Sanjou (1998) identified incorporation of greater than 60% cereal rye biomass in conjunction with a legume as resulting in increased net nitrogen immobilization. Wells et al. (2013) determined that soil nitrogen availability is reduced following rolling of a rye cover crop relative to conventional tillage, and that peak nitrogen immobilization occurred 4-6 weeks after rolling.
There was a cover by nitrogen rate effect that persisted throughout the 2013 sorghum growing season. In all 4 months, soil nitrate-nitrogen availability for sorghum under moderate fertility was significantly higher than all other treatments, indicating that nitrogen availability to the crop was substantially higher. Based on van Oosterom et al. (2010), continued availability of nitrate-nitrogen in the rooting zone may allow for a near-linear increase in total plant nitrogen on a land-area basis, though the effect on yield may not be as distinct. However, Olson et al. (2013) reported that total plant nitrogen did not significantly increase from 60 to 150 days after emergence in a high-biomass producing sorghum, indicating that the majority of nitrogen may be taken up in the first 60 days. Thus, a rapid breakdown of clover under low fertility may result in the loss of nitrates from the rooting zone, and a subsequent lack of benefit to the following crop if not planted shortly after incorporation.
In 2014, similar results for monthly nitrate-nitrogen availability were observed in the month following incorporation. Due to stand emergence issues however, sorghum was re-planted approximately 6 weeks following cover crop incorporation, versus 2 weeks post-incorporation in 2013. In the first month following incorporation (May), clover produced the highest levels of soil available nitrate-nitrogen, and was comparable to beet tops and higher than all other crops (Table 4-3). Beet tops were not significantly different from camelina, and higher than fallow and rye plots across nitrogen rates. However, the elevated level of soil available nitrate-nitrogen observed under moderate N clover in 2013 were not present in 2014, potentially due to heavy early-season rainfall, which is known to accelerate decomposition and leaching of nitrogen (St. Luce et al., 2011). Rye, camelina and fallow plots showed no differences in the first month following incorporation in 2014. During the second month (June) following incorporation no significant differences were observed between any of the incorporated covers (p=0.24), nor between the moderate and low fertility treatments (p=0.99). In the fourth month following incorporation (August), the last split application of nitrogen fertilizer to the moderate fertility plots resulted in a seven-fold increase in available nitrates at 20 cm. No differences in soil available nitrates were observed during the last (5th, September) month of sorghum growth in 2014. The lack of differentiation between cover crops after the first month post-incorporation indicated that the succeeding crop must be planted shortly after cover crop termination to maximize potential nitrogen uptake.
During rotational crop production over the fall, winter and spring of 2013-14, nitrate-nitrogen levels were generally low except immediately following tillage activities when higher nitrate-nitrogen levels were uniformly observed across all plots (3.09 kg N ha-1 in September 2013), or in plots fertilized with 40 kg N ha-1 in a single month (beets in Jan 2013 and Jan 2014) (Fig. 1). Tillage and incorporation of sorghum stubble produced an increase in observed nitrate-nitrogen at 20 cm, but the magnitude of the increase was not significantly different between treatments. Fertilization of 25-30 kg N ha-1 to beets, rye, and camelina in December of 2013 did not significantly affect soil available nitrate-nitrogen (p = 0.18), but observed nitrate-nitrogen levels were numerically smaller in beet, rye and camelina plots relative to clover and fallow. Nitrate-nitrogen availability in the soil was relatively low and consistent across clover, rye, camelina and fallow plots, suggesting that N applied as fertilizer was either taken up by the crop or soil microbial biomass. This was supported by cover crop nitrogen removal rates, which equaled or exceeded applied N.
Sorghum Season Available Nitrogen
Cool season rotational crop effects on cumulative soil available nitrate-nitrogen during the sorghum growing season were not different between moderate and low nitrogen application treatments in either year. The observed effects of rotation crop were comparable between 2013 and 2014, but only marginally significant (p = 0.08) for 2014. In both years, clover incorporation resulted in increased soil available nitrate-nitrogen (251% in 2013 and 76% in 2014 relative to fallow), while incorporation of beet tops, rye and camelina stems did not significantly affect mineralized N. However, beet tops showed a trend towards increasing available soil N, while rye and clover showed a trend towards decreasing available N. Similarly, sugar beet tops were fertilized with 100 kg N ha-1 each winter, which was primarily apportioned to leaves, and incorporated into the soil at harvest. While the effect of sugar beet leaves was not significant, the observed trend was likely attributable to higher N fertility on the cool season crop. Interestingly, rye and camelina incorporation showed decreases in soil available nitrate-nitrogen in 2013 relative to fallow plots (66% and 39% respectively).
Educational & Outreach Activities
This work was part of a Ph.D. dissertation by Jeffrey R. Fedenko that will be defended in Fall of 2015 at the University of Florida. It is anticipated that that these results will soon be submitted for publication to peer-reviewed agronomy journals. It is also expected that an outreach publication will be produced that addresses the nitrogen credit from the winter legume along with release dynamics following incorporation to assist growers with managing with winter rotation crops in the region. Additionally other Ph.D., M.S., and undergraduate students were involved in the project and educated on the use, potential benefits, and production of winter rotation crops with summer cash crops in the region. The main Ph.D. student on the project has already begun a research position in industry.
The short-term impacts of this research are better understanding of the use of potential new winter cover crops in the region, like camelina sativa for oil production. This project along with others has demonstrated that current camelina cultivars do not appear to be a very viable winter rotation crop. This has resulted in farmers no longer showing an interest in camelina and looking at other crops like carinata for a potential oil seed crop. Similarly, sugar beets do seem to have some potential, particularly in South Florida, if allowed to grow late into spring and early summer, but this may be too late for many summer crops. Clover was very successful in this study and is already used in the region, the results from this study will lead to better management of clover for N credit to subsequent crops. We anticipate that the benefit of clover will continue to grow in our system due to the observed yield decline in the summer sorghum at low fertilizer N. So we have continued the project for an additional year, so over the long-term we expect this to increase adoption of winter legumes in some systems in the region, especially sorghum or other energy crop systems, over time as these systems be more established.
As mentioned above, the Ph.D. student on the project has accepted a position with industry now and frequently interacts with farmers and other industry partners. Through this role, the results of his project are now reaching farmers and industry.
Areas needing additional study
The study showed some very interesting results regarding the potential N credit from different winter rotation crops along with temporal dynamics of nitrate-nitrogen from breakdown of residues of incorporated winter rotation crops. One interesting aspect of these results that warrants further research is the fate of inputs of N other than crop uptake or available soil nitrate-nitrogen from some of the cover crops. This would be especially for the clover legume as nitrogen inputs were far greater than the apparent N credit.