Bioone.org will be down for maintenance on 16 May, 2022 from 18:00-22:00 Pacific US. We apologize for any inconvenience.
Open Access
Translator Disclaimer
1 June 2018 Dry matter partitioning and residue N content for 11 major field crops in Canada adjusted for rooting depth and yield
Arumugam Thiagarajan, Jianling Fan, Brian G. McConkey, H.H. Janzen, C.A. Campbell
Author Affiliations +
Abstract

To improve the estimates of C and N inputs to soil, we developed new estimates of partitioning between the harvested portion, aboveground residue, and belowground residue for 11 major crops based on depth-adjusted root/shoot ratios and grain yield-adjusted harvest indices. We updated the mean N concentration of each partition.

Introduction

More than 3.7 Pg of crop residues are produced worldwide each year and are one of the major sources of C to sustain cropland soil organic carbon (SOC) stocks (Lal 2004). Residue C (i) contributes new decomposable organic C that provides the energy that driving biogeochemical cycles, (ii) represents extracted atmospheric CO2, and (iii) ultimately dictates the overall health and productivity of soils (Wang et al. 2016). In addition, globally, almost 12 Tg of N is contributed to the soil through crop residues (Cassman et al. 2002). The N concentration strongly governs the rate of residue decomposition (Janzen and Kucey 1988). Together, the C and N from crop residues are essential inputs to the belowground food web (Johnen and Sauerbeck 1977; Malhi et al. 2012). Owing to these indispensable roles of C and N, a precise and accurate estimation of C and N inputs from crop residues is critical to agro-ecological models and studies assessing nutrient budgets, carbon pools, soil organic matter (SOM), and soil quality (Bolinder et al. 2007; Fan et al. 2016).

Crop residues add C and N through (i) aboveground residues (AGR) (including stems, leaves, seed-holding plant structures, and threshable seed hulls) and (ii) the belowground (BGR) root biomass. The AGR for grain crops is often estimated using the harvest index (HI), the ratio of seed mass to total shoot mass (i.e., seed + AGR). The BGR is frequently estimated from shoot mass based on the root/shoot ratios (RSR). The RSR is the ratio of belowground plant biomass to aboveground plant biomass measured at or near maturity. The belowground biomass is based on observable roots and does not include difficult-to-measure belowground biomass lost in sloughed roots and rhizodeposition. The C deposition and RSR values vary widely among species due to differing root growth patterns (Jackson et al. 1996). For example, grasslands and savannahs tend to have high soil C stocks (Scurlock and Hall 1998) because perennial grass and forage species exhibit higher RSRs than annual species (Fan et al. 2016). Literature values for BGR are based on measurements taken from varying depths, creating inconsistencies in reported RSR values.

Based on literature review, Janzen et al. (2003) provided values for partitioning crop dry matter (DM) between grain yield (G), AGR, and BGR for a wide range of crops grown in Canada, and also provided estimates of N concentration for these DM components. This reference is widely used, including estimation of N input to soil for Canada’s National Inventory Report on Greenhouse Gas Emissions (Environment and Climate Change Canada 2017).

Recently, Fan et al. (2017) analyzed published data and showed that there is a linear relationship between HI and G:

(1)

cjss-2017-0144eq1.gif
where Ic and Sc are the intercept and slope of the relationship between HI and G for crop c. This relationship captures much of the HI variability due to growing conditions and cultivars and so improves estimates of AGR from G. Fan et al. (2016) also developed a set of root distribution functions with depth for 11 major crops for Canada, so that the observed RSR values can be adjusted to a uniform depth.

Our first objective was to incorporate recent developments in describing the root mass distribution with depth and the HI-yield relationship into a new approach to estimate the DM partitioning between G, AGR and BGR for 11 important crops in Canada: wheat (Triticum aestivum L.), maize (Zea mays L.), oat (Avena sativa L.), barley (Hordeum vulgare L.), pea (Pisum sativum L.), chickpea (Cicer arietinum L.), lentil (Lens culinaris L.), soybean (Glycine max L.), canola (Brassica napus L.), flax (Linum usitatissimum L.), alfalfa (Medicago sativa L.) and for a conglomerate of other forages: bromegrass (Bromus inermis Leyss), crested wheat grass (Agropyron spp.), fescue (Festuca arundinacea Schreb), red clover (Trifolium pratense L.), ryegrass (Lolium sp.), Russian wild rye (Psathyrostachys juncea Fisch.), sweet clover (Melilotus officinalis L.), switchgrass (Panicum virgatum L.), and timothy (Phleum pratense L.). Our second objective was to improve the estimates of mean N concentration of G, AGR, and BGR by adding new values from the literature to those already used in Janzen et al. (2003)

Materials and Methods

We compiled a dataset to adjust RSR to uniform soil depth for major crops and forages in Canada by searching databases of Scopus and Google Scholar. We added data from 58 studies (refer to the  Supplementary data 1 1; crop-wise), each comprising of one or more experimental datasets. New data points for mean calculations varied by biomass fractions and by crops. For example, new G mean for wheat included 195 new data points, whereas chickpea had mere two values. The data were aggregated for 11 field crops, and for alfalfa and other forages (bromegrass, crested wheatgrass, fescue, red clover, ryegrass, Russian wild rye, sweet clover, switchgrass, and timothy). For studies when the plants were grown in lysimeters outdoors or in containers within a greenhouse or when authors stated that the roots grown in the field had been sampled to maximum rooting depth, we assumed the RSR was for the whole root profile. Otherwise, the observed root mass used to calculate the RSR was adjusted to selected depth up to maximum rooting depth, dmax, for each crop according to the following equation:

(2)

cjss-2017-0144eq2.gif
where Rd is the root mass for selected depth, d; cjss-2017-0144ieq1.gif is the root mass for the observed depth, dobs; and da , dmax, and c are fitted equation parameters for each crop (Fan et al. 2016). For forage crops, the fitted parameters were those derived for fescue (Fan et al. 2016). Mean values and 95% confidence interval were generated by bootstrapping with 1999 iterations using the “boot” function in R (Canty and Ripley 2012).

Equations 35 describe the relationships between major DM partitions. Because G is the most widely available plant production data from experiments and national statistics, a formulation is provided to calculate each of the other dry mater partitions from G.

(3)

cjss-2017-0144eq3.gif

(4)

cjss-2017-0144eq4.gif

(5)

cjss-2017-0144eq5.gif
where DM d is the whole plant DM to soil depth of d, AGR is aboveground residue, BGR d is belowground residue to depth d (note DM d , AGR, and BGR are in same units as G), Ic and Sc are the intercept and slope of the HI-yield relationship for crop c, and RSR d is the root/shoot ratio to a selected soil depth, d. From eqs. 3 to 5, it is useful to express partitions as a ratio to DM since all are then comparable on a common basis:

(6)

cjss-2017-0144eq6.gif

(7)

cjss-2017-0144eq7.gif

(8)

cjss-2017-0144eq8.gif

The literature values of N concentration used by Janzen et al. (2003) were updated with more recent literature reporting N contents of grain, straw, and (or) roots of the selected crops for Canadian studies. In some cases, we found no new information on N content (e.g., roots in flax). The AGR consists of a range of components, such as stems and leaves, but it was often unclear what components were included in measurements of N concentration although several references described the measured portion as stems. The average N contents in grain, root, and shoots were calculated from observing the following criteria (i) each year, location, and cultivar were treated as a separate data point; for example, a study evaluating two cultivars in 2 yr yielded four values for N content that were then included in calculation of species average, and (ii) when studies included fertilization or related treatments, values were individually scrutinized for variability and extreme values were excluded when the N content had greater than >25% variation from the overall experiment mean.

Results and Discussion

Our depth-adjusted values for RSR (Table 1) will be consistent with the measured values from the literature when all values are referring to the same depth of root mass determination. However, our depth-adjusted root mass estimates used for RSR will produce more comparable and consistent RSR values for a single specified depth than that derived from measured root data from multiple studies in which the depth over which roots were measured varied widely. An important advantage of our new depth-adjusted values is the ability to estimate root material input for a specific depth. From eq. 7 with Table 1, the mean partitioning of DM for grain crops to roots to 20 cm is 13%, whereas it is 20% to full rooting depth. The values available from Janzen et al. (2003) or Bolinder et al. (2007) do not enable distinction in root partitioning for different rooting depths.

Table 1.

Root/shoot ratios (RSR) for different soil depth (cm) for major field and forage crops in Canada.

cjss-2017-0144tab1.gif

For lentil, soybean, and flax, we found no new Canadian observations for N concentration in BGR (Table 2). Overall, we increased the number of observations of N concentration by about 40% compared with earlier work of Janzen et al. (2003) (data not shown). Still, many values remain based on small sample sizes. For six crops (oat, pea, chickpea, soybean, potato, and alfalfa), the sample size was seven or less for residue (Table 2). We took advantage of the increased sample sizes to calculate a measure of variation (i.e., standard deviation) that was not done in the earlier study (Table 2). As expected, the N concentrations from simply increasing sample size generally did not change N concentrations fundamentally after considering their underlying variability (Table 2). We attribute apparent changes in N concentrations from those in Janzen et al. (2003) to assumed better representation of the population by our larger sample of values from the literature. To illustrate, our mean N concentration of oat G was 24.26 g N kg-1 and appears higher than the 18 g N kg-1 value in Janzen et al. (2003). The latter value was based on a sample of only four points to which we added five more points to the sample and, therefore, presumably, the difference reflects a better estimate of the population mean due to the larger sample size. Another example of change we attribute to the effect of increasing sample size is our higher N concentration of canola AGR for which we added 13 new points to produce a nearly doubled total sample size of 27. The updated pea BGR N concentration (21.99 g N kg-1) was twice the previous value (10 g N kg-1). We found no original rigorous value of pea BGR N concentration within the references in Janzen et al. (2003) so our higher value is derived entirely from three new data points we included.

Table 2.

N concentrations of grain (G), aboveground residues (AGR), and belowground residues (BGR) of the major crops in Canada from Janzen et al. (2003) and from this study.

cjss-2017-0144tab2.gif

Compared with Bolinder et al. (2007) values for cereal crops (wheat, oat, barley, and maize), our values for entire root partition are all higher, their values did not all represent the whole rooting depth; when calculated for our RSR to 20 cm, the two data sets were comparable (results not shown). Our values for AGR were within 5% units of those in Bolinder et al. (2007) except for oat, where our value was higher. Because our partitioning to G varies considerably with yield, it is difficult to compare with constant values in other studies. For comparable crops in Janzen et al. (2003) and Bolinder et al. (2007), their G partition values are within 5% units of our values calculated from eq. 6 for typical yields (Table 3). Compared with our yield-adjusted partition values, the constant G partition values from these references will overestimate AGR for high grain yields and underestimate it for low grain yields. This not only lowers accuracy of AGR estimates over a range of grain yields in 1 yr but also compromises the accuracy of a time series of estimates whenever there is a yield trend. Fan et al. (2017) showed that the aboveground partitioning value for maize from Bolinder et al. (2007) was reasonable for mean Canadian grain yields during 2004–2009. Because maize yields have generally increased over time, the yield-adjusted partition value shows that the constant partition value would cause mean Canadian maize AGR to be underestimated before 2004 and overestimated after 2009. Higher yielding cultivars typically also have higher partitioning to G so that the yield-adjusted values also removes much of the cultivar effect on partitioning (Fan et al. 2017).

Table 3.

Calculated plant partitioning of total plant dry matter (DM) into belowground residue (BGR) for entire crop rooting depth, into grain (G), and into aboveground residue (AGR), for dry-matter grain yields of 2, 4, and 8 t ha-1.

cjss-2017-0144tab3.gif

To estimate C input from our partitions, it is necessary to multiply the DM by carbon concentration. Bolinder et al. (2007) estimated the C concentration of all crop partitions is 0.45 g g-1. Additional belowground C input to soil would come from root exudates and sloughing. Bolinder et al. (2007) estimated that this additional C input for all crops is approximately 65% of the BGR based on measured roots.

Conclusion

In summary, we incorporated recent developments in describing the root mass distribution with depth and the HI-yield relationship into a new approach to estimate the DM partitioning between G, AGR, and BGR for 11 major field crops in Canada. This general approach is unique because it adjusts the belowground DM partitioning for a specified soil depth and adjusts the aboveground DM partitioning for harvested yield. We also updated the Janzen et al. (2003) estimates for N content in above- and belowground crop partitions based on literature survey. These developments will improve the estimates of C and N partitioning in the crops and, particularly, for estimates of C and N input from crop residues to the soil.

Notes

1 Supplementary data are available with the article through the journal Web site at  http://nrcresearchpress.com/doi/suppl/10.1139/cjss-2017-0144.

Acknowledgements

This work was funded by Agriculture and Agri-Food Canada. J. Fan acknowledges the Natural Sciences and Engineering Research Council of Canada for the opportunity to work at AAFC as a postdoctoral fellow. We thank Elijah Atuku of AAFC for dedicated work of extracting data from scientific papers.

References

1.

Bolinder , M.A., Janzen , H.H., Gregorich , E.G., Angers , D.A., and VandenBygaart , A.J. 2007. An approach for estimating net primary productivity and annual carbon inputs to soil for common agricultural crops in Canada. Agric. Ecosyst. Environ. 118(1–4): 29–42. https://doi.org/10.1016/j.agee.2006.05.013Google Scholar

2.

Canty, A., and Ripley, B. 2012. boot: Bootstrap R (S-Plus) functions. R package version 1(7).Canty, A., and Ripley, B. 2012. boot: Bootstrap R (S-Plus) functions. R package version 1(7).

3.

Cassman , K.G., Dobermann , A., and Walters , D.T. 2002. Agroecosystems, nitrogen-use efficiency, and nitrogen management. Ambio, 31(2): 132–140. https://doi.org/10.1579/0044-7447-31.2.132. PMID:12078002. Google Scholar

4.

Environment and Climate Change Canada. 2017. National Inventory Report 1990–2015: greenhouse gas sources and sinks in Canada. Environment and Climate Change Canada, Gatineau, QC, Canada.Environment and Climate Change Canada. 2017. National Inventory Report 1990–2015: greenhouse gas sources and sinks in Canada. Environment and Climate Change Canada, Gatineau, QC, Canada.

5.

Fan , J., McConkey , B., Wang , H., and Janzen , H. 2016. Root distribution by depth for temperate agricultural crops. Field Crops Res. 189: 68–74. https://doi.org/10.1016/j.fcr.2016.02.013Google Scholar

6.

Fan , J., McConkey , B., Janzen , H., Townley-Smith , L., and Wang , H. 2017. Harvest index–yield relationship for estimating crop residue in cold continental climates. Field Crops Res. 204: 153–157. https://doi.org/10.1016/j.fcr.2017.01.014Google Scholar

7.

Jackson , R.B., Canadell , J., Ehleringer , J.R., Mooney , H.A., Sala , O.E., and Schulze , E.D. 1996. A global analysis of root distributions for terrestrial biomes. Oecologia, 108(3): 389–411. https://doi.org/10.1007/bf00333714. PMID:28307854. Google Scholar

8.

Janzen , H., Beauchemin , K., Bruinsma , Y., Campbell , C., Desjardins , R., Ellert , B., and Smith , E. 2003. The fate of nitrogen in agroecosystems: an illustration using Canadian estimates. Nutr. Cycling Agroecosyst. 67(1): 85–102. https://doi.org/10.1023/a:1025195826663Google Scholar

9.

Janzen , H., and Kucey , R.M. 1988. C, N, and S mineralization of crop residues as influenced by crop species and nutrient regime. Plant Soil, 106(1): 35–41. https://doi.org/10.1007/bf02371192Google Scholar

10.

Johnen , B., and Sauerbeck , D. 1977. A tracer technique for measuring growth, mass and microbial breakdown of plant roots during vegetation. Ecol. Bull. 25: 366–373. Google Scholar

11.

Lal , R. 2004. Soil carbon sequestration impacts on global climate change and food security. Science, 304(5677): 1623–1627. https://doi.org/10.1126/science.1097396Google Scholar

12.

Malhi , S.S., Nyborg , M., Goddard , T., and Puurveen , D. 2012. Long-Term tillage, straw management, and nitrogen fertilization effects on organic matter and mineralizable carbon and nitrogen in a Black Chernozem soil. Commun. Soil Sci. Plant Anal. 43(20): 2679–2690. https://doi.org/10.1080/00103624.2012.711880Google Scholar

13.

Scurlock , J.M.O., and Hall , D.O. 1998. The global carbon sink: a grassland perspective. Global Change Biol. 4(2): 229–233. https://doi.org/10.1046/j.1365-2486.1998.00151.xGoogle Scholar

14.

Wang , X., McConkey , B.G., VandenBygaart , A.J., Fan , J., Iwaasa , A., and Schellenberg , M. 2016. Grazing improves C and N cycling in the Northern Great Plains: a meta-analysis. Sci. Rep. 6: 33190. https://doi.org/10.1038/srep33190. PMID:27616184. Google Scholar
© Her Majesty the Queen in right of Canada 2018. This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.
Arumugam Thiagarajan, Jianling Fan, Brian G. McConkey, H.H. Janzen, and C.A. Campbell "Dry matter partitioning and residue N content for 11 major field crops in Canada adjusted for rooting depth and yield," Canadian Journal of Soil Science 98(3), 574-579, (1 June 2018). https://doi.org/10.1139/cjss-2017-0144
Received: 29 November 2017; Accepted: 7 April 2018; Published: 1 June 2018
JOURNAL ARTICLE
6 PAGES


Share
SHARE
KEYWORDS
azote
carbon
carbone
crop residue
nitrogen
paille
pousse
ARTICLE IMPACT
RIGHTS & PERMISSIONS
Get copyright permission
Back to Top