Nutritional and environmental effects on biological nitrogen fixation in soybean: A meta-analysis
Introduction
Nitrogen (N) is usually the most limiting nutrient for crop productivity (Vitousek and Howarth, 1991). Despite the high N content in the atmosphere, N is not present in soil parent material (Hedin et al., 2009). Therefore, soil N inputs to the biosphere relies on organic matter, synthetic fertilizer applications, and biological N2 fixation (BNF) via nitrogenase enzyme activity (Galloway et al., 2008; Vitousek et al., 2013). Biological fixation is accomplished in part by some blue-green algae, but predominantly by free-living or symbiotic bacteriae. The last group (i.e. Rhizobia) establish a mutualistic association with legumes species, fixing N from the atmosphere that is used by both the bacteria and the host (Herridge et al., 2008). Rhizobia fixes more than 30 × 109 kg N every year in 250 million hectares worldwide under legume cultivation, (Kinzig and Socolow, 1994) and is predicted to increase to 50 × 109 kg N year−1 by 2050 (Galloway et al., 2004). Global legume cultivation is dominated by soybean, a key component of global human and livestock diet (Brumm and Hurburgh, 1990).
Soybean seed yield is positively correlated with total N uptake (Salvagiotti et al., 2008; Rotundo et al., 2014; Santachiara et al., 2017), and 44–72% is usually accounted by BNF (Ciampitti and Salvagiotti, 2018). Maximum BNF occurs when the seed filling period begins (increasing seed demand) and decreases through physiological maturity (Zapata et al., 1987). Even though fixation provides a high proportion of total N demand, high rates of BNF are usually not required to maximize soybean seed yield in soils with adequate mineral N supply because of a trade-off between both N sources (Santachiara et al., 2017). However, high rates of BNF are necessary to maximize yields in soils with low to medium N fertility and to reach near-to-neutral soil N balances after soybean cultivation (Collino et al., 2015; Santachiara et al., 2017). In this context, understanding the factors affecting BNF is especially relevant for both soybean production and long term sustainability of agriculture.
Biological N2 fixation is considered an extremely sensitive process influenced by nutrient and environmental conditions (Graham, 1992; Purcell et al., 1998; Zahran, 1999; King and Purcell, 2005; Liu et al., 2011). There are several published studies focused on quantifying the effect of multiple factors on BNF. Previous reports have shown that BNF is reduced by high soil nitrate availability (Eaglesham et al., 1983; Vessey and Waterer, 1992; Arrese-lgor et al., 1997), soil water deficit (Minchin and Pate, 1975; Bennett and Albrecht, 1984; Sall and Sinclair, 1991), and long periods of soil water saturation (Rajaramamohan-Rao, 1976; Bennett and Albrecht, 1984;). Likewise, departures from temperatures ranging from 22 to 25 °C resulted in reductions in BNF (Lindemann and Ham, 1979). Non-optimal temperatures, water stress, high soil N availability, and soil pH have been reported to have inhibitory effects (Sprent, 1971; Pankhurst and Sprent, 1975; Munévar and Wollum, 1981; Graham, 1992; Bordeleau and Prévost, 1994; Zhang et al., 1996; Zahran, 1999; Hungria and Vargas, 2000; Prudent et al., 2016). In contrast, radiation, carbon dioxide concentration (Murphy, 1986), and nutrient availability have been reported to have enhancing effects (Graham, 1992; Zahran, 1999; Divito and Sadras, 2014). While the relationship between these factors and BNF has been analyzed before, a quantitative appraisal summarizing effect sizes and factors affecting these effect sizes is currently lacking. Previous meta-analysis on N, P, K, and S effects on soybean (Divito and Sadras, 2014) focused on mostly on nodulation (both number and weight) but not on direct estimations of BNF as in the present study.
Meta-analyses are designed to integrate information from various experiments and conditions evaluating common treatment effects (Curtis and Wang, 1998). Meta-analyses are particularly informative since they provide effect sizes estimates, calculated as the response of a treatment relative to an untreated control. Different meta-analysis has been successfully used to estimate effects of environmental conditions of soybean seed quality (Rotundo and Westgate, 2009), soybean responses to increases in CO2 (Ainsworth et al., 2002), and ozone concentration (Morgan et al., 2003). Using this approximation, we seek to summarize and quantify overall soybean BNF responses to environmental and nutritional variables. Therefore, the objective of this study was to perform a meta-analysis summarizing published data regarding the effect of nutrient addition and environmental variation on soybean BNF. Interactions among treatment timing, rate and intensity were also explored.
Section snippets
Database compilation and treatments partitioning
Published articles about soybean BNF were searched using Scopus®, Science Direct® and Google Scholar® databases of peer reviewed literature. Employed search criteria included a hierarchy structure: (a) soybean as object of study (“common” and/or “scientific name”), (b) nutrient (application or fertilization) and environmental condition (e.g. water deficit) as intervention (various synonymous were used for each search, for instance “water stress" or “drought” or "water deficit" or "water
Response of BNF to nutrient addition
For the unweighted analysis, N fertilization reduced BNF by 49% and by 70% as estimated for NDFA and ARA, respectively (Fig. 1a). Relative reduction in NDFA was 44% in field experiments while it was more than 70% in greenhouse experiments (Qt = 17.2; P < 0.001; Sup. Table I.1). This difference between experimental setting was not observed when BNF was estimated as ARA (Qt = 0.4; n.s.). Increased rate of N application in field experiments reduced BNF when estimated by both NDFA and ARA (Fig. 1b;
Discussion
Biological N2 fixation has been reported as a very sensitive process affected by several environmental factors. Even when the relative impact of N nutrition on soybean BNF has been extensively analyzed (Salvagiotti et al., 2008; Ciampitti and Salvagiotti, 2008), the effects of other environmental conditions and nutrients other than N, has not been extensively summarized. The current meta-analysis integrates numerous data sources to attain general response patterns of BNF to soil nutrient
Conclusions
The current meta-analysis reported the different effects of nutrient availability and environmental factors on BNF in soybean. Results from this work would serve as possible starting point for generating new insights and additional hypotheses regarding BNF. Environmental (water deficiency, flooding and temperature) and N addition during initial vegetative stages had the most detrimental effects of BNF. This pattern might be extended to other factors affecting negatively BNF not analyzed here.
Acknowledgments
This research was partially funded by Agencia Nacional de Promoción Científica y Tecnológica, Argentina (PICT2011-1292). Authors wanted to thanks Ryan McCormick for insightful comments on the manuscript. G. Santachiara & J.L. Rotundo are members of CONICET, The Argentina National Research Council.
References (80)
- et al.
How do phosphorus, potassium and sulphur affect plant growth and biological nitrogen fixation in crop and pasture legumes? A meta-analysis
Field Crop Res.
(2014) - et al.
Nitrogen fixation, growth and yield of soybean grown under saturated soil culture and conventional irrigation
F. Crops Res.
(1993) - et al.
Environmental factors affecting N2 fixation in grain legumes in the tropics, with an emphasis on Brazil
Field Crop Research
(2000) - et al.
The response and recovery of nitrogen fixation activity in soybean to water deficit at different reproductive developmental stages
Environ. Exp. Bot.
(2013) - et al.
Assessment of the quality of meta-analysis in agronomy
Agric. Ecosyst. Environ.
(2012) - et al.
Nitrogenase: standing at the crossroads
Curr. Opin. Chem. Biol.
(2000) - et al.
Meta-analysis of environmental effects on soybean seed composition
Field Crop Research
(2009) - et al.
Nitrogen uptake, fixation and response to fertilizer N in soybeans: a review
Field Crop Res.
(2008) - et al.
Plant stand, nodulation and seed yield in soybean as affected by phosphate fertilizer placement, source and application method
Eur. J. Agron.
(2013) Water and nitrogen limitations in soybean grain production. I Model development
Field Crop Research
(1986)