Letter to the EditorTrade-offevaluation using carbon dioxide equivalent and hazard index of a paddy soil with application of organic liquid fertilizer

Dear Editor,

Rice is an important staple food crop in Asia,but the anaerobic soil conditions in rice fields cause emission of the potent greenhouse gas,methane(CH4)(Le Mer and Roger,2001),and promote the accumulation of arsenic(As)in the grain(Le Mer and Roger,2001;Khanamet al.,2020).To facilitate sustainable agriculture,using wastewater as an organic fertilizer is essential for reducing wastewater treatment-associated energy consumption and the recycling of wastewater nutrients(Zouet al.,2009).However,increases in both CH4emissions and grain As levels because of organic matter supply,which enhances anaerobic conditions and is also a source of CH4,have been reported in rice fields receiving organic liquid fertilizers such as biogas slurry(Jiaet al.,2013;Huanget al.,2014),sewage effluent(Zouet al.,2009),and liquid cattle slurry(Riyaet al.,2015).

The creation of aerobic soil conditions through water management is an effective option for reducing CH4emissions and grain As levels(Linquistet al.,2015),because this results in the inactivation of methanogenic archaea(Le Mer and Roger,2001)and the association of As with soil minerals(Khanamet al.,2020).However,the creation of oxidized soil conditions can also cause an increase in the emission of another potent greenhouse gas,nitrous oxide(N2O),and lead to the accumulation of cadmium(Cd)in rice grains(Araoet al.,2009;Zouet al.,2009).Therefore,it is difficult to determine the optimal water management strategy that minimizes both greenhouse gas emission and toxic metal accumulation in rice grains.

The impact of water management on global warming can be evaluated by integrating both CH4and N2O emissions into carbon dioxide(CO2)equivalent(CO2-eq)using their global warming potentials(GWPs),a widely used approach in the evaluation of greenhouse gas emissions from rice paddy fields(Zouet al.,2009;Linquistet al.,2015).Additionally,hazardous trace elements in rice grains must be lower than the recommended national or international upper limits.However,the trade-offbetween As and Cd accumulation in rice has made it difficult to develop cultivation methodologies that produce rice grains containing low levels of both As and Cd.

The concept of hazard index(HI)has been applied in health risk assessments(USEPA,1986).The HI measures the overall potential risk of non-carcinogenic effects from more than one hazardous trace element.It has been applied in health risk assessments of market milled rice(Qianet al.,2010),soil remediation(Wanet al.,2020),and the application of multi-element-polluted amendments(Zhanget al.,2021).The advantage of using HI to evaluate cultivation methodologies or practices for the reduction of As and/or Cd levels in rice grains is its ability to evaluate the combined adverse effects of several hazardous trace elements.Thus,the application of both CO2-eq and HI would be useful for clarifying an optimal management strategy that involves trade-offs.This study aimed to investigate how water management affects greenhouse gas emissions and human health risks of a paddy soil with application of organic liquid fertilizer in a pot experiment with rice cultivation.

Uncontaminated soil was taken from the plow layer of a paddy field at the Field Museum Hommachi,Field Science Center,Tokyo University of Agriculture and Technology,Fuchu,Tokyo,Japan.The soil samples were sieved to 2 mm and stored at room temperature until use.Total carbon(C),total nitrogen(N),total As,and total Cd contents of the soil were 48.6g kg−1,5.9 g kg−1,11.7 mg kg−1,and 1.5 mg kg−1,respectively.The total As and Cd contents of the commercial chemical fertilizer(80 g kg−1each of NH4-N,P2O5,and K2O)were 3 and 6mg kg−1,respectively.The organic liquid fertilizer,an aerated liquid phase of cattle waste,was obtained from the Menuma-Machi Organic Center(Saitama,Japan).The total N,NH4-N,and dissolved organic C(DOC)concentrations of the liquid fertilizer were 1 513,1 076,and 726mg L−1,respectively.The liquid fertilizer contained 3.3 mg L−1of total Cd,while As was not detected.

Nine pots(30 cm in height and 25 cm in inner diameter)were prepared by first placing 2 kg of gravel in the bottom and then covering the gravel with 9.8 kg of soil at a bulk density of 1.0 g cm−3.Three treatments in triplicate were prepared as follows:i)chemical fertilizer and continuous flooding(CC),ii)organic liquid fertilizer and continuous flooding(OC);and iii)organic liquid fertilizer and water management(OW).Fertilizers were applied at 50 kg NH4-N ha−1as basal fertilizer 12 d before transplanting and 40 kg NH4-N ha−1as topdressing 40 d after transplanting(DAT)for CC and OC and 64 DAT for OW after the midsummer drainage to reduce CH4emissions(Riyaet al.,2015).The total NH4-N application rate(90 kg ha−1)was the recommended rate for rice cultivation in Japan(MAFF,2019a).The total As and Cd applied with the chemical fertilizer were less than 1%of their total amounts in the soil.The soil was amended with 279 g ha−1of total Cd from the organic liquid fertilizer,accounting for 9.31%of the total Cd in the soil.

Three rice seedlings were transplanted into a pot at 0 DAT at 20 hills m−2,and drainage was started at a rate of 1 cm d−1using a peristaltic pump(MINIPULS Evolution,Gilson,USA)connected to the bottom of the pots.During rice cultivation,the pots in CC and OC were continuously flooded at 1–5 cm depth until the final drainage started at 101 DAT.For OW,midsummer drainage started at 57 DAT,and the pots were reflooded at 64 DAT after exposed to the atmosphere for 5 d.Then,the pots were intermittently irrigated from 78 to 101 DAT.During intermittent irrigation,the soil surface was exposed to the atmosphere seven times for 1–4 d each time and a total of 14 d.

During cultivation,soil redox potential(Eh)at the 5 cm depth was measured using an EP-201 platinum electrode(Fujiwara Scientific Co.Ltd.,Japan).Soil pore water at the 5 cm depth was taken with a DIK-300B soil pore water sampler(Daiki Rika,Japan),filtered with a 0.45-µm membrane filter,and analyzed for DOC concentration using a total organic C(TOC)analyzer(TOC 5000 A,Shimadzu,Japan).

The closed-chamber technique was used to measure CH4and N2O fluxes(Riyaet al.,2015).The cumulative CO2-eq emission was calculated using the GWPs of CH4(28 g CO2-eq g−1)and N2O(265 g CO2-eq g−1)(IPCC,2013;Riyaet al.,2015).

Harvested brown rice was digested with HNO3,and As and Cd concentrations were quantified using a 7500c inductively coupled plasma mass spectrometer(Agilent,Japan).Soil samples for As and Cd analyses and fertilizer samples for As analysis were digested with H2SO4-HNO3-HClO4.Fertilizer samples for Cd analysis were digested with HCl-HNO3.Then,As and Cd concentrations in the digested liquid of the soil and fertilizer samples were measured using a ZA3300 atomic absorption spectrometer(Hitachi High-Tech Corporation,Japan).

The potential health risks of Cd and As in harvested grains were evaluated using HI(USEPA,1986;Qianet al.,2010).First,the estimated daily intake(EDI)of each metal was calculated(Qianet al.,2010),assuming 365 d year−1of exposure frequency,81 years of exposure duration(average lifetime of a Japanese male)(MHLW,2017),147 g rice person−1d−1of the daily rice ingestion rate(average Japanese rice consumption rate)(MAFF,2019b),and 62 kg person−1of body weight(BW)(average value of a Japanese male)(MHLW,2018).Then,target hazard quotient(THQ)was calculated to assess the health risks associated with rice consumption(Qianet al.,2010),assuming 350 and 7µg kg−1BW week−1of the provisional tolerable weekly intake of As and Cd,respectively(FAO/WHO,1989,2003).Finally,HI,the potential risk of adverse health effects from a mixture of As and Cd,was calculated by summing the THQ values of As and Cd.An HI value of 1 or lower suggests that product(brown rice)consumption is unlikely to cause adverse non-carcinogenic health effects over a lifetime.

One-way analysis of variance(ANOVA)was performed to identify statistical differences among treatments in cumulative CH4,N2O,and CO2-eq emissions,metal contents in the grain,EDI,THQ,and HI,followed by least significant difference(LSD)test using SPSS 16.0(IBM Corp.,USA).

Although the OC treatment tended to increase CH4flux(Fig.1),organic liquid fertilizer did not significantly increase the cumulative CH4emissions when compared with CC(Table I).The OC treatment tended to increase DOC concentration in soil pore water and decreased soil Eh when compared with the CC treatment(Fig.1).The amount of organic C supplied by the organic liquid fertilizer,61 kg C ha−1(34 and 27 kg C ha−1from basal fertilization and topdressing,respectively),might not have been sufficient to decrease soil Eh and increase the cumulative CH4emissions.A similar result was reported for a rice field receiving digestate at a low application rate(Huanget al.,2014).Furthermore,the Eh in CC showed that the soil was relatively aerobic throughout the experiment.The lowest soil Eh in OC was−154 mV,which did not reach the threshold value for active methane emission(−200 mV)(Le Mer and Roger,2001).The relatively high percolation rate applied in our pots(1 cm d−1)would cause oxygen influx into the soil and retard the development of reducing conditions(Yagiet al.,1998).This may explain why there was no significant difference in CH4emission between CC and OC.

Water management did not affect the cumulative CH4and N2O emissions under organic liquid fertilizer application(Table I).Midsummer drainage and intermittent irrigation increased soil Eh and N2O flux but decreased CH4flux(Fig.1).However,the lack of significant difference in CH4emission between OC and OW suggests that the extent of soil aeration(1–5 d for one drainage event)in OW may not be sufficient to reduce CH4emission.According to Linquistet al.(2018),the dry-down period plays a more important role than the number of dry-downs in CH4emission reduction.There was no significant difference in N2O emission between OC and OW.According to Nishimuraet al.(2004),N2O fluxes during drainage periods were comparatively low at an application rate of 100 kg N ha−1or less of chemical fertilizer.Thus,the low N application rate and the limited number of dry-downs resulted in the low level of cumulative N2O emission.

Fig.1 Temporal changes in CH4 and N2O fluxes,soil redox potential(Eh),and dissolved organic carbon(DOC)concentration in soil pore water in the different treatments of the pot experiment with cultivation of rice.CC=chemical fertilizer and continuous flooding;OC=organic liquid fertilizer and continuous flooding;OW=organic liquid fertilizer and water management.The vertical solid and dotted arrows indicate topdressing under continuous flooding(CC and OC)and water management(OW)conditions,respectively.The horizontal solid and dotted arrows indicate midsummer drainage and intermittent irrigation,respectively,in OW.Error bars are standard deviations(n=3).

TABLE ICumulative CH4 and N2O emissions and their CO2 equivalent(CO2-eq)in the different treatments of the rice cultivation pot experiment and summary of one-way analysis of variance(ANOVA)

Cumulative CO2-eq emissions were not significantly different between treatments(Table I).The CH4emission accounted for approximately 100% of the CO2-eq emission in CC and OC and approximately 83%in OW.Thus,water management decreased the contribution of CH4and increased that of N2O to CO2-eq emission when compared with continuous flooding.More work is needed to clarify the extent of the influence of drainage duration and percolation rate on CH4and CO2-eq emissions as discussed above.

Grain yield was not significantly different between treatments(Table II).In addition,the total grain As and Cd levels were lower than their upper acceptable limits of 0.35 and 0.4 mg kg−1grain,respectively(Codex Alimentarius Commission,2005,2016).Thus,organic liquid fertilizers can be substituted for chemical fertilizers.

The As content in rice grain was not significantly different between treatments(Table II).Rice plant As uptake increases when the soil becomes anaerobic from flooding and/or organic matter application,which leads to As mobilization(Araoet al.,2009;Jiaet al.,2013).In contrast,water management increases soil Eh and reduces As accumulation in rice grains(Araoet al.,2009).The ineffectiveness of organic liquid fertilizer application and water management in reducing As accumulation in this study might be attributed to the relatively low organic liquid fertilizer application rate and short dry-down periods as noted above.

The total Cd content in OC was comparable to that in CC,whereas the Cd content in OW was significantly higher than those in CC and OC(Table II).Therefore,water management enhanced Cd accumulation in rice grains under organic liquid fertilizer application.The higher level of soil oxidation with water management in OW compared with OC was reflected in the increased soil Eh in this treatment(Fig.1).This resulted in the dissolution of soil Cd and itsaccumulation in rice grains in OW.The addition of exogenous Cd to the soil by organic liquid fertilizer(approximately 9%of soil Cd)would also increase grain Cd accumulation in OW compared with OC,where the soil was more reduced and Cd would be more prevalent as insoluble CdS(Khanamet al.,2020).

TABLE IIBrown rice yields and the contents,estimated daily intakes(EDIs),target hazard quotients(THQs),and hazard indices(HIs)of As and Cd in different treatments of the rice cultivation pot experiment and summary of one-way analysis of variance(ANOVA)

Although the trade-offbetween As and Cd accumulation in rice grains with water management is well known(Araoet al.,2009),few reports are available on how such trade-offaffect health risk.The calculated EDIs of grain As and Cd(Table II)were comparable to those previously found in China(Qianet al.,2010).The significantly higher EDI,THQ,and HI in OW than in OC suggest that water management increased the potential Cd consumption and non-carcinogenic health risk.The THQ and HI values for all treatments were less than 1,which suggests no noncarcinogenic health risk under the study conditions.The contribution of Cd accounted for more than 90%of the HI,even though the total Cd contents in brown rice were lower than those of As.This resulted from the much lower provisional tolerable weekly intake of Cd than As.Similar results were also reported by Qianet al.(2010),who investigated the THQs and HIs of milled rice in the Chinese market.These results suggest that Cd control is more important than As in terms of non-carcinogenic health risk in the studied soil.This study revealed that HI is useful to quantify the health risk of the grain as well as in discussing the contribution change of each toxic metal in the health risk trade-offcaused by water management.

Given the much higher application of organic liquid fertilizer(83.6t ha−1)compared with chemical fertilizer(1.13 t ha−1),Cd accumulation from the continuous application of organic liquid fertilizer,as well as the CO2emission from fossil fuel consumption for liquid fertilizer transport and application,should be taken into account.Further studies are needed to evaluate the life-cycle analysis of greenhouse gas emissions and the combined carcinogenic and non-carcinogenic human health risks.

ACKNOWLEDGEMENTS

This study was supported by the Yashima Environment Technology Foundation and partly by a Grant-in-Aid for Young Scientists(No.20K19994),Scientific Research(C)(No.17K00594),and Fostering Joint International Research(B)(No.18KK0288)from the Japan Society for the Promotion of Science(JSPS).We thank Drs.Lesley Benyon and Catherine Dandie from Edanz Group(https://jp.edanz.com/ac)for editing the draft of this manuscript.