Complementary effect of zoo compost with mineral nitrogen fertilisation increases wheat yield and nutrition in a low-nutrient soil

Muhammad S.A.KHANLynette K.ABBOTTZakaria M.SOLAIMAN*Peter R.MAWSONIan S.WAITE and Sasha N.JENKINS

1 School of Agriculture and Environment,Institute of Agriculture,The University of Wester n Australia,Per th WA 6009(Australia)

2 Perth Zoo,Depar tment of Biodiversity,Conser vation and Attractions,Perth WA 6151(Australia)

3 Agricultural Officer(Headquar ters),Directorate,Soil Fertility Research Institute,Agriculture Department,Gover nment of the Punjab,Lahore 53700(Pakistan)

ABSTRACT Excess nitrogen(N)fertiliser use in agriculture is associated with water pollution and greenhouse gas emissions.While practices and programs to reduce N fertiliser application continue to be developed,inefficient fertiliser use persists.Practices that reduce mineral N fertiliser application are needed in a sustainable agricultural ecosystem to control leaching and gaseous losses for environmental management.This study evaluated whether fully or partially replacing mineral N fertiliser with zoo compost(Perth Zoo)could be a good mitigation strategy to reduce mineral N fertiliser application without affecting wheat yield and nutrition.To achieve this,a glasshouse experiment was conducted to assess the complementary effect of zoo compost and mineral N fertiliser on wheat yield and nutrition in a sandy soil of southwestern Australia.Additionally,a chlorophyll meter was used to determine whether there was a correlation between chlorophyll content and soil mineral N content,grain N uptake,and grain protein content at the tillering(42 d after sowing(DAS))and heading(63 DAS)growth stages.The standard practice for N application for this soil type in this area,100 kg ha-1,was used with a soil bulk density of 1.3 g cm-3 to calculate the amount of mineral N(urea,46%N)and Perth Zoo compost(ZC)(0.69%N)for each treatment.Treatments comprised a control(no nutrients added,T1),mineral N only(100 kg N ha-1,T2),ZC only(100 kg N ha-1,T7),and combinations of mineral N and ZC at different rates(mineral N at 100 kg N ha-1+ZC at 25 kg N ha-1(T3),mineral N at 75 kg N ha-1+ZC at 25 kg N ha-1(T4),mineral N at 75 kg N ha-1+ZC at 50 kg N ha-1(T5),and mineral N at 50 kg N ha-1+ZC at 50 kg N ha-1(T6)).The T6 treatment significantly increased grain yield(by 26%)relative to the T2 treatment.However,the T7 treatment did not affect grain yield when compared to the T2 treatment.All treatments with mineral N and ZC in combination significantly improved the 1 000-grain weight compared to the T2 treatment.Chlorophyll content was better correlated with soil mineral N content(r=0.61),grain N uptake(r=0.62),and grain protein content(r=0.80)at heading(63 DAS)than at tillering(42 DAS).While ZC alone could not serve as an alternative to mineral N fertiliser,its complementary use could reduce the mineral N fertiliser requirement by up to 50%for wheat without compromising grain yield,which needs to be verified in the field.Chlorophyll content could be used to predict soil mineral N at the heading stage,and further studies are warranted to verify its accuracy in the field.Overall,the application of ZC as part of integrated nutrient management improved crop yield with reduced N fertiliser application.

Key Words: chlorophyll meter SPAD measurement,integrated nutrient management,nitrogen use efficiency,reduced N fertiliser application,wheat yield

INTRODUCTION

With the world’s population expected to reach 10 billion by 2050,an estimated 50%increase in agricultural production compared with the 2013 values is required to meet future food demands and provide food security(Fisket al.,2015;FAO,2017).This trend in global population increase has led to a nine-fold increase in mineral nitrogen(N)fertiliser use globally since the 1960s,and a further increase of up to 40%—50%was predicted by 2050(Suttonet al.,2013).The role of N in terms of increasing photosynthetic leaf area by accelerating chlorophyll production,promoting tillering,and enhancing the content of grain protein is well documented.However,these N functionalities depend upon the availability of Nat specific developmental stages of wheat growth(Langer and Liew,1973;Anderson and Garlinge,2000).In Australia,the use of N fertiliser increased significantly during the 1980s and the 1990s primarily to achieve high wheat yield and high N content in wheat grain(Angus,2001).However,the frequent application of N fertiliser has disturbed the natural balance of N and P biogeochemical cycles in soil,causing an estimated financial damage of$170 billion annually to global ecosystem services(Suttonet al.,2013).Furthermore,excessive nutrient leaching(Mantoviet al.,2005;Paramashivamet al.,2016),especially in degraded and sandy soils,can lead to eutrophication of nearby aquatic systems and fish kills(Willénet al.,2017).The availability of some nutrients(e.g.,P and K)is at stake due to limited reserves,and other nutrients(e.g.,N)are becoming expensive to manufacture because of the high energy requirement for their production(Rengel,2013).Thus,there is an urgent need for more sustainable N fertiliser use without compromising crop performance.

Soils of southwestern Australia are highly weathered and commonly sandy in texture with typically low-N status(Moore,2001;Jenkinset al.,2016).Consequently,these soils require considerable nutrient inputs to sustain the productivity of major crops(Anderson and Garlinge,2000;Moore,2001;Yamaguchiet al.,2004).Wheat is the main crop grown in this region,accounting for almost half of the total Australian cropland(ABARES,2016).However,the average wheat yield(1.7 t ha-1)between 1996 and 2010 was relatively low when compared with the water-limited yield potential of 3.5 t ha-1(Hochmanet al.,2016),suggesting that one or more factors constrain productivity.One of the biophysical factors widening the gap between attainable and actual wheat yield is poor nutrient management in sandy soils of southwestern Australia.Despite the high use of N fertiliser,the N use efficiency(NUE)is low in Australia(Chenet al.,2009).Improvement in NUE can reduce mineral N consumption without affecting wheat yield.Progressive farming practices in Australia have been reported to produce up to 82%of the water-limited yield potential with the adoption of reduced tillage and timely application of N fertiliser(Tilmanet al.,2002;van Reeset al.,2014).Also,grain growers are increasingly using organic amendments(manure and compost)to reduce their N fertiliser application and address soil constraints,especially in South Australia(Quilty and Cattle,2011).Adding organic matter to soil has been shown to increase soil quality and crop performance(Jenkinset al.,2009;Solaimanet al.,2019)and to improve the resilience,structure,and function of soil(Abbottet al.,2018).

Zoo animals,particularly herbivores,produce large volumes of dung that can be difficult to dispose of.A solution is to compost the dung into a high-quality soil amendment(Pérez-Godínezet al.,2017).Perth Zoo compost(ZC)is a recently developed resource for the Perth Zoo in Western Australia(Perth Zoo,2017).However,its potential in terms of bioavailability of N and enhancement of NUE has not been evaluated.The critical aspect of considering ZC as an additional source of N is the timely release of N through mineralisation.

Methods for estimating potentially available N released after mineralisation of organic soil amendments(compost or manure)involve either aerobic or anaerobic incubation of soil over varying time spans or chemical extraction of potentially available soil organic N.These traditional methods are time consuming and do not represent actual field conditions(Curtin and McCallum,2004;Roset al.,2011).Handheld chlorophyll meters,including SPAD(soil plant analysis development),have been shown to provide a non-destructive and rapid technique for estimating soil mineral N(Debaekeet al.,2006;Yuanet al.,2016;Ravieret al.,2018;Fiorentiniet al.,2019).A potential benefit of this method is that N deficiency could be identified in the field and rectified through a top-up of Nfertiliser to improve yield with minimal N loss to the environment.A chlorophyll meter measures the transmittance of red and infra-red radiation emitted through leaf and gives a relative value for soil N status(Follettet al.,1992;Spaneret al.,2005;Poole and Hunt,2014).If chlorophyll measurements are correlated with leaf N,an empirical relationship between leaf N and soil mineral N could be developed to estimate soil mineral N at a specific plant growth stage(Spaneret al.,2005;Rayment and Lyons,2011;Poole and Hunt,2014).Debaekeet al.(2006)and Fiorentiniet al.(2019)demonstrated a strong correlation between chlorophyll content in durum wheat and soil N status.However,to our knowledge,chlorophyll meters have not been evaluated in crops receiving organic N fertiliser inputs or combinations of organic and mineral fertilisers.

This study aimed to test the hypotheses that i)combining ZC and mineral N fertiliser can reduce N fertiliser requirements for wheat in a sandy soil without compromising wheat yield under glasshouse conditions and ii)a chlorophyll meter can be used to predict soil mineral N content,grain N uptake,and grain protein content at tillering and heading for wheat grown in a sandy soil.

MATERIALS AND METHODS

Experimental design

The experiment was set up in a complete randomised block design with seven treatments and three replicates per treatment.Wheat(Triticum aestivumL.)was sown in pots(1.63 kg air-dried soil)under glasshouse conditions at The University of Western Australia(UWA),Crawley Campus,Perth,Australia(31°59′03.4 S,115°49′10.0 E).The districtrecommended N dose of 100 kg N ha-1was used with a soil bulk density of 1.3 g cm-3to calculate the amount of mineral N(urea,46%N)and ZC(0.69%N)for each treatment.Mineral N and ZC were applied at the following rates:no nutrients added(control,T1),urea only at 100 kg N ha-1(T2),urea at 100 kg N ha-1plus ZC at 25 kg N ha-1(T3),urea at 75 kg N ha-1plus ZC at 25 kg N ha-1(T4),urea at 75 kg N ha-1plus ZC at 50 kg N ha-1(T5),urea at 50 kg N ha-1plus ZC at 50 kg N ha-1(T6),and ZC only at 100 kg N ha-1(T7)(Table I).A basal dose of phosphorus was applied at a rate of 25 kg P ha-1for all treatments except T1.

TABLE I Soil amendments and their nutrient equivalents in different treatments of this study

Soil sampling and analysis

The soil(top 0—10 cm)used for the glasshouse experiment was collected from UWA Farm Ridgefield at Pingelly(32°30′23 S,116°59′31 E).The farm is situated in the semiarid wheatbelt region,which produces approximately 40%of Australia’s grain.The site is characterised by an average annual rainfall of 445 mm(57.8 rainy days),evaporation rate of approximately 1 700 mm,and temperature ranges of 6—16°C in winter and 15—31°C in summer.The parent rock type is colluvium over truncated,deeply weathered lateritic profile(granite as original parent rock).The land has been used historically for cropping and pasture(Jenkinset al.,2016).The soil samples were dried,sieved(2 mm),and thoroughly mixed to obtain a homogenised sample prior to measuring the electrical conductivity(EC),pH,total organic carbon(TOC),total carbon(TC),total N(TN),organic matter(OM),and particle size distribution.Soil EC and pH were measured in a soil-water(1:5)suspension.Soil TC and TN contents were determined with a CN combustion analyser(vario MACRO;Elementar,Germany).Soil C:N ratio was determined by dividing the TC by TN.Soil TOC was determined using the Dumas high-temperature combustion method,and OM was calculated by multiplying the TOC by a factor of 1.72.Analysis of each parameter was carried out in triplicate following standard methods as described by Rayment and Lyons(2011).Mean values and standard errors were calculated for all treatments.

Soil samples for ammonium N(NH+4-N)and nitrate N(NO-3-N)analysis were collected after each harvest,thoroughly mixed,labelled,sieved through a 2-mm sieve,and stored in a cold room at 4°C until analysed.Soil mineral N(sum of NO-3-N and NH+4-N)was determined spectrophotometrically(Hood-Nowotnyet al.,2010;Wanget al.,2017).Soil moisture was determined gravimetrically,and results were reported on a dry weight basis.Data interpretations were made according to Hazelton and Murphy(2016).

The soil is as a Tenosol according to the Australian Soil Classification(Isbell,2002)and a sandy loam based on particle size analysis(83.4%sand,6.5%silt,and 10.1%clay)according to the Australian Soil and Land Survey Field Handbook(McDonaldet al.,1998).It is moderately acidic(pH 5.81)with low EC(0.14 dS m-1).The TOC(21 g kg-1)and TN(1.8 g kg-1)contents were relatively low,typical for the dryland soils of Western Australia.The C:N ratio of the soil was 10:1,suggesting that it had poor structural condition and stability(Hazelton and Murphy,2016).

Quality assessment of ZC

Perth Zoo compost is a high-quality compost that complies with the AS 4454-2012 production protocol(Standards Australia,2012)for compost,soil conditioners,and mulches(Perth Zoo,2017).To comply with the AS 4454-2012 production protocol,ZC is produced by maintaining the core temperature of the compost mass at 55°C or higher for 15 d or longer,with the windrow being turned for a minimum of five times,in a dedicated stream offsite and not mixed with material sourced from any other site.The raw materials used to produce ZC are composed of 34%garden waste(tree and shrub pruning and course wood chips),52%herbivorous animal waste(primarily derived from large herbivores such as Asian elephants and white rhinoceros),and 14%carnivorous animal waste(faeces and soiled bedding),and their quantities vary with season.The ZC used in this study was analysed for essential macronutrients(N,P,K,Ca,Mg,and S)and micronutrients(Zn,Cu,Fe,Mn,and B)using inductively coupled plasma optical emission spectroscopy(ICP-OES,Model 5300DV;Perkin Elmer,USA).The TC and TN contents were determined using a CN combustion analyser(vario MACRO;Elementar,Germany),and the C:N ratio of the ZC was calculated by dividing the TC by TN.Analytical methods described in Standards Australia(2012)were followed for chemical analysis of compost.

The macronutrient analysis revealed that the ZCconsisted of 6.9 g kg-1total N,2.1 g kg-1total P,3.8 g kg-1total K,11.3 g kg-1total Ca,1.7 g kg-1total Mg,and 1.1 g kg-1total S.The compost also contained a range of micronutrients(Zn,Cu,Fe,Mn,and B of 102,21,5 929,112,and 32 mg kg-1,respectively).The C:Nratio,generally used to evaluate compost quality(Sharmaet al.,2012),of the ZC was low(11:1),with 76.5 g kg-1TC,indicating that decomposition of the compost is likely to proceed at a high rate under environmental conditions(Hazelton and Murphy,2016).We would also expect the low C:N ratio of the compost to favour N mineralisation over N immobilisation in soil(Fisket al.,2015).In addition,the soil had a low C:N ratio,further stimulating N mineralisation.Therefore,the compost would break down quickly and contribute to plant nutrition with respect to macro-and micronutrients.

Glasshouse conditions

Soil moisture was maintained at 80%of the field capacity throughout the experiment.The glasshouse temperature fluctuated between 8.3 and 24.3°C with an average of 17.3°C during the experiment,whilst the average relative humidity during this period was 69.8%.Wheat was harvested at the tillering(42 d after sowing(DAS)),heading/flowering(63 DAS),and ripening(92 DAS)stages of development.

Correlation analysis

The SPAD(chlorophyll meter SPAD-502,Konica Minolta,Japan)measurements were taken at three points in fully emerged leaves:the tip,middle,and near the base of the leaf at tillering(42 DAS)and heading(63 DAS).The mean value was used to assess the correlation between SPAD measurements and soil mineral N content,grain N uptake,and grain protein content.

Shoot and root analyses

Aboveground biomass and corresponding root mass material were oven dried at 70°C,and the mean shoot and root oven-dry weights of three replicates was determined.Root:shoot dry weight ratios were calculated,and shoot samples were ground in a mill and analysed for TNusing a CN analyser following the Dumas high-temperature combustion method(Rayment and Lyons,2011).Three replicates of each sample were analysed in consecutive order to minimise the carryover effect of the previous sample.Standards and blanks were repeated in the analysis after every 10th sample.

Nitrogen use efficiency parameters

Both grain and straw yields were recorded after wheat harvest at ripening(92 DAS)by drying the samples at 70°C for three days.Grain N and straw N contents were determined using the Dumas high-temperature combustion method(Rayment and Lyons,2011).Grain N uptake(grain yield×grain Ncontent),straw Nuptake(straw yield×straw N content),N uptake(or recovery)efficiency(total N uptake(grain and straw)/total N applied),N physiological efficiency(grain yield/total N uptake),and N agronomic efficiency(grain yield/total N applied)were determined by following the methodologies described by Jenkinson and Smith(1988)and Gagnonet al.(1997).Total amount of N applied to a pot was calculated by taking the sum of soil mineral N(NH+4-N and NO-3-N)measured in the control and the amount of N applied as a treatment to that respective pot.

Wheat quality parameters

Thousand-grain weight,grain N content,and grain protein content were considered as quality parameters.Grain protein content was calculated by multiplying the grain N content by 5.83(Baker,1979).Thousand-grain weight was calculated based on the grain weight per pot and the number of grains per pot.

Statistical analysis

Analysis of variance(ANOVA)was performed to determine the effects of treatments on wheat yield parameters.To determine which treatments were significantly different,Tukey’s test was performed for comparison of means.All statistical analyses were performed using the R statistical package(Ihaka and Gentleman,1996).

RESULTS

Nitrogen uptake and biomass production

Plant biomass(shoot plus root),root:shoot dry weight ratio,and shoot N content did not differ among treatments at the tillering(42 DAS)and heading(63 DAS)stages(Fig.1,Table II).However,root:shoot dry weight ratio and shoot N content were higher at tillering and decreased by the heading stage in all treatments.At heading,the T3 treatment produced significantly higher shoot N(28.13 g N kg-1DM)compared with the T1 treatment.Plant biomass at heading was significantly higher in the T5 treatment than in the T1 treatment.The T7 treatment produced significantly lower grain N content compared with the T2 treatment and all combination treatments.

Fig.1 Complementary effect of Perth Zoo compost with mineral N on wheat biomass production at the tillering(42 d after sowing(DAS))(a)and heading(63 DAS)(b)stages on a sandy soil under glasshouse conditions.Bars with the same letter are not significantly different at P≤0.05.See Table I for details of the treatments.CK=control.

Wheat yield components and quality parameters

All combination treatments(T4,T5,and T6)resulted in a significantly higher grain N uptake than the T2 and T7 treatments(Table III).The T2 treatment produced the lowest grain yield among all treatments.The T2 treatment produced a significantly lower 1000-grain weight compared to all the other treatments.For grain protein content,all combination treatments produced similar grain protein content;however,the T7 treatment resulted in significantly lower grain protein content compared to all the other treatments.

Nitrogen use efficiency parameters

The T7 treatment had the lowest N uptake efficiency among all the treatments(Table IV).Lower grain yield and N agronomic efficiency were observed in the T2 treatment than in all the other treatments except T1.All combination treatments had higher N uptake efficiency,N physiological efficiency,and N agronomic efficiency compared to the T2 treatment(Tables III and IV).

SPAD measurements

The SPAD(chlorophyll meter)measurements were significantly correlated with soil mineral N(r=0.61)and grain N uptake(r=0.62)at heading(Table V,Fig.2).A stronger correlation was observed between the SPAD measurements and grain protein content(r=0.80)at heading.The SPAD measurements explained 38%of the variations in soil mineral N,39%of variation in grain N uptake,and 80%of variation in grain protein content.

Fig.2 Linear regression models between SPAD-502 measurements and soil mineral N(a),grain N uptake(b),and grain protein content(c)at the heading(63 d after sowing)stage of wheat grown under glasshouse conditions.Grey shades represent standard errors.

TABLE II Complementary effect of Perth Zoo compost with mineral N on shoot,straw,and grain N contents at the tillering(42 d after sowing(DAS)),heading(63 DAS),and ripening(92 DAS)stages of wheat grown on a sandy soil under glasshouse conditions

TABLE III Complementary effect of Perth Zoo compost with mineral N on N uptake and yield components a)of wheat grown on a sandy soil under glasshouse conditions

DISCUSSION

Nitrogen use efficiency and agronomic parameters

All treatments had a similar effect on plant biomass and shoot Ncontent at tillering of wheat,demonstrating adequate N availability in all treatments at tillering.However,the lowest plant biomass at heading and the smallest grain N content at ripening were observed in the treatment with ZC only(T7),which could be related to N deficiency at heading and ripening.This implies that the ZC applied singly did not fulfil the total N requirement of wheat,possibly owing to incomplete mineralisation.Generally,composted organic materials release N at much slower rates(1%—3%of total N year-1)than mineral N fertiliser because they have higher C:N ratio and organic N content(Al-Batainaet al.,2016).All combination treatments increased N uptake and grain yield.Iqbalet al.(2017)observed a similar trend in maize,where a mixture of compost and mineral N increased the yield more than the compost alone.Montemurro(2009)also found that treatments receiving a combination of mineral N and compost produced significantly higher grain yield and N uptake.Similarly,Abediet al.(2010)and Demelashet al.(2014)reported an increase in wheat grain yield when mineral N was applied in combination with compost.Treatment with a 50%reduction in mineral N(T6)did not reduce grain N content and grain yield.Thus,the amount of mineral N applied to the sandy soil investigated here can be reduced by up to 50%of its recommended dose through the addition of an equivalent amount of ZC N.Demelashet al.(2014)also reported a 50%reduction in the use of inorganic fertiliser when combined with compost.However,the amount of compost that can be substituted for mineral N fertiliser can differ and is dependent upon the quality of compost used.Several long-term experiments conducted in different agroecological zones across the globe inferred that crop yields from treatments of organic amendment and synthetic fertiliser were comparable when the total nutrient supply was equivalent(Celestinaet al.,2019).The recommended dose of mineral N can be derived either from organic amendments or synthetic fertilisers(Sistaniet al.,2017;Celestinaet al.,2018).

TABLE IV Complementary effect of Perth Zoo compost on N use efficiency parameters a)of wheat grown on a sandy soil under glasshouse conditions

TABLE V Correlation coefficient(Pearson)values between SPAD-502 measurements and soil mineral N(NO-3-N and NH+4-N),grain Nuptake,and grain protein content recorded at the tillering(42 d after sowing(DAS))and heading(63 DAS)stages of wheat grown on a sandy soil under glasshouse conditions with application of mineral N only,Perth Zoo compost only,or a mixture of Perth Zoo compost and mineral N

Plants receiving a combination of mineral N and ZC exhibited an increased uptake of N and showed the best N uptake efficiency.The treatment with ZC only(T7)showed the lowest N uptake efficiency among all treatments except the control(T1).The lowest grain N content and low N uptake efficiency in the T7 treatment indicate that ZC alone cannot fulfil the total nutrient requirements of a crop in its first year of application owing to the slow release of nutrients and incomplete mineralisation(Al-Batainaet al.,2016).

Complementary effects of composts improve soil structure and reduce nutrient losses through leaching(Abbottet al.,2018).Sandy soils are more prone to losses of plant nutrients owing to leaching(Knops and Tilman,2000).Compost provides additional exchange sites to adsorb nutrients,which explains why the combination treatments had higher N uptake and N uptake efficiency(Abbottet al.,2018).

Despite having the lowest N uptake efficiency,the treatment with ZC only(T7)exhibited the highest Nphysiological efficiency,and hence N agronomic efficiency of this treatment was not different from those of the other treatments.This could be explained in the light of Mitscherlich’s law of diminishing returns(Ferreiraet al.,2017).Typically,crop yields have linear responses to increasing dose of a nutrient(e.g.,N)until the nutrient dose reaches a critical value or sufficiency level;afterwards,the yields remain constant.At this point,another factor becomes limiting to crop yield.Nitrogen in the T7 treatment could not reach the sufficiency level;therefore,the yield was in the linear phase of the N response curve.Consequently,the yield:N uptake ratio and N physiological efficiency were higher in this treatment than in other treatments.

It is not surprising that the treatments receiving the lowest mineral N showed the greatest N physiological efficiency.Complementary effect of the compost applied with mineral N fertiliser not only increased nutrient mobilisation through induced microbial catabolic activities,but also improved N physiological efficiency(Fisket al.,2015;Abbottet al.,2018).This could be the reason why the treatment with ZC only(T7)had the highest N physiological efficiency.Our findings agree with those of Montemurro(2009),who reported a negative correlation between N uptake efficiency and N physiological efficiency.Hence,the T7 treatment presented a higher Nphysiological efficiency and 1000-grain weight,but lower N uptake than the other treatments.

Generally,composts are mostly comprised of organic N forms(e.g.,proteins,peptides,and amino acids within microbes),which must be first mineralised into inorganic N forms(ammonia and nitrate)before nutrients are accessible to plants(Fisket al.,2015).Organic N is released more slowly than inorganic N,and its turnover is controlled by abiotic factors such as temperature and water(Hoyleet al.,2006,2013;Farrellet al.,2013).In contrast to the composts reported in those studies,the compost used in this study(ZC)had a much lower C:N ratio(11:1),which favoured N mineralisation over N immobilisation and retention in the sandy soil studied,a typical C-limited soil of Western Australia(Fisket al.,2015).Therefore,more N will be available to plants as heterotrophic microorganisms are less able to consume inorganic N at low C:N ratio(Hoyleet al.,2006).This could be the reason for the better response from the treatment with ZC only(T7)than from the treatment with mineral N only(T2).Conversely,we found that the T2 treatment had the lowest grain yield and N agronomic efficiency compared to all other treatments except for the control(T1).This could be due to N leaching in the sandy soil with N applied at sowing.Mixing the compost with mineral N in all the combination treatments might slow N loss,resulting in better grain yield and agronomic efficiency(Ahmadet al.,2008).

Compost and wheat quality

The partial substitution of urea with the compost on a N basis(T2—T6)increased the grain protein content by up to 4.1%and the 1 000-grain weight by up to 8.1%compared with that of the control(T1).The treatment with ZC only(T7)significantly increased the 1 000-grain weight compared to the treatment with mineral N only(T2),which could be due to the high quality of the compost used in this study.However,the T7 treatment did not improve grain protein content in comparison to the T2 treatment.Grain protein content is linked to the uptake of N at later developmental stages of wheat,and the lower grain protein content in the compost probably reflects N deficiency.Montemurro(2009)also reported a similar response of treatment with compost only in comparison with treatment with mineral N only with respect to 1 000-grain weight and grain protein content.Although an equivalent amount of total N(100 kg N ha-1)was applied in both the T2 and T7 treatments,not all the N in the compost was available to wheat during the experiment.Therefore,complete N mineralisation and nitrification of the compost did not take place,leading to N deficiency and consequently reduced grain protein content in the T7 treatment.This is consistent with other studies showing that composted organic materials release N at much slower rates than mineral N fertilisers(Al-Batainaet al.,2016).It can be inferred that compost N could not fully substitute mineral N;however,when applied in combination with mineral N,the complementary effect of compost increased wheat yield and plant nutrition.

SPAD and soil mineral N

Values of SPAD(chlorophyll meter)measurements did not differ at tillering and ranged from 47.7 to 49.7,which could be an indication of sufficient available soil mineral N in all treatments.Hence,there was no correlation between SPAD measurements and soil mineral N at tillering.At heading,SPAD measurements ranged from 49.4 to 54.2 and had significant positive correlations with soil mineral N.This concurs with data from a study of Monostoriet al.(2016),who reported a significant correlation between SPAD value(R2=0.929)and soil mineral Nafter three cropping seasons in Hungary.Fiorentiniet al.(2019)also identified SPAD as a potential tool for indirectly evaluating N availability to crops at their growth stages.However,despite significant correlations between SPAD measurements and soil mineral N,it is difficult to explicitly declare a critical SPAD value for prediction of soil mineral N because a number of factors,such as soil type,crop cultivar,sampling technique,and time,affect SPAD measurements(Spaneret al.,2005).Further research using a wider range of soil and agroecological conditions(including field conditions)is needed to develop a comprehensive and reliable correlation between SPAD values and soil N for precise prediction of soil N.

SPAD and wheat quality prediction

Positive correlations of SPAD measurements with grain N uptake and grain protein content were observed in this as well as in other studies(Spaneret al.,2005;Fiorentiniet al.,2019).However,these reported correlations are site and crop specific,and it is difficult to generalise the correlations with a prediction model.

CONCLUSIONS

Perth Zoo compost and mineral N fertiliser applied in combination to wheat on a sandy soil could increase grain yield and reduce the mineral N fertiliser dose by half under glasshouse conditions.Therefore,Perth Zoo compost could be used to reduce the dependency on mineral N fertiliser without compromising yield.Furthermore,the use of Perth Zoo compost may provide additional benefits and more balance to soil health when combined with mineral fertiliser.The low volume of supply and availability of Perth Zoo compost would preclude its widespread use in agriculture,and other forms of compost that might be more readily available in bulk quantities should be investigated in a similar way.The use of the chlorophyll meter could be beneficial for predicting soil mineral N and hence N requirements,but further tests are required to verify its use under field conditions.

ACKNOWLEDGEMENTS

M.S.A.Khan thanks the Australian Government for providing postgraduate degree scholarship under its Australia Awards Program.The Sir Eric Smart Family contributed funds for this research through the Institute of Agriculture,The University of Western Australia.The salaries of S.N.Jenkins and I.S.Waite were partially supported by RnD4Profit-14-1-022-Waste to Revenue:Novel Fertilisers and Feeds,Australian Pork Limited,and Australian Government(Department of Agriculture and Water Resources)as part of the Rural Research and Development(R&D)for Profit Program.