Decker, David L. : D.L. Decker was postgraduate researcher and is presently graduate student, University of Nevada, Reno.

Fogg, Graham E

Grismer, Mark E.

Louie, Dianne T. : D.T. Louie is Staff Research Associate, Department of Land, Air and Water Resources, UC Davis

Rolston, Dennis E.

Nitrogen isotope ratios identify nitrate contamination sources Dennis E . Rolston P Graham E . Fogg o David L . Decker o Dianne T . Louie P Mark E . Grismer may be contributing significant Nitrate contaminationof ground - amounts of NO3 to groundwater . water is becominga widespread The nitrogen ( N ) isotope method of problem in California.Toevaluate distinguishing sources is based on the utility of stable nitrogen iso - measuring the two stable isotopes of N topes for identifying sources of and 15N ) in NO3 of the sample . A ( 14N nitrate contamination , nitrogen common way of representing these ra - isotope ratios ( V5N ) were mea - tios is delta 15N ( 615N ) , which is the sured on nitrate extracted from differencebetween the ratio 15N / 14N core samples taken below natural , of a sample and that of a standard ( in fertilizer , on - site sewage disposal the atmosphere ) divided by the r'itio ( septic ) and animal sources in the of the standard , with the entire quan - Sacramentoand Salinas valleys . tity multiplied by 1,000 . This results in The mean 6I5Nvalue from natural a 615N value that typically varies less sources was not significantly dif - than several per mil ( % o ) from that of ferent from that of fertilizer the standard . The percentage of the sources . Themean 6l5Nvalue two isotopes is nearly constant in the from animal sources was signifi - atmosphere at 0.366 % 15N . However , cantly different from that of septic because of the slight differencein sources and natural andor fertil - atomic mass of the two isotopes , cer - izer sources . Nitrogen isotope ra - tain chemical and physical processes often preferentiallyutilize one isotope , tios tend to be site specific and causing a relative enrichment of that sus - should be measured below isotope in the product and a relative pected sources in the subsurface enrichment of the other isotope in the and in groundwater . remaining reactants . Because of these isotopicfractionationprocesses , NO3 Nitrate ( N03 ) concentrationshigher from various N sourceshas been than the drinking water standard in shown to have different N isotope ra - groundwater are common in many tios . This ratio in a sample depends on parts of the world . Multiple sources of the series of reactions that formed the NO3 contamination generally occur N compound and the composition of within the same watershed or ground - its precursors . If the sample is en - water basin . Sources include animal riched in 15N relative to the standard , feedlots , horse corrals , dairy waste la - then the 615N value is positive , and if goons , manure applied to land , mu - the sample is depleted in 15N , then the nicipal sewage effluent , on - site sewage 615N value is negative . From literature disposal systems , urban and agricul - sources , the 615N values of NO3 from tural fertilizer , natural soil organic " Sol1 samples were obtained by drilling soil organicnitrogen , fertilizer , animal matter and , in some cases , geologic 10feet below the water - table surface . waste and septic tank effluentrange sources.It is often difficult to ascertain from approximately + 2 to + 8 % 0 , - 3 to which of these sourcesin a watershed CALIFORNIA AGRICULTURE , VOLUME 50 , NUMBER 2 32 Davis source 1 areas 8 4 0 + 2 % 0 + , 9 to + 25 % 0and + 6 to + 25 % 0 , re - vacuum filteringwere centrifuged and spectively . the decantant refiltered.A small This study evaluates the use of N amount of the thawed soil was taken - + - FM1IIIwr 20 isotope ratios as a tool for identifying - - 4 . - Feedlot to determine moisture content . E - - 0 . . saptlc NO3contaminationsourcesin ground - Steam distillationwas used to con - 5 30 - a - Nature1 water for certain California conditions . vert the inorganic forms of N in the + Groundwatersamples The 615N values in soil water and soil - waterextracts to a stable form of groundwater were measured in verti - ammonium salt needed for N isotope I Fertlllzer + Feedlot cal profiles directly beneath various analysis.These samples were then sent 0 septic 60 sources of NO3 contamination ( for ex - to a laboratory for N isotope analysis . 100 150 A 0 50 ample , fertilized fields , animal waste Standard deviation of 6 to 12 samples Nitrate sites , septic systems and unfertilized , of reference standards was no larger Fig . 1 . NO , concentrationswlth depth for uncultivated land ) , from the source to than 0.8 615N units . Multiple extrac - lnorganlcfertilizer , anlmal feedlot , septic tank ( on - site sewage disposal ) and natural the groundwater . tions of field samples were generally soil organlc matter sites In the Davis area within this same range . Thus we do ( southern Sacramento Valley ) . Solid sym - Sampling in Davis , Salinas Valley not consider samples significantly dif - bols represent balled groundwater Two study areas were chosen for ferent if they fall within * 1 815N unit . samples . Soil - water extract concentrations are In pug oven - dried soil , and groundwa - field sampling , based on differing ter concentrationsare In ppm . NO , varies with location , depth hydrogeologic conditions and ease of access for drilling . These areas were in We drilled 26 boreholes during this 0 Davla source erne the southern SacramentoValley , study . Results are shown in terms of where all sampling was performed in NO3 concentrationsand 815N values - = - Fmrilllzer the vicinity of Davis , and in the Salinas versus depth for the Davis area ( figs . 1 - - Q . - Feedlot Valley . Each study region contained and 2 ) and the Salinas Valley area - - 0 - - & pnC sampling sites representing four dif - ( figs . 3 and 4 ) . Some gaps occur in the - A - Niturml ferent sources of NOS:natural soil or - 615N values due to insufficient NO3 to Groundwater urnplea 40 ganic matter , inorganic fertilizer , ani - conduct isotope analysis.Concentra - m Fertlllzer + Feedlot mal feedlot / dairy and septic tank tions of ammonium , chloride and sul - 50 0 SOptlC effluent.Soil and subsurface sampling fate were also determined and used in - A Natural 60 was accomplished by drilling and the interpretation , but are not plotted - 5 0 5 10 15 20 25 6 â?? s N ( % ) samplingwith an & inch - diameter , here . In general , ammonium concen - continuous - flight , hollow - stem auger trations were negligibly small . Fig . 2 . 615Nconcentratlonswlth depth for and with a 2.5 - inch - diameterCalifor - Inorganicfertilizer , animal feedlot , septic NO3 concentrationsvary consider - tank ( on - site sewage disposal ) and natural nia split spoon sampling ahead of the ably with location and depth owing to soil organic matter sites In the Davis area auger . Cores were taken continuously differing land uses , soil - water fluxes ( southern Sacramento Valley ) . Solid sym - from the ground surface to the water and geochemicalprocesses ( figs . 1and bols representballed groundwater samples . table . Drilling continued into the wa - 3 ) . Most of the data represent nitrate in ter table to a depth of 10feet below the soil - waterextractsexpressed as pg / g 0 SalinasValley water - table surface whenever possible . oven - dried soil . source ems A water sample was collected from the As expected , NO3 concentrations Soll - water extracts ( @ g ovendrlodsoll ) bottom of eachborehole upon reach - beneath the natural sources were low . 50 - Fwtlllzeer - ing groundwater , using a bailer inside No3 concentrationsbeneath the septic E . - - + - F - eedlot the auger . Subsamples of the core sources also tended to be low . This - 43 - - saptlC ~ - A - Natural were taken approximately every 5 feet could be caused by the boreholes not and were preserved for analysis . After intercepting the main discharge zones the core was taken in the field , the of the on - site sewage effluent , which I Fenllker sample was placed in a polyethylene can seep from the leach lines or pits in + Feedlot 150 A Natural zip - lock bag and placed in an ice chest a spatially and temporally erratic fash - 0 50 100 150 200 250 filled with dry ice . The sample was ion . Nevertheless , we are confident Nitrate Fig . 3 . N0 , concentrations wlth depth for then transferred to a freezer until it that our sampling captured enough inorganicfertilizer , anlmal feedlot , septic was processed . No3 to register 615N signatures of the tank ( on - site sewage disposal ) and natural various sources . The high NO3 concen - soil organlc matter sites in the Sallnas Nitrogen isotope analyses Valley . Solid symbols representbailed trations in shallow groundwater be - groundwater samples . Soil - water extract Nitrate in the soil sampleswas ex - neath the fertilizer , feedlot and septic concentrationsare in pg / g oven - driedsoil , tracted with deionized water ( 1 : 5 ) . sitesin the Davis area and beneath the and groundwaterconcentrationsare in Samples that were still cloudy after fertilizerand feedlot sites in the Sali - PPm . CALIFORNIA AGRICULTURE , MARCH - APRIL 1996 33 tistics are based on all of the 615N data , years , the 615N value was consistent at 0 Salinas Valley source arms including those boreholes not shown about throughout the vertical 25 Soil - waterexlracia in figures 1through 4 . The mean 615N subsurface profile ( fig . 2 ) . The ground - - m - Fertilizer value for the agricultural ( fertilizer ) water sample showed a slightly lower - - 4 . - Feedlot 50 - 4 - - SeptlC source was not significantly different 615N value than the soil water . At the 5 - A - Natural P from the mean of the natural back - Salinas Valley animal waste site , three d 75 Groundwater samples ground source at the 95 % confidence boreholes were drilled in two of four rn Fertlllzer level . The animal sources were sig - abandoned dairy waste evaporation 100 Feedlot nificantly different from the septic ponds and the animal feeding pens . A Natural sources and the fertilizer and natural The borehole with the most consistent 125 0 5 10 15 20 25 - 5 sources . 615N values of the three boreholes had 6 â?? = N ( % ) In addition , we made statistical a nearly constant value of about + 17 % 0 Fig . 4 . 615Nconcentrationswith depth for comparisons between mean 615N val - in the upper 50 feet of the profile ( fig . 4 ) . inorganic fertilizer , animal feedlot , septic ues between areas for each source Below 50 feet , the 615N values de - tank ( on - site sewage disposal ) and natural type . There was no significant differ - creased to around + 6 % 0 ; however , the soil organic matter sites in the Salinas ence in means for the natural , septic Valley . Solid symbols representbailed other two boreholes at this site groundwater samples . and animal waste sources between the showed high 815N values both deep Davis and Salinas areas , but means for and shallow . The spatial variability of the fertilizer sites were significantly nas Valley could reflect both local and 615N values at this site can be attrib - different . The elevation of mean 615N off - site sources . uted to subsurface heterogeneity and values at the Salinas fertilizer site over to the fact that the site contains several Natural soil organic matter those at the Davis fertilizer site may be point sources of animal waste . sources . 615N values beneath the un - partly due to the occasional applica - fertilized , natural soil organic matter Fertilizer sources . Five boreholes tion of compost containing animal sites were very consistent among the were drilled at the Davis site and 10 at waste at the Salinas location and 1or to borings in the unsaturated zone and the Salinas site . Figures 2 and 4 give a significant amount of nitrogen - fixing varied between about zero and + 5 % 0 examples of 615N values with depth legumes in the cropping history of the ( figs . 2 and for the two sites . The Davis site has 4 ) . The 615N values in the Davis location . been cropped with corn , alfalfa and groundwater samples for the Salinas sugar beets for many years . The Sali - site were all too high for a natural or - Mixed sources of groundwater N nas site is in an irrigated agricultural ganic matter source and suggest a The 615N data shown in figures 2 field that has been in a mixed veg - likely non - natural source in the up - and 4 are chiefly from the unsaturated etable cropping system for several gradient direction . decades . Although most of the fertil - zone directly beneath known source On - site sewage disposal types . In groundwater investigations izer applied was inorganic sources . 615N values from the Davis N , some of compost containing animal waste was NO3 contamination , however , much on - site sewage disposal ( septic ) site of the 615N data will represent deeper ( fig . 2 ) showed an increasing trend also applied occasionally at the Salinas groundwater pumped from produc - with depth , but were well below the site . The 615N values for these sites tion wells . If this deeper groundwater largest ratios expected for animal were nearly constant with depth , con - contains NO3contamination from waste . This borehole was drilled be - sistent with the long - term use of fertil - more than one source , the 615N values tween two leach lines spaced 25 feet izer . The constant 615N values with depth also indicate that denitrification will reflect a mixture of those sources . apart . We suspect that water from the Therefore it is typically not possible to leach lines did not move far enough in the subsurface must have been small . The cyclic high - low trend in â?? fingerprint â?쳌 the source of NO3 con - horizontally to intersect the borehole tamination by merely measuring 815N location until a substantial depth was NO3 concentrations beneath the Sali - nas site ( fig . 3 ) is due to geologic layer - in a few water wells . Clearly , deter - reached ; this would explain the low mining the sources of NO3 contamina - concentrations of ing of the subsurface profile at this lo - NO3 and low 615N tion in groundwater commonly re - cation , with the lower values tending values in the shallow section . At the quires careful hydrogeologic analysis to locate in the coarser - grained sedi - Salinas site ( fig . 4 ) , the borehole was ments . The lack of similar cyclicity in of groundwater migration rates and drilled beside a deep vertical sewage flow paths as well as an exhaustive in - disposal pit . Except for one very high 615N at the same borehole ( fig . 4 ) at - ventory of potential tests to the independence of 615N val - NO3 sources . In 615N value at the Salinas site , most of this analysis , the 615N technique is ues with respect to concentration . the 615N values from the human waste an important tool rather than â?쳌 the Comparisonsof different disposal systems lie between + 5 and solution . â?쳌 Table 1shows statistical sources . + 10 % 0 . Even if an area contains only one comparisons of the differences in Animal waste sources . At the source of NO3 contamination that is mean Davis animal waste site , a horse corral 615N values for the various theoretically distinguishable by 615N source types and locations . These sta - that had been in use for at least 30 CALIFORNIA AGRICULTURE , VOLUME 50 , NUMBER 2 34 techniques , it is possible that the 615N values measured in groundwater or deep unsaturated zone will be measur - ably higher than in the shallow zone directlybeneath the source . The rea - son is that denitrification , the process whereby NO3 is converted to N20 and N2 gases , can raise the 615N values above those of the original source . Denitrificationis most likely when anaerobicconditions occur in a bio - logically active soil zone , which in turn most likely appears when the wa - ter table is within a few feet of land surface.When the water table is shal - low , there is greater likelihood that anaerobicconditions of the saturated zone extend sufficiently into the bio - logically active soil zone to drive deni - trification.Clearly , in addition to hydrogeologic analysis , application of the 615N technique requires collecting data on 615N values beneath the sources of concern to detect whether denitrificationis occurring and to verify the 615N signatures of the sources.Additional measurement of 615N values beneath septic sources is particularly needed because of the dearth of such data in the literature . In this study , little to no evidence of denitrificationwas found , presumably because the water table was 40 to more than 100 feet below land surface at all sites , well below the soil zone . That is , 615N values tended not to increase Isotope ratios as indicators with depth , except beneath the Davis septic site ( fig . 2 ) and in one of the Nitrogen in the subsurface zone ex - values of about + 7.3 % 0at Davis and boreholes drilled at the Salinas septic isted primarily as nitrate , except for + 8.7 % 0at Salinas Valley . It is clear from site . At the Davis septic site , the bore - one septic and one dairy lagoon site in our data that the 615N methodology hole did not intersect the plume until a the SalinasValley where significant cannot be used to distinguish between depth of approximately 25 feet , which amounts of ammonium existed . The fertilizer and natural soil organic mat - explains the increase in 615N value ter sources . However , in the sites that 615N values determined from the vari - with depth . At the SalinasValley sep - we examined , the ous sites fall within the range of values 615N method can be tic tank site , the 615N value increased reported in the literature , but with used to distinguish between animal with depth and the NO3 concentra - lower variances . The mean sources and fertilizer and / or soil or - 615N value tions decreased with depth , which for the natural background sites was ganic matter sources.In addition , our could be indicative of denitrification . about + 2.6 % 0in both the SalinasValley data show that there is a significant In addition , a few locally elevated 615N differencebetween mean and Davis study areas . Mean 615N 615N of the values were found at various sites , such animal and septic sources.These re - value for the fertilized sites was as the deepest samplebeneath the Sali - slightly greater than for the natural sults exemplify the necessity of mea - nas Valley fertilizer site ( fig . 4 ) . How - sites with + 4.4 % 0in the SalinasValley suring the specific 815N of suspected ever , 615N values from the otherbore - and slightly lower at sources of + l . 6 % , i , n the NO3 contamination for the holes and from deeper groundwater Davis area . The animal waste sites had particular area of interest . Simply us - verify that these elevated values are iso - a mean value of ing values from the literature may + 10 % 0at Davis and lated anomalies , apparently causedby + 14 % 0in SalinasValley . The on - site not allow clear separation of source highly localized processes . sewage disposal sites had mean 615N types . CALIFORNIA AGRICULTURE , MARCH - APRIL 1996 35 How does water price The 615N values at most sites were fairly consistentwith depth from the affect irrigationtechnology surface to the water table . There is no evidencefrom our data that denitrifi - adoption ? cation is a significant process at any of the sites , with the possible exception of the Salinas Valley septic tank site . P David Zilberman o David Sunding Slareth Green Thus , except for one site in the eight Collup Steve Doug Parker D Cliff Trotter o main test sites , the results demonstrate that measuring the 615N value imme - diatelybelow the NO3 source can be fewer resources . In this article , we as - an accurateindicator of the fingerprint The use of water price or best sess whether technology choice is con - of that source and that , under the con - managementpractices have been sistent with the assumption of profit ditions prevailing at these sites , the advocated by some commentators maximizationand , if so , determine fingerprint will not change much dur - to induce adoption of low - volume which factorsmost influence technol - ing NO3 transport to groundwater . irrigation technologiesand to en - ogy choice . This is a very important conclusion for courage water use efficiency . Some commentatorshave advo - use of the N isotope technique to indi - However , the method of water ap - cated the use of water price as a policy cate sources of NO3 in groundwater . plication is only one of many in - tool to induce adoption of low - volume Nevertheless , users of the & 5N ap - puts and constraints in agricul - irrigation technologies and to encour - proach should be aware of the poten - tural production . Californiaâ??s age increased water - use efficiency . tial for mixing of 615N from multiple highly diverse topography , soil Specifically , environmentalistsand sourcesand of denitrificationunder types and variety of crops influ - many economists frequently assert some circumstances . Careful hydro - ence irrigation technology that irrigation water should be priced geologiccharacterizationas well as choices , therefore a policy man - to encourage adoption of modern tech - sampling of both the unsaturated and dating adoption of modern tech - nologies and reflect the value of water saturated zonesbeneath potential nologies is likely to have undesir - outside agriculture . However , the ef - sources are therefore typically re - able impacts . Crop type appears fectiveness of water price to achieve quired for successful application of the to be a major considerationin these goals may be limited because the 615N approach . technologychoice , as some tech - method of water application is only nologies may be incompatible one of many crucial inputs and con - D.E . Rolston , G.E . Fogg and M.E . straints in agricultural production . with some types of crops . Grismer are Professors of Soil Science , Our model demonstrates that large Hydrogeology and Hydrologic Science , re - increases in the price of water gener - Continued urban population growth , spectively , and D.T . Louie is Staff Re - ally encourage heavier reliance on heightened public awareness of the search Associate , Department of Land , Air drip and other low - pressure irrigation environmentalbenefits of in - stream and Water Resources , UC Davis ; and D.L . systems for certain crops , but may water flows , and the virtual halt of Decker was postgraduate researcher and is have only modest effects on adoption water supply development in Califor - presently graduate student , University of decisionsfor other modern irrigation nia have increased pressure on state Nevada , Reno . technologies . and federal agencies to reallocate wa - This research was supported by the ter away from agriculture.Many public - California Water Resources Control Irrigation decisions in Arvin interest groups and policy makers Board , Interagency Agreement No . 1 - 155 - We selected the Arvin Edison Wa - have suggested that growers could in - 253 - 0 , Monterey County Water Resources ter StorageDistrict , located in the crease their use of low - volumeirriga - Agency and the U.S . Environmental Pro - southern San Joaquin Valley at the ter - tion technologies while maintaining tection Agency Assistance Agreement minus of the Friant - Kern Canal , as our current production levels . Some inter - Nos . 1009564 - 91 - 0 and C9009532 - 90 - 1 . study area . There is wide variation in ests have even advocated imposing The contents do not necessarily reflect the the types of irrigation technologies agricultural â?? best management prac - views and policies of the U.S . EPA , employed in the District : 25 % furrow tices â?쳌 mandating the adoption of irri - Monterey County or the California Water or flood , 49 % high - pressure sprinkler gation technologies . California grow - Resources Control Board . The authors and 26 % low - pressure drip and ers have been criticized for their appreciate the assistance of Heike microsprinkler ( table 1 ) . This variation â?? irrational â?쳌 and â?? inefficient â?쳌 irrigation Clausnitzer , Amy Wong , Mike Tolin and makes the District ideal for analysis technology choices . It has been sug - Brian Lindsay in laboratoy analytical because there is a large amount of gested that growers could maintain or work and of Dr . Richard Mulvaney at the variability , yet the area is relatively increasetheir profitabilitywhile using University of Illinois for B5N analysis . CALIFORNIA AGRICULTURE , VOLUME 50 , NUMBER 2 36