1 Manuscript accepted March 1987.2 Hawaii Institute of Marine Biology, University of

Hawaii, 1000 Pope Rd ., Honolulu , HI 96822.30 ceanic Institute, Makapuu Point, Waimanalo, HI

96795.4U.S. Army Corps of Engineers, Pacific Ocean Divi­

sion, Constru ction Opera tions Branch , Fo rt Shafter , HI96858.

ANCHIALINE PONDS are land-locked brackishwater pools adjacent to the sea (Holthuis1973). These pools have subterranean connec­tions to the ocean and pond volumes respondto tidal fluctuations. Anchialine ponds aregeographically restricted and are found inporous coralline or recent volcanic substrata.They are characterized by an unusual array oforganisms, many of which are found only inthe anchialine habitat. Anchialine ponds andtheir biota have only recent ly received atten­tion in the literature (Holthuis 1973; Maciolek1983; Kensley and Williams 1986). Most workhas been concerned with descriptions of thebiotope, taxonomy of the fauna, or hypoth­eses to explain observed faunal distribution;litt le quantitative information is available onthe biota or any other component of the sys­tem. Reasons for this lack of information in­clude the cryptic and hypogeal nature (sensuMaciolek 1983) of some anchialine speciesand the difficulty in sampling these systems.

Pacific Science (1987), vol. 41, nos. 1-4© 1988 by the University of Hawaii Press. All rights reserved

Characteristics of Water Quality in Anchialine Ponds of the Kona,Hawaii, Coast .1

RICHARD E. BROCK, JAMES E. NORRIS,2 DAVID A. ZIEMANN3 AND MICHAEL T. LEE4

ABSTRACT: A study of the water quality characteristics of anchialine pondsof the Kona, Hawaii, coast suggests that groundwater is a major source ofdissolved nutrients for these systems. These groundwater sources apparentlyshow high spatial and temporal variability with respect to dissolved nutrients.Changes are apparent in the water quality characteristics of one anchialine pondsystem that has been subjected to considerable surrounding development. Thesechanges are within the range of natural variability suggesting that this perturba­tion, at least over the short term (ca. 9 years), is not damaging since thesenutrients frequently occur naturally in excess of concentrations which wouldcontrol biological proce sses. Within an anchialine pond system that we havestudied, spatial variability in water quality may be explained by a simple modelof groundwater diluti on with proximity to the sea.

In the Ha waiian Islands, anchialine pondsare found in the geologicall y young lava flowson Maui and Hawaii islands . More than 70%of the estimated 520 Hawaii Island anchialineponds occur in a 53 km contiguous sectionof the Kona coast. The lands surroundingmany of these Kona coast anchialine pondshave been proposed for or are undergoingresort and residential development. As aresult, much of what is known regardingthese anchialine systems is found not in thepublished literature, but rather in impactassessments written to fulfill environmentalregulations.

One recently initiated development at theWaikoloa, Kona, area (between Waiulua andAnaehoomalu Bays) destroyed more than 130anchialine ponds in late 1985; approximately66 adjacent ponds with a combined water sur­face area of 1.4 ha were set aside in a 4.9 hapreserve (the Waikoloa Anchialine Pond Pre­servation Area- see Figure 1). Since the ear ly1970's, selected ponds in this Waiulua-An­aehoomalu complex (some of which are nowin the preserve) were occasionally sampled fororganism abundance or water quality charac­teristic s. The literature on water quality char­acteristics of anchialine ponds is scant, andthere are no comprehensive studies. Cox et al.(1969) reported nitrate, phosphate, and sili-

200

Water Quality in Anchialine Ponds-i-Bnocx, NORRIS, ZIEMANN, AND LEE 201

cate levels from four Kona coast ponds; allother data are given in unpublished reports.With increasing human presence along theKona coast, changes in anchialine pond waterquality are likely. Such changes could have aprofound effect on the biota. This study ex­amines the water quality characteristics of ananchialine pond system undergoing surround­ing development and tests the hypothesis thatthere has been no change in the water qualityof this system since the commencement ofdevelopment. We also propose a parsimoni­ous hydrological model that explains the spa­tial variability observed in water quality char­acteristics of these ponds.

MATERIALS AND METHODS

Study Site

This study was carried out in the Wai­koloa Anchialine Pond Preservation Area(WAPPA). The WAPPA is situated in theWaiulua Bay-Anaehoomalu section of theSouth Kohala District in Kona, Hawaii (Fig­ure I). The largest concentration ofanchialineponds on Hawaii Island was located in theWaiulua Bay area. In December 1985 morethan 130of these ponds were destroyed duringthe ongoing construction of a hotel complex.The WAPPA was established as a mitigativemeasure. Other development adjacent to andin the vicinity of the WAPPA has been under­taken over the last nine years; a golf courseborders the preserve on the inland side and aroad bisects it (completed in 1979, see Figure1).About 550 m southeast of the managementarea is a second resort completed in 1981.

Three natural ponds were chosen for rou­tine monthly sampling in the WAPPA. Sam­pling commenced in April 1986. The pondswere selected on the basis of their location inthe system relative to nearby developmentsand the shoreline. The sampled ponds repre­sent a range of situations from the inland/development margin of the preserve adjacentto the golf course to the natural shoreward/ocean border. Figure 1 presents a map of theWAPPA with existing borders and all extantponds along with the three ponds (numbers

48,155, and 188) that were routinely sampled;roadways and other nearby development arealso indicated. Pond 48 was selected as repre­sentative of the inland location (approximately308 m from the shoreline) that is close to de­velopment activities. Pond ISS is located inthe center of the preserve (about 148m fromthe ocean), and Pond 188 is situated close tothe ocean (56 m away) and farthest away fromdirect construction activities.

Several other locations have been sampledmonthly. One of these is a coastal wel1 devel­oped by the resort for irrigation purposes. Thewell is located about 1 km southeast and 850m inland of the preserve, is well removed fromcoastal development, and serves as a source ofnatural (low salinity) groundwater for thisstudy. As part of the permit requirements, theresort developer has dug the first of severalman-made anchialine ponds (see Figure I) .These artificial ponds are designed to provideadditional anchialine habitat. This first arti ­ficial water body is located just outside theWAPPA border about 50 m to the north ofPond 155. Water quality sampling of thispond commenced in July 1986 following itsconstruction. In October 1986, water qualitysampling of the irrigation water which is usedon the golf course inland of the preserve wasinitiated. For comparative purposes, 24 an­chialine ponds situated in the South Kohala­North Kona districts were sampled betweenNovember 1985 and September 1986. Theseponds are representative of Kona coast an­chialine systems and thus the water qualitydata serve as a control.

Data Collection and Laboratory Methods

In the field, replicate samples of several pa­rameters were collected for laboratory anal­ysis. Water samples that were analyzed fornutrients were filtered in the field throughprecombusted GFC (2.1 cm diameter) filters.All glassware and sample bottles were acidrinsed . With the exception of samples col­lected for salinity or chlorophyl1 analyses, allwater quality parameters were sampled intriplicate. Although temperature and salinitystratification was occasionally observed indeeper (> 70 ern) parts of some ponds, all

202 PACIFIC SCIENCE, Volume 41, 1987

HAWAII

f

d 12miles

d SOm

FI GURE I. Map showing the location of four sampling sites (ponds 48, 155, 188, and mm-I) in the WaikoloaAnchialine Pond Preservati on Area (WAPPA) The preservation boundary is shown as a dashed line. The locat ion ofthe WAPPA on Hawaii Island is given in the inset.

samples were taken at a 15 to 20 em depth inthe water column to sample surface water s.Samples for salinity analyses were not filtered.Chlorophyll samples were collected and pro­cessed following the extraction and fluores­cence procedures recommended by Jeffrey(1974), Jeffrey and Humphrey (1975), and

Strickland and Parsons (1972). Salinity valueswere calculated from conductivity measure­ments obtained on a Plessey-Grundy model6230N laboratory salinometer.

Micronutrient concentrations of sampleswere determined according to somewhatmodified versions of the methods listed by

Water Quality in Anchialine Ponds-s-Bnocx, NORRIS, ZIEMANN, AND L EE 203

Strickland and Parsons (1972). The analyseswere performed on a Technicon AutoanalyzerII system. Using the autoanalyzer, PO;3 (asorthophosphate) was determined followingmethods as outlined by Murphy and Riley(1962), N0i" + N03 by the techniques ofArmstrong, Williams, and Strickland (1966),NH1 following the procedures of Solorzano(1969) and Si02 was determined after themethods of Strickland and Parsons (1972).More detailed information on the laboratorymethods are given in Smith et al. (1981).

We commenced sampling the three naturalponds (nos. 48,155 and 188) in April 1986. InMay we began sampling the coastal well andin July , following its construction, the man­made pond was added to the list of regularlysampled localities. In October 1986 we ini­tiated sampling of the water used to irrigatethe golf course inland of the WAPP A. Thisirrigation water is drawn from the coastalwell; the well water is used to dilute treatedsewage effluent from the resort which is thenutilized for irrigation purposes.

Data for the water quality parameters ex­amined in this study were compared betweennatural ponds (pooling data for a given pondover all surveys) and between surveys (poolingdata for all natural ponds in each survey) . Forthe above comparisons and others made inthis study, the Kruskal-Wallis analysis ofvari­ance was used (SAS Institute 1985).

RESULTS

The water quality data are summarized inTable 1. These data are presented sequentiallywith respect to station distance from theshoreline for each survey. Thus of the lo­cations sampled, the most inland from theshoreline is the well, followed by Pond 48, theman-made Pond 155, and Pond 188 which isin closest proximity to the sea. The results ofthe Kruskal-Wallis one-way analysis thatcompares variables between the natural ponds(pooling data for a given natural pond over allsurveys) and between surveys (pooling datafor all natural ponds in each survey) arepresented in Table 2. The results of theseANOVAs are discussed below.

Surface salinity measurements of the sam­pled locations in the WAPPA are given inTable 1. The tide state in all ponds was nearmean low tide (0.0 m) during the first threesample periods and high (+0.05 to +0.8 m)during the last three. The tide during the Octo­ber sample was unusually high. The sampledponds in the preserve have maximum depthsat mean tide of less than 1 m. The three natu­ral ponds exhibited a between-pond gradientwhere surface salinity increased in a seawarddirection. The deeper man-made pond exhib­ited a departure from this trend; in three offour samples, salinity was higher than in PondISS.

Changes in surface salinity from one surveyto the next are small; both Ponds 48 and 155show a decrease in surface salinities at hightide while this is not apparent in either Pond188 or the coastal well. The analysis of vari­ance using pooled salinity data for eachnatural pond (48, 188, 155) over all surveysshowed that significant differences exist be­tween these ponds (Table 2). This horizontalsalinity gradient is apparent in Table I andseems to be related to pond location relativeto the sea. This gradient appears to be a rela­tively constant feature, for no statistically sig­nificant differences were found between sur­veys (Table 2).

The concentrations of nitrate plus nitrite(hereafter called nitrate) in the sampled pondsis given in Table 1. The data in Table 1 sug­gest some patterns: nitrate values were con­sistently lowest in the coastal well water,which presumably is representative of naturalgroundwater. The highest values were foundin Pond 48, the most distant from the shore­line, and a trend of decreasing values is appar­ent in ponds located progressively closer tothe shoreline. Similar to the salinity data thisnitrate gradient is statistically significant be­tween the natural ponds in the WAPPA but itis not between surveys (Table 2). The irriga­tion water sampled in October had a nitratelevel of 46.21 jiM.

The concentration of phosphate (PO;3) inthe various WAPPA ponds from April to Oc­tober is given in Table 1. The distribution ofphosphate shows a trend similar to nitrate inthe WAPPA system: within anyone sampling

204 PACIFIC SCIENCE, Volume 41, 1987

TABLE I

MEASUREMENTS OF WATER QUALITY PARAMETERS FROM THREE NATURAL PONDS, O NE M AN-MADE POND AND THECOASTAL WELL FROM THE WAPPA MADE OVERA SEVEN MONTH PERIODFROMA PRIL THROUGH OCTOBER, 1986.

The tide state of each pond at the time samples were taken is presented simply as either low (L) or high (H)

SALINITY NO; PO ; 3 NRi Si02 Ch I aLOCATION SURVEY TIDE (0/00) (jlM ) (jlM) (jl M) (jl M) (mg jm3)

Coastal MAY 1.7 49.8 1.9 0.09 811well JUL 1.7 47.6 2.0 0.05 814

A UG 1.7 46.5 2.2 0.03 778SEP 1.7 47.4 2.4 0.64 791OCT 1.7 45.5 1.8 0.13 860X 1.7 47.4 2.1 0.18 811± S.D . 0.03 1.6 0.2 0.24 31

Pond APR L 6.3 74.7 5.1 0.90 745 0.0748 MAY L 6.2 83.0 5.9 1.30 706 0.11

JUL L 6.4 79.4 4.5 1.20 705 0.09AUG H 3.0 86.4 5.1 0.90 708 0.06SEP H 4.1 98.1 5.7 0.91 738 0.08OCT H 7.0 97.9 5.5 0.59 762 0.Q7X 5.7 86.6 5.3 1.00 732 0.08± S.D . 2.35 4.6 0.2 0.25 21 0.06

Man-made JUL L 9.5 64.9 4.0 0.92 657 0.42Pond AUG H 7.3 66.3 3.8 0.47 681 0.67

SEP H 6.3 78.8 3.9 1.02 707 0.12OCT H 7.8 78.6 4.2 1.29 691 0.04X 7.7 72.1 4.0 0.93 684 0.31± S.D. 1.3 7.0 0.2 0.31 21 0.29

Pond APR L 6.7 51.1 3.0 1.40 731 0.09155 MAY L 6.5 78.3 3.6 1.03 701 0.11

JUL L 6.7 60.3 3.3 1.42 694 0.16AUG H 5.2 60.6 3.6 1.11 708 0.09SEP H 8.8 63.7 3.1 1.56 712 0.10OCT H 4.1 68.6 3.2 0.71 751 0.Q7X 6.4 63.7 3.3 1.21 720 0.10± S.D . 1.4 8.6 0.3 0.30 21 0.03

Pond APR L 8.3 55.4 2.6 0.69 703 0.05188 MAY L 10.5 50.4 2.4 0.85 638 0.08

JU L L 8.9 53.4 2.6 0.77 647 0.03A UG H 9.6 45.0 2.7 0.57 628 0.09SEP H 8.5 48.8 2.3 1.23 658 0.21OCT H 15.2 44.4 2.4 0.72 598 0.08X 9.7 48.6 2.5 0.80 660 0.09± S.D . 2.3 4.6 0.2 0.22 40 0.06

TABLE 2

CHI-SQUARE VALUES FROM THE KRUSKAL-WALLIS ANOVA OFWATER QUALITY PARAMETERS. A na lyses wereperformed with each respon se vari a ble po oled across surveys (Pond ana lysis) and acro ss ponds (Survey analysis).

Significa nt differences a t the P < 0.0 I level are den oted with an as terisk

CLASS

PondSurvey

SALINITY

14.80*1.14

43.46*2.40

47.16*1.29

15.72*21.26*

12.85*7.53

ChI a

2.845.54

Water Quality in Anch ialine Ponds-e-Bao cx, NORRIS, ZIEMANN, AND L EE 205

period, phosphate concentration was lowestin the low salinity coastal well, highest in themost inland pond (48) and decreased in pondslocated closer to the shoreline. These differ­ences in phosphate levels between ponds arestatistically significant (Table 2). This trend isapparently stable through time for there areno statistically significant differences betweensurveys (see Table 2). The phosphate concen ­tration of the single irrigation water samplewas 19.7 jlM , over three times the concentra­tion found at any other location.

Ammonium (NHt) concentrations in thesampled ponds through the April -Octoberperiod are also given in Table 1. The lowsalinity coastal well ammonium levels are anorder of magnitude lower than the concentra­tions seen in the ponds. Although highestvalues were found in the central part of thepreserve (Pond 155), the trend of decreasingammonium concentrations with proximity tothe shore persists . These differences are statis­tically significant between ponds as well asbetween surveys (Table 2) suggesting consid­erable fluctuation in concentration throughtime. The single irrigation water sample pro­vided the highest ammonium concentrationrecorded in this study (9.4 jlM), almost sixtimes the concentration observed at any otherlocation.

As with the other nutrients, silicate levelswere highest inland and they decreased to­wards the sea (Table 1). The differences in thelevel of silicate between ponds are statisticallysignificant but no significant differences existbetween surveys (Table 2) suggesting that theseaward gradient in silicate is temporally sta­ble. The silicate concentration from the singleirrigation sample was 887 jlM.

Chlorophyll a (Chi a) concentration in thesampled ponds did not exhibit any obvioustrend; these data are given in Table 1. In thenatural ponds (48, 155, and 188) mean Chi aconcentrations are similar with no statisticallysignificant differences apparent either betweenponds or surveys. The highest Chi a valueswere found in the man-made pond.

Twenty-four anchialine ponds well removedfrom the Waikoloa pond preserve along theKona coast were sampled for comparativepurposes. Other than one pond with nearby

hotel construction, these pools have been sub­jected to little or no surrounding develop­ment. The results ofwater quality sampling inthese ponds are as follows ex ± S.D.): salinityranged from 1 to 14ppt (6.3 ± 3.0 ppt), nitratevalues were between 0.5 to 62.4 jlM (38.1 ±19.6 jlM), phosphate concentrations were be­tween 0.5 to 6.6 jlM (2.0 ± 1.9 jlM) andammonium ranged from 0.3 to 14.8 jlM witha mean of 3.1 ± 3.6 jlM.

DISCUSSION

A characteristic feature of west Hawaii isits diffuse groundwater discharge at the shore­line (Cox et al. 1969). Estimates range from2300 to 9400 m' day " km - I of coastline inthe vicinity of the WAPPA (Kanehiro 1977).This discharge is a result of the island's geolo­gically young lavas. The high porosity of theselavas will not support water contained abovesea level near the shoreline (Cox et al. 1969).Thus, anchialine ponds are restricted to de­pressions in the lava that extend down into thewater table . Anchialine ponds are defined ashaving brackish water; this mixohalinity is theresult of seaward-flowing groundwater mov­ing through the porous substratum and mix-·ing with warmer, more saline waters below .Typically the residence time of water in an­chialine ponds is on the order of hours and isrelated to high substratum porosity (Kane­hiro 1977). Because of the subterranean con­nection, anchialine ponds are tidally influ­enced; groundwater mixes with seawater andthe water quality of the ponds reflect this in­teraction. Since groundwater and seawatershow distinct differences in the water qualitycharacteristics we measured, the pond charac­teristics are expected to display variabilitycompatible with relative influence of a givenwater type prevailing at a given tidal condi­tion. Thus tide state and the resulting variabledegree of mixing contribute to the extantchemical conditions in anchialine ponds.These conditions may be further altered byother physical (e.g., solar radiation, basinpermeability, and location, etc.) and biologi­cal processes (e.g., nutrient assimilation byplants).

206

Along the Kona coast, the greatest naturalcontribution of nutrients comes from ground­water rather than surface seawater (Bienfang,1980). This author noted nearshore oceanwater nutrient concentrations from a location35 km south of the WAPPA as follows: nitrate0.3 liM, phosphate 0.1 liM and ammonium0.3 liM . Swain (1973) reported groundwaternitrate concentrations (X = 55.7 liM, n = 2)and silicate concentrations (X = 1550 liM,n = 2) from an upland well. Kay et al. (1977)noted Kona coast groundwater nitrate levelsranging from 27 to 108 liM (X = 57 ± 15.3liM, n = 8) and phosphate concentrations from1 to 3.9 liM (X = 2.1 ± 0.3 liM, n = 8). Inthe present study, groundwater nitrate, phos­phate, and silicate, as represented by thecoastal well samples, are all relatively high.Reported natural groundwater nutrient levelsin other localities may be greater. Johannes(1980) reported groundwater nitrate levels be­tween 115 and 380 liM from Perth, Australia,and Marsh (1977) noted nitrate concentrationsin Agana, Guam groundwater of 178 liM.

The subterranean flow of the low-salinity,high-nutrient water towards the shoreline atWaikoloa is driven by the hydrostatic headdeveloped inland; because of low rainfall andhigh substratum permeability this water tablegradient is only about 19 cmkm" (Kanehiro1977). This suggests that the water chemistryof the WAPP A ponds is controlled by the netseaward flow of groundwater. The data fornitrate, phosphate, and silicate show a statis­tically significant gradient relative to pondlocation, decreasing with proximity to theshoreline. Salinity shows an opposing statis­tically significant gradient with respect to dis­tance from the sea. These gradients appear tobe relatively stable features (no statisticallysignificant changes through time) and areprobably caused by the dilution of ground­water with seawater on its movement towardsthe sea, as well as by modification due tobiological activity (assimilation) in the ponds.Ammonium does not seem to follow the trenddescribed for the other nutrients. The statis­tically significant spatial (between ponds) andtemporal (between surveys) differences inammonium concentrations are probably re­flective of rapid turnover of this nutrient.

PACIFIC SCIENCE, Volume 41, 1987

Most of the plant biomass in these anchialinesystems is comprised of benthic (attached)algae; phytoplankton which our ChI a sam­ples measure comprises a small proportion ofthe total. Phytoplankton response to the nu­trient gradient as measured by ChI a was lowand not significant suggesting that pond waterresidence time is short relative to phytoplank­ton turnover.

The seaward flow of high nutrient ground­water and its dilution by nutrient-poor sea­water provides a simple model that explainsthe observed trends in water chemistry of theWaikoloa anchialine ponds. Nutrient levelsother than silicate in the natural groundwater(as represented by the coastal well) are, how­ever, consistently lower than concentrationsfound in the inland ponds. Thus, if this modelis correct, there must be additional nutrientinput occurring between the coastal well andthe inland border of the WAPPA (Pond 48).One possible source is the Waikoloa golfcourse situated about 30 m inland of the pre­serve. Water used to irrigate the grounds isenriched with treated sewage effluent. We ob­tained the following nutrient concentrations(in liM) from this irrigation water: nitrate46.2, phosphate 19.7, ammonium 9.4, and sili­cate 887. Dry fertilizers (21-7-14, N: P: Kratio) are also applied to the golf course at anapproximate rate of 276 kg ha -I yr" . Ni­trogen and phosphorus from these sourcesmay be entering the WAPPA groundwateralthough Chang and Young (1977) found noleaching to groundwater beneath a golfcourseon Oahu receiving a similar nutrient subsidy.This (1977) study however was carried out onolder, much less permeable substrate than ispresent at Waikoloa. Autochthonous input ofnitrogen is also possible through nitrogenfixation by the kiawe tree (Prosopis pallida)which is present in the pond preserve. If sig­nificant autochthonous nutrient productionwas occurring throughout the preserve, onemight expect a gradient of increasing nitratelevels towards the shoreline. This is not thecase.

Comparative anchialine pond nutrient dataare available from Cox et al. (1969). Theseauthors noted nitrate, phosphate and silicatelevels in four anchialine ponds adjacent to the

Water Quality in Anchialine Ponds-e-Baocx, NORRIS, ZIEMANN, AND LEE

TABLE 3

207

COMPARISON OF MEAN SALINITIES (%0 ± SD) AND MEAN NUTRIENT CONCENTRATIONS (pM ± SD) TAKEN IN 1977FROM FIFTY PONDS LOCATED INOR NEAR THE WAPPA (BIENFANG 1977) AND THE THREENATURAL WAPPA PONDS

SURVEYED INTHIS STUDY. Significant differences exist between the means of the two studies for nitrate andpho sphate (P < 0.01, Kruskal-Wallis ANOVA)

SOURCE

Bienfang (1977)Present Study

N

8818

S (0/00)

6.6 ± 1.87.4 ± 2.8

NO )"

17.81 ± 9.5266.33 ± 17.70

1.17 ± 0.333.70 ± 1.26

NHt

0.94 ± 0.840.99 ± 0.30

WAPPA prior to any development. Meanconcentrations (JiM ± SO) are as follows: ni­trate 54 ± 3.4, phosphate 1.5 ± 0.4, and sili­cate 609 ± 100. Other than phosphate whichis low relative to the present study, these val­ues are similar. More water quality data forthe Waikoloa anchialine system prior to anydevelopment are available from Bienfang(1977). This author reported water qualitydata for 50 anchialine ponds overlapping withand in the vicinity of the WAPPA. The grandmeans for Bienfang's (1977) data as well asthose from the present study are given inTable 3. A Kruskall-Wallis ANOVA of thesedata point to statistically significant positivedifferences in the concentrations of nitrateand phosphate between the two studies. Thesesignificant increases may be related to the useof enriched irrigation water (i.e., sewage efflu­ent diluted with coastal well water) and fertil ­izers on the Waikoloa golf course constructedsubsequent to Bienfang's (1977) survey.

In conclusion , this study documents statis­tically significant increases in nitrate andphosphate concentrations in the Waikoloaanchialine ponds since 1977. Although statis­tically greater, nutrient concentrations in theWAPPA presently fall into the range ofvaluesobserved in the groundwater or other anchia­line ponds along relatively undeveloped sec­tions of the Kona coast. Periodic sampling ofthe biota in these ponds from 1972 (Maciolekand Brock 1974) to present has yielded noobvious changes, suggesting that the biota ofanchialine pond systems are insensitive to theincreased nutrient concentrations observed inthis study. It appears that these nutrient spe­cies are in excess and thus are not limiting.Possible mechanisms to this apparent bioticinsensitivity are the characteristic short water

residence time of ponds and the usual pres­ence of large numbers of endemic herbivo­rous crustaceans. Through their grazing thesecrustaceans appear to keep many macroalgalspecies from dominating the system. Any per­turbation affecting these mechanisms couldresult in major shifts in the structure of an­chialine pond communities.

ACKNOWLEDGMENTS

The authors wish to thank Drs. S. V. Smith,P. Bienfang, and J. H. Brock for their reviewof the manuscript. This work was supportedin part by the Waikoloa Anchialine Pond Pre­servation Area Trust Fund. HIMB contribu­tion no . 733.

LITERATURE CITED

ARMSTRONG, F. A. J., P. M. WILLIAMS, andJ. D. H. STRICKLAND. 1966. Photo-oxida­tion oforganic matter in sea water by ultra­violet radiation, analytical and other ap­plications. Nature 211 :481 -483.

BIENFANG, P. K. 1977. Survey of the aquaticbiota and water quality characteristics ofthe anchialine ponds at Anaehoomalu,Hawaii. Unpublished report prepared forBoise Cascade Co., Honolulu, HI. Novem­ber 1977. Oceanic Institute, Makapuu,Waimanalo, HI. 150 p.

BIENFANG, P. 1980. Water quality character­istics ofHonokohau Harbor: a subtropicalembayment affected by groundwater in­trusion. Pacif. Sci. 34: 279-29l.

CHANG, S. Y. K., and R. H. F . YOUNG. 1977.An investigation into environmental effects

208

of sewage effluent reuse at the KaneoheMarine Corps Air Station Klipper golfcourse. Water Resources Research CenterTech. Memorandum 53, University ofHawaii, Honolulu. 51 p.

Cox, D . c.. F. F. PETERSON, W. M. ADAMS,C. LAU, J . F . CAMPBELL, and R. D . HUBER.1969. Coastal evidence of groundwaterconditions in the vicinity of Anaehoomaluand Lalamilo . South Kohala, Hawaii .Water Resources Research Center Tech .Report. 24. Uni versity of Hawaii , Hono­lulu. 53 p.

HOLTHUIS, L. B. 1973. Caridean shrimpsfound in land-locked saltwater pools atfour Indo-West Pacific localities (SinaiPenin sula, Funafuti Atoll, Maui, and Ha­waii Islands), with the descriptions of onenew genus and four new species. Zoolo­gische Verhandelingen 128 : 1-48.

JEFFREY, S. W. 1974. Profiles ofphotosynthe­tic pigments in the ocean using thin-l ayerchromatography. Mar. BioI. 26 : 101-110.

JEFFREY, S. W., and G. W. HUMPHREY. 1975.New spectrophotometric equation for de­termining chlorophylls a, b, eland c2in higher plants, algae and natural phyto­plankton. Biochem. Physiol. Pflanzen 167:191-194.

JOHANNES, R. E. 1980. The ecological sig­nificance of the submarine discharge ofgroundwater. Mar. Ecol. Prog. Ser. 3 :365­373.

KANEHIRO, B. Y. 1977. A hydrogeologic studyof the Kiholo to Puako area on the islandof Hawaii. M. S. Th esis, Univ. of Hawaii,Honolulu. ix + 116p.

KAY, A. E., L. S. LAU, E. D . STROUP, S. J .DOLLAR, D. P. FELLOWS, and R. H. F.YOUNG. 1977. Hydrological and ecologicalinventories of the coastal waters of WestHawaii . Water Resources Research Cen­ter Tech . Report 105. Uni versity of Ha­waii, Honolulu. 94 p.

PACIFIC SCIE NCE , Volume 41, 1987

MACIOLEK, J. A., and R. E. BROCK. 1974.Aquatic survey of the Kona coa st ponds,Hawaii Island. Uni v. Hawaii Sea GrantAdvisory Report AR-74-04. 73 p.

KENSLEY, B., and D. WILLIAMS. 1986. Newshrimps (Family Procarididae and At­yidae) from a submerged lava tube onHawaii. J. Crust BioI. 6 :417-437.

MACIOLEK, J . A. 1983. Distribution and bi­olog y of Indo-Pacific insular hypogealshrimps. Bull. Mar. Sci. 33: 606-61 8.

MARSH, J . A., JR. 1977. Terrestrial inputs ofnitrogen and phosphorus on fringing reefsin Guam, p . 332-336. In: Proceedings ofthe 2nd International Coral Reef Sympo­sium , Vol. I. Great Barrier Reefcommittee,Brisbane, Au stralia.

MURPHY, J. , and J . P. RILEY. 1962. A modi­fied simple solution method for the deter­mination of phosphate in natural wat ers.Anal. Chim. Acta 27 :31- 26.

SAS INSTITUTEINC. 1985. SAS User's Guide:Stati stics, Version 5 Edition. SAS InstituteInc . 956 p.

SMITH, S. V., W. H. KIMMERER, E. A. LAWS,R . E. BROCK, and T. W. WALSH. 1981. Ka­neohe Bay sewage diversion experiment:Perspecti ves on ecosystem responses tonutritional perturbation. Pacif. Sci. 35 :279- 402.

SOLORZANO, L. 1969. Determina tion of am­monia in natural waters by the phenolhy­pochlorite method. Limnol. Oceanogr.14:799- 801.

STRICKLAND, J . D . H. , and T. R. PARSONS.1972. A practical handbook of seawateranalysis. 2nd. ed. Bull. Fish . Res . Bd. Can.,167.310 p.

SWAIN, L. A. 1973. Chemical quality ofground water in Hawaii. U . S. Geol. Sur­vey Report. R-48.