Leaching characteristics of anions and cations from …Bull. Minamikyushu Univ. 39A:57-66 (2009) Leaching characteristics of anions and cations from evergreen leaves supplied to the

Bull. Minamikyushu Univ. 39A: 57-66 (2009)

Leaching characteristics of anions and cations from evergreenleaves supplied to the stream bed and influences on stream

water composition in the Southern Kyusyu Mountains

Hirokazu Kitamura

Laboratory of Landscape Conservation, Minami-Kyushu University,Takanabe, Miyazaki 884-0003, Japan

Received October 8, 2008; Accepted January 28, 2009

The amount of allochthonous material that enters lotic ecosystems annually from terrestrialriparian forests is estimated to be considerable, and substances derived from the decompositionof leaf litter are thought to have an important impact on stream water quality. The purpose ofthe present study is to clarify the leaching characteristics of leaf litter of the evergreen speciesthat are dominant in the riparian zone of the investigated stream and to examine the influence ofleachates on mountain stream water ionic composition by laboratory leaching experiments andfield measurements.

Five fresh fallen leaves and fallen leaves of ten species of evergreen trees dominant in the ripar-ian zone of the investigated reach were placed into a beaker filled with 500cc of distilled waterand allowed to stand for 30 days at ambient temperature between 25~30

。C after well washing

with distilled water, air-dried for one week and oven-dried at 80。C for 48 hours and the electric

conductivity (EC)), hydrogen ion concentration ((pH)) and the concentrations of cations ((Li+, Na+,NH4

+, K+, Ca2+ and Mg2+)) and anions ((F－, Cl－, NO2－, Br－, NO3－, PO43－ and SO4

2－)) were measuredat 1 and 30 days after submersion. On the field measurements, stream water samples were col-lected from four different environments ((springs, riffles, pools and side-pools)) and were trans-ported to our laboratory and EC, pH and the concentrations of the cations and anions weremeasured. The field measurements were conducted monthly from April 2006 to March 2007.

EC, pH, and cation and anion concentrations, continued to increase in all samples for 30 daysfollowing the immersion of evergreen tree leaves in distilled water for being attributed to theleaching of some components from the leaves. The EC values of water samples containing sub-merged green Symplocos theophrastifolia Sieb. et Zucc. leaves exceeded 150 S/cm, and pH valuesincreased following the rapid decrease immediately after submergence, from 6.0 to 7.0. The mostdominant cation in the water samples was K+, accounting for more than 70% of all cations inmost sample waters. On the 30th day after submergence, the order of average cation releaserates from green leaves was K+ > Mg2+ > Ca2+ > Na+, which coincided to the experimental resultsin previous studies. The order of average anion release was Cl－ > PO4

3－ > SO42－. In contrast to

the high K+ ratios in the water of the leaching experiments, the monthly K+ concentrations in thestream water were relatively lower than those of the other cations and anions. The lower concen-tration of K+ in the stream water during the growing season would suggest that there is a differ-ential utilization of K+ by the biota and stream sediment acts as an important agent for removalof leachate from water. Therefore, the clay component of the sediment may serve as a reservoirof chemically bound K+ and biota may also provide a reservoir for K+ if there is some netincrease in biomass. The EC value is lower in the spring than in the other environments and thatthe K+, Mg2+ and Ca2+ concentrations are considerably higher in the riffle, pool and side-pool, inwhich litter had been supplied, than in the spring where leaf litter is not deposited.

These results indicate that the primarily source of ions is rock weathering in the spring andsimultaneous ion loading from multiple sources occurs in the riffle, pool and side-pool and thatleaf litter is one of the primary sources of K+, Mg2+ and Ca2+ in the stream water. Based on theseresults, it is demonstrated that anion and cation concentrations throughout the stream reach arenot always uniform due to heterogeneous distribution of ionic material caused by leachate fromleaves, although stream water has been regarded as providing a uniform continuous body for theionic materials, and changes in water composition due to the natural input and subsequent leach-ing from leaf litter were predicted.

Key words: evergreen trees, leaf litter, leachate, cations and anions, stream environments.

57

58 Leaching characteristics of ions from evergreen leaves supplied to the stream bed and influences on stream water composition

1. INTRODUCTION

Allochthonous material, such as leaf litter, enteringsmall streams represents the primary energy source avail-able to the consumers in these systems. This organic mat-ter has been shown to remain within streams where it isdecomposed by a variety of physical, chemical and bio-logical processes 1). Arising from the instream processingof this organic material is the production of dissolvedorganic matter (DOM) 6) 25), nutrients 26) and physicochemi-cal changes in water chemistry. The amount ofallochthonous material that enters lotic ecosystems annu-ally from terrestrial riparian forests is estimated to beconsiderable 7), and substances derived from the decom-position of leaf litter are thought to have an importantimpact on stream water quality.

Pioneering studies that described the effect of leaf litteron the chemical quality of streams, including Schneller(1955) 31) and Slack (1955) 33), reported that the decay ofleaves in pools generally induced greater water color andlow dissolved-oxygen concentrations. Hynes (1960) 12)

indicated that the decomposition of spruce and red cedarneedles released a substance toxic to fish, and Myers(1961) 27) suggested that manganese was leached from oaktrees in the watershed of the San Clemente Reservoir.These studies suggested that various tree species maymake widely differing chemical contributions to streamsthrough leaf litter leachates.

While many subsequent studies have focused on thephysical degradation of leaf litter 5) 11) 12) 22) 24) or on thechemical changes associated with degradation instreams 6) 18) 20) 23) 27), few studies have examined the effectsof materials leaching from leaf litter on stream waterquality. The purpose of the present study, therefore, is toclarify the leaching characteristics of leaf litter of theevergreen species that are dominant in the riparian zoneof the investigated stream and to examine the influence ofleachates on mountain stream water ionic composition.The present study focuses on one of the most importantstream water quality indices, anions and cations, asleachates from leaf litter in the stream water.

2. SITE DESCRIPTION

Experiments were conducted in the Takeo River inSaito City of Miyazaki Prefecture. The Takeo River, atributary of the Hitotsuse River, ranges in elevation from100m to 600m and originates in the Southern KyusyuMountains. The Takeo River Basin overlays the MiyazakiFormation, which was formed during the Cretaceous tomiddle Paleogene periods. The formation consists ofblack slate, sandstone and shale. The geologic structureof the basin, therefore, is characterized by the prevalenceof weak and extensively folded rock strata with numerousfaults that are susceptible to weathering. These geologicconditions have formed steep and unstable basin slopeswhere several mid-sized landslides have occurred in thelast fifty years. To prevent sediment disasters caused bylandslides resulting from the weakness of this basin,seven 3 to 5m- high sabo dams have been constructed inthe stream since 1965 18).

The study site consists of an approximately 1 km riverreach between sabo dam No. 5 and approximately 50mdownstream of sabo dam No. 6 (Fig. 1). The stream is afirst-order stream with a mean width of 10 m, mean depth

(at modal flow) of 0.3m, mean slope of 0.25, and meancurrent velocities ranging from 20―30 cm/sec. Thestream bottom is composed primarily of gravel, pebblesand cobble substrates and the riparian vegetation consistsof evergreen deciduous trees, such as Symplocostheophrastifolia Sieb. et Zucc., Machilus japonica Sieb.et Zucc., Meliosma rigida Sieb. et Zucc., and Litseaacuminata Kurata. Since the slope of the stream basin issteep, tree leaves supplied to the forest floor of the slopetend to be removed to the stream bed.

3. METHODS

3-1. Materials and laboratory leaching experimentsTen species of evergreen trees, shown in Table 1, are

common in the riparian zone of the investigated reachand make a substantial contribution to annual litter fall.Approximately fifty pieces of fresh fallen leaves and fall-en leaves of each tree species were collected from thestream bank of the reach. Fresh fallen leaves and fallenleaves are whole leaves categorized as follows. Freshfallen leaves maintain the original shape without anyskeletonization and still retain their original colorbecause of the relatively short time elapsed after falling.Fallen leaves also hold the original shape without anyskeletonization and appear to be brown. The leaves werecollected on May 30, 2006 because in the investigatedreach the evergreen species fall from May to July eachyear.

The collected leaves were well washed with distilledwater to eliminate aerosol dust and atmospheric gases 8),air-dried for one week and oven-dried at 80

。C for 48

hours. Five fresh fallen leaves and fallen leaves of eachspecies were selected at random and weighed. Then,each sample was placed into a beaker filled with 500ccof distilled water and allowed to stand for 30 days atambient temperature between 25~30

。C. The electric con-

ductivity (EC) and hydrogen ion concentration (pH) weremeasured with a water quality probe (WQC-20A, TOAElectronics Ltd., Japan), and the concentrations ofcations (Li+, Na+, NH4

+, K+, Ca2+ and Mg2+) and anions(F－, Cl－, NO2－, Br－, NO3－, PO4

3－ and SO42－) in the water

were measured using ion chromatography (DX-120,NIPPON DIONEX K.K.) at 1 and 30 days after submer-sion, because maximum loss of soluble components

Fig. 1. Stream reach investigated and sampling sites.

59

through abiotic leaching occurs during the first 24 hoursexposure of leaf litter to water and is completed withinapproximately four weeks 35).

3-2. Stream water sampling and estimation of sampleion content

Stream water samples were collected from four differ-ent environments (Fig.1): (a) springs from cracks on thevertical cliff of bedrock slope along the side of thestream approximately 150cm above the stream margin;(b) riffles; (c) pools; and, (d) side-pools, which were con-sidered to be stagnant areas of water along the streammargin. Since a series of sampling points (spring, riffle,pool and side-pool) are not continuously distributed, thewater quality of a sampling point dose not affect that ofeach neighboring sampling point.

Samples were collected by hand in a 250ml polyethyl-ene bottle. Water from springs was collected from cracksin the bedrock and 10cm below the water surface in rif-fles. In pools and side-pools, samples were collectedapproximately 5cm above leaf litter pack on the substrateusing a 250ml polyethylene bottle. A water sampler wasused to collect water in pools and side-pools that weredeeper than 50cm. At the time of sampling, water tem-perature, EC and pH of the water were measured. Thecollected water samples were transported to our laborato-ry and the concentrations of cations (Li+, Na+, NH4

+, K+,Ca2+ and Mg2+) and anions (F－, Cl－, NO2－, Br－, NO3－,PO4

3－ and SO42－) were measured. The measurements

were conducted using the same methods applied to thestream water described above.

These measurements were conducted monthly fromApril 2006 to March 2007.

4. RESULTS

4-1. Leaching characteristics of evergreen leaves4-1-1. Changes in EC and pH values of water fromsubmerged leaves

The pH and EC of the distilled water in which leafsamples were submerged varied with time as shown inTable 1. The initial EC and pH of distilled water was1.36μS/cm and 6.9, respectively. Marked increases inEC were observed one day after leaf submersion, indicat-ing that some materials leached from submerged leavesimmediately after submergence, resulting in rapid leach-ing during the initial 24 hours 10) 16) 38). The EC increasedwith time in all water samples with significant differ-ences observed after 30 days, and the mean EC values ofwater after 30 days' submergence of green leaves ofSymplocos theophrastifolia Sieb. et Zucc., Machilusjaponica Sieb. et Zucc., Meliosma rigida Sieb. et Zucc.,Litsea acuminata Kurata, and Machilus thunbergi Sieb.et Zucc. were 152.6, 116.0, 131.1, 144.9, and 133.4μS/cm, respectively. The mean EC values on the 30thday were higher in green leaves than in brown leaves,with the exception of Castanopsis Spach. Based on thesemean values, the increasing EC rates attributed to leach-ing from submerged leaves should tend to be lower inleaves with a relatively thicker outer layer of the epider-mis, which consists of cutin.

The mean pH values of water decreased from the ini-tial pH value one day after submersion, with the excep-tion of brown Symplocos theophrastifolia Sieb. et Zucc.leaves, which was attributable to aerosol dust adhered tothe leaves. The mean pH values increased with time inall water samples after 30 days' submergence, recovering

Leaching characteristics of ions from evergreen leaves supplied to the stream bed and influences on stream water composition

Days after submergence

EC30/1 ECg/bpH EC pH EC

301

EC30/1： The ratios of mean EC value after 30 days to that after one dayECg/b： The ratios of mean EC value after 30 days of green leaves to that of brown leaves

6.36.95.06.05.25.05.45.45.95.85.35.65.65.25.95.86.16.05.85.9

33.682.653.223.8

129.389.746.139.66.77

10.0319.4016.4514.4133.96.237.467.6524.334.19.93

6.17.07.06.87.16.97.17.06.86.66.36.36.76.66.56.67.06.86.86.4

152.6126.4116.051.8

131.194.7

144.976.037.522.224.326.664.947.741.833.776.449.7

133.425.6

4.541.532.182.181.011.063.141.925.542.211.251.624.501.416.714.529.992.053.912.58

1.21

2.24

1.38

1.91

1.69

0.91

1.36

1.24

1.54

5.21

green leavesbrown leaves










CategoriesSpecies

Quercus myrsinaefolia Blume

Pasania edulis Nakai

Machilus thunbergii Sieb. et Zucc.

Litsea acuminata Kurata

Quercus gilva Blume

Castanopsis Spach

Quercus glauca Thunb.

Symplocos theophrastifolia sieb. et Zucc.

Machilus japonica sieb. et Zucc.

Meliosma rigida sieb. et Zucc.

Table 1. Change in pH and EC (μμS/cm) of water containing submerged leaves

60Leaching characteristics of ions from

evergreen leaves supplied to the stream bed and influences on stream

water com

position


Species Category



Average

Quercus gilva Blume

Castanopsis Spach







F－ Cl－ NO3－ PO43－ SO4

2－ total F－ Cl－ NO3－ PO43－ SO4

2－ total F－ Cl－ NO3－ PO43－ SO4

2－

Species Category Na＋



Average

Quercus gilva Blume

Castanopsis Spach







（a）cation


green leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leaves

（b）anion

0.1380.3080.0890.4120.6850.7010.2200.7980.2790.2662.6530.3980.3120.1450.4800.0690.1530.5280.2270.3050.5240.393

1 30 Rk/t

0.3000.1830.1200.1980.0790.0620.0560.1340.0000.1140.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0550.069

NH4＋

3.7413.5554.4014.533

11.82713.520

4.0176.0131.6952.7206.2193.6251.3381.1722.9480.4840.7012.2793.9801.3974.0873.930

K＋

0.0720.6740.0890.1320.6590.4930.0330.1230.0180.0460.9640.1820.0890.0980.1030.0080.0330.0390.0510.1120.2110.191

Mg2＋

0.1460.8980.1410.1870.5440.7670.0400.1940.0570.0970.4560.1700.2160.1070.0980.0190.0800.0610.0880.1330.1870.263

Ca2＋

4.3975.6184.8405.463

13.79515.544

4.3667.2632.0503.243

10.2924.3751.9561.5223.6290.5800.9672.9074.3471.9475.0644.846

total0.1220.1790.0990.2380.2480.2930.2010.4080.0950.1230.4350.2320.1910.0510.2620.0530.0870.3030.2320.2630.1970.214

Na＋

0.5930.6650.0000.0000.3550.4110.0450.4340.9670.7060.6240.5270.2350.0000.2450.0300.0000.0200.0000.0000.3070.279

NH4＋

15.82810.785

9.3275.128

12.55814.44415.12815.504

8.6847.3508.3274.1528.5961.166

11.0821.9827.4996.677

16.8509.027

11.3887.621

K＋

2.0632.5860.3381.1120.7250.5520.1390.5040.3260.6731.8660.6360.7560.1890.8750.1620.1360.2280.3170.4080.7540.705

Mg2＋

1.9682.3790.2042.1170.8280.7260.1020.2900.2290.5691.1390.4800.3060.1340.6070.1360.1040.2460.2210.3410.5710.742

Ca2＋

20.57516.593

9.9688.594

14.71416.42615.61517.14110.300

9.42112.392

6.02810.085

1.54113.072

2.3647.8277.473

17.62010.03913.217

9.562

total0.61.11.02.81.71.81.32.40.91.33.53.91.93.32.02.31.14.01.32.61.52.2

Na＋

2.94.00.00.02.42.50.32.59.47.55.08.72.30.01.91.30.00.30.00.02.32.9

NH4＋

76.965.093.659.785.387.996.990.584.378.067.268.985.275.784.883.895.889.495.689.986.279.7

K＋

10.015.6

3.412.9

4.93.40.92.93.27.1

15.110.6

7.512.3

6.76.91.73.11.84.15.77.4

Mg2＋ Ca2＋

9.614.3

2.024.6

5.64.40.71.72.26.09.28.03.08.74.65.71.33.31.33.44.37.8

0.1140.0890.0000.0000.0000.0000.0440.0700.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0160.016

1.6301.5800.3520.4000.6730.7410.5791.1400.5920.7081.0700.6510.8560.1340.7720.2980.2440.4191.1450.8860.7910.696

0.6500.3520.0831.9940.2880.2180.0530.0780.0000.0000.0000.0000.0000.2270.5670.0970.2770.0000.0000.0000.1920.296

2.71.80.00.00.00.02.52.70.00.00.00.00.00.00.00.00.00.00.00.00.70.9

39.133.034.515.914.620.933.044.646.2

100.0100.0100.0

48.237.251.875.516.966.729.849.535.238.8

15.67.38.2

79.36.26.13.03.00.00.00.00.00.0

62.838.124.519.2

0.00.00.08.5

16.5

0.00.0

44.60.0

38.019.046.026.753.8

0.00.00.0

27.60.00.00.0

58.533.351.626.531.311.4

42.557.912.7

4.841.154.015.423.0

0.00.00.00.0

24.20.0

10.20.05.40.0

18.624.024.232.5

1 30 Rk/t

green leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leavesgreen leavesbrown leaves



4.1644.7941.0192.5164.6123.5521.7542.5551.2820.7081.0700.6511.7760.3611.4900.3951.4430.6283.8491.7902.2461.795

1.7702.7740.1300.1221.8981.9180.2700.5870.0000.0000.0000.0000.4300.0000.1520.0000.0780.0000.7160.4300.5440.583

0.0000.0000.4540.0001.7540.6750.8080.6810.6900.0000.0000.0000.4910.0000.0000.0000.8440.2091.9880.4740.7030.204

0.0780.0990.6760.1930.4340.4560.0310.1730.1290.2550.2190.2040.0670.0340.2260.0280.0360.1000.7890.0860.2680.163

1.2321.0940.1770.4170.6700.6600.5761.2150.3310.4640.4240.4400.4790.1640.6260.2210.1690.3940.9540.4720.5640.554

0.3710.2060.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0000.0370.021

0.0000.0000.4060.0002.2041.1010.0850.0000.2010.0000.0000.0000.0000.0770.0000.0000.0000.0000.6960.0000.3590.118

0.3231.4220.1420.1262.4212.4170.3290.7190.0000.0000.0000.0000.1940.0510.1060.0000.0640.0000.6380.1600.4220.489

2.0042.8221.4020.7365.7294.6341.0212.1080.6600.7180.6430.6440.7400.3260.9580.2490.2700.4933.0770.7171.6501.345

Table 2. Changes in cation ((anion)) concentration par 1g of leaves ((mg/l)). Rk/t is the ratio of cation ((anion)) concentration to total cation ((anion)) concentration on the 30th day fromsubmergence of leaves

to the initial pH value attributable to leaching. UnlikeEC, there were no marked differences in pH after 30days of being submerged among tree species and cate-gories of leaves, although many complex interactionsbetween pH and nutrient cycles were observed and theireffect on accelerating the decomposition of leaves inaquatic ecosystems will be explained below 10) 11) 13) 38).

4-1-2. Ion concentrations leached from submerged leavesTable 2 shows the concentrations of cations and anions

leached from leaves submerged in distilled water. Intable 2, the concentrations per 1g leaves are indicated.The ions detected in all samples during the experimentperiod were Na+, K+, Mg2+, Ca2+ and Cl－. The cations andanions appeared to leach from the leaves immediatelyafter submersion and the ion concentrations increasedwith time in all water samples. The mean total concentra-tions of K+, Mg2+, Ca2+, Na+ and Cl－ per 1g leaves on the30th day were 11.39, 0.75, 0.57, 0.20 and 0.79 mg/l,respectively. The order of leaf type from highest to low-est total cation concentration after 30 days wasSymplocos theophrastifolia Sieb. et Zucc. green leaves,Machilus thunbergii Sieb. et Zucc. green leaves, andLitsea acuminata Kurata brown leaves, with total cationconcentrations of 20.58, 17.62 and 17.14mg/l, respec-tively. The water in which brown Castanopsis Spach,Quercus glauca Thunb., and Pasania edulis Nakai leaveswere submerged had the lowest total cation concentra-tions of all the leaves examined, below 7.00mg/l.

Among the cations, K+ was leached most abundantlyby all plant species in the present study. The ratios of K+

concentration to total cation concentration on the 30thday exceeded 95% in water with submerged green Litseaacuminata Kurata, Pasania edulis Nakai, and Machilusthunbergii Sieb. et Zucc. leaves, and greater than 90% inwater with submerged Machilus japonica Sieb. et Zucc.green leaves and Litsea acuminata Kurata brown leaves.Relatively low Rt/k values less than 70% were observedin water with Symplocos theophrastifolia Sieb. et Zucc.brown leaves, Machilus japonica Sieb. et Zucc. brownleaves and Castanopsis Spach leaves.

The anion concentrations on the 30th day of after leafsubmergence were in the same range as those of cations,with the exception of K+. The difference in ion concen-trations between categories depended on the leaf speciesand varieties of ion.

4-2. Stream water fluctuations4-2-1. Overall comparisons of measured values instream water

The ranges and means of stream depth and measuredparameters, corresponding to the environments of therespective sampling sites during the experiment period,are shown in Table 3. The mean water temperature in thespring was 1

。C higher than the other stream environments

sampled. The highest pH value was observed in thespring and the mean pH values were lowest in the side-pool. The range and mean of pH in the riffle was exactlythe same as those in the pool, indicating the medianvalue between spring and side-pool. The observed ECvalues were extremely low in the spring but higher at theother sample sites.

61Leaching characteristics of ions from evergreen leaves supplied to the stream bed and influences on stream water composition

Spring

Riffle

Pool

Side-pool

Depth（cm） Water temperature（℃）

range

―

5－40

80－130

5－30

mean range mean range mean range mean

―

16.3

99.7

21.4

pH EC（μS/cm）

59.3

86.4

85.2

82.0

7.3

7.4

7.4

7.2

6.4－8.1

6.6－8.0

6.6－8.0

6.5－7.8

16.5

15.3

15.3

15.4

7.0－22.6

8.2－22.7

8.2－22.7

6.7－22.7

37.2

66.0

69.4

52.3

100.6

99.2

97.0

100.3

－

－

－

－

Table 3. Overall comparison of ranges and means of measured values

Spring 1.693－6.597 3.562 0.001－0.648 0.116 0.168－0.836 0.454 0.012－2.732 1.276 0.003－8.546 2.831Riffle 2.181－3.894 2.796 0.001－0.317 0.055 0.473－1.359 0.653 1.840－3.740 2.828 6.042－13.839 9.368Pool 2.156－5.808 2.819 0.001－0.879 0.095 0.455－1.646 0.652 1.041－3.853 2.715 1.660－13.559 8.856Side-pool 2.189－4.874 3.071 0.002－0.296 0.070 0.470－2.872 0.943 1.170－3.692 2.658 4.242－13.765 8.486

range mean range mean range mean range mean range mean

(b) cation

Spring 0.001－0.137 0.055 2.240－6.144 4.691 0.168－3.105 1.576 1.834－6.954 4.420Riffle 0.038－0.116 0.062 3.554－5.696 4.309 1.865－4.046 2.572 3.819－6.530 5.217Pool 0.001－0.129 0.061 3.609－10.036 4.491 1.758－3.933 2.494 2.154－6.888 5.061Side-pool 0.033－0.106 0.061 3.560－7.575 4.818 0.039－0.039 0.039 3.607－6.635 5.212

range mean range mean range mean range

(a) anion

mean

　　　　　　　　　　Na＋　　　　　　　　　NH4＋　　　　　　　　　K＋　　　　　　　　　Mg2＋　　　　　　　　Ca2＋

　　　　　　　　　　　F－　　　　　　　　　　　　 Cl－　　　　　　　　　 NO3－　　　　　　　　 SO42－

Table 4. Overall comparison of ranges and means of ion concentrations in the stream water (mg/ l)


The ranges and means of each ion concentration in thestream water, corresponding to the environments, areshown in Table 4. The measurements in September weredisregarded due to an experimental apparatus malfunc-tion. In the present study, Li+ (cation) and NO2－, Br－,PO4

3－ and F－ (anions) were not detected in the streamwater during the measurements. The mean ion concentra-tions were lowest in the spring and highest in the side-pool, with the exception of NO2－, Na+, and NH4

+.Furthermore, the mean concentrations of K+, Mg2+, andCa2+ were two to three times higher in the side-pool thanin the spring. Other than the three springs that were sam-pled in the present study, no additional springs or tribu-taries flow into the reach and there are no artificialsources of ions. Consequently, the stream water wasinferred to contain naturally ionic material in the reachinvestigated.

4-2-2. Monthly fluctuations in anion and cation con-centrations

The monthly mean concentrations of the major ionscorresponding to the environments of the respective sam-ple sites during the experiment period are shown inFig.2. Each ion concentration shows a unique monthlyfluctuation. Na+ and Cl－ indicated similar seasonal fluc-tuations in concentration for all stream bottom types. Na+

is generally found in association with Cl－, indicatingtheir common origin. Although weathering of NaCl-con-taining rocks accounts for most of the Na+ found in riverwater, the constant supply of Na+ and Cl－ through rain-water input is thought to originate from sea water in thewatershed and contributes significantly to the ion supplyalong the coasts 1).

Mg2+ and Ca2+ concentrations were lowest in the springthroughout the experiment period. The monthly meanconcentrations of Mg2+ and Ca2+ in the side-pool equaledto those in the riffle and pool. The mean K+ concentra-tions were highest in the side-pool except in Novemberand lowest in the spring throughout the experiment peri-

od, as observed for Mg2+ and Ca2+. The mean K+ concen-trations of the side-pool and pool were found to increaseduring the warm period from May to August. The con-centrations in the spring slightly fluctuated from a mini-mum value of 0.27mg/l in February.

The SO42－ concentrations in the riffle, pool, and side

pool were observed to fluctuate similarly with a higherperiod from November to April and lower period fromMay to October. Those in the spring fluctuated each monthwithin a range from a minimum value of 3.58 (mg/l) inOctober to a maximum value of 5.52 (mg/l) in November.

The results of multivariate analysis of monthly changesin mean concentrations of each ion (independentvariable) for each sampling site (riffle, pool, and side-pool) and sampling season divided into four temporalgroups (March to May, June to August, September toNovember, and December to February) (dependent vari-able) indicated significant relationships among thedependent variables (P<0.005). Furthermore, Sheffe'smultiple range test indicated a significant difference inthe monthly change in SO4

2－ concentrations between thespring and the side-pool and riffle (P<0.005), betweenthe side-pool and the other sites for K+ (P<0.001) andbetween the spring and the other sites for Mg2+ and Ca2+

(P<0.0001).

5. DISCUSSION

5-1. Differences in K+ concentrations between labora-tory experiments and field measurements

In the present study, physicochemical parameters, suchas EC, pH, and cation and anion concentrations, contin-ued to increase in all samples for 30 days following theimmersion of evergreen tree leaves in distilled water.This is attributed to the leaching of some componentsfrom the leaves. The EC values of water samples con-taining submerged green Symplocos theophrastifoliaSieb. et Zucc. leaves exceeded 150S/cm, and pH values

Fig. 2. Monthly fluctuations in major ion concentrations of the site.


increased following the rapid decrease immediately aftersubmergence, from 6.0 to 7.0. The most dominant cationin the water samples was K+, accounting for more than70% of all cations in most sample waters. On the 30thday after submergence, the order of average cationrelease rates from green leaves was K+ > Mg2+ > Ca2+ >Na+, which coincided to the experimental results in pre-vious studies 4) 10) 24). The order of average anion releasewas Cl－> PO4

3－> SO42－.

In contrast to the high K+ ratios in the water of theleaching experiments, the monthly K+ concentrations inthe stream water were relatively lower than those of theother cations and anions, reflecting the lowest abundanceand least variability of K+ in river water among the majorcations 1). In a previous study 20), the author found that theevergreen tree defoliation period in the riparian forests ofthe study reach is from May to July and the supply anddeposition of newly fallen leaves are accelerated duringthis period, promoting subsequent litter decomposition inthe investigated area. Furthermore, K+ concentrations instream water generally increased with increasing dis-charge21). In the study reach, stream discharge is higherfrom spring to summer and lower during autumn andwinter. K+ concentrations in the stream water, therefore,were predicted to increase during the period from May toJuly because of increased K+ supply by leaching fromfallen leaves and higher discharge. Contrary to thisexpectation, the K+ concentrations in the stream waterwere considerably lower than the other major cations.

The lower K+ concentrations may have been observedbecause K+consumption is increased during plant andanimal growing seasons as an indispensable mineral forplant and animal growth. In particular, plant growth dur-ing the warm period coincides with the high K+concen-tration period and this seasonal change in biologicaldemand corresponds to seasonal changes in flow condi-tions. The lower concentration of K+ in the stream waterduring the growing season would suggest that there is adifferential utilization of K+ by the biota. K+appears to becompletely water soluble in plants but may combine withorganic compounds 24) and stream sediment acts as animportant agent for removal of leachate from water 22).Therefore, the clay component of the sediment may serveas a reservoir of chemically bound K+. Biota may alsoprovide a reservoir for K+ if there is some net increase inbiomass. These biological consumption may occur inPO4

3－ which was not detected in the stream water duringthe measurements.

5-1-2. Possibility of alteration of stream water compo-sition from leaf leachate

Minerals in stream water originate from varioussources. For example, it is said that Mg2+ and Ca2+ instreams originate almost entirely from the weathering ofsedimentary carbonate rocks, and approximately 90% ofK+ originates from the weathering of silicate materials,especially potassium feldspar and mica1). Although pol-lution and atmospheric inputs are small sources, atmos-pheric inputs are minimal, and pollution contributes onlyslightly. Based on the results of the leaching experimentin the present study, however, biomass contribution tostream water composition, i.e. through leachate from leaflitter, is also pronounced. Although there are numerousmethods for detecting the origins of stream water compo-sition, differences in composition between stream waterand groundwater are due only to the ratios of rock weath-ering 8).

SO42－ was detected every month at all sampling sites

in the stream water but was not detected in the leachatesfrom evergreen tree species in the leaching experiments;thus rock weathering can be identified as the primarySO4

2－ source, although atmospheric gasses and aerosoldust may contribute the certain part of the SO4

2－ source.Consequently, the concentrations of the major ions thatcame from other sources, such as Cl－, K+, Mg2+, Ca2+

and Na+, should vary independently of SO42－ concentra-

tion. Figure 3 shows the correlation of the monthly meanconcentrations in the stream water between SO4

2－ andthe other major ions. In the figure, relatively positive cor-relations are observed, the ion concentrations increasewith increasing SO4

2－ concentration in the spring, withthe exception of K+ and Cl－ which is contributed throughrainwater input from sea salts. Conversely, positive cor-relations between SO4

2－ and the other major ions are notobserved in the other environments, with the exceptionof Ca2+. These results indicate that the primarily sourceof ions is rock weathering in the spring and simultaneousion loading from multiple sources occurs in the riffle,pool, and side-pool. Sulfide minerals occur in minorquantities in many different rock types, particularly infine-grained black shale 36), which is dominant in thestudy watershed. During certain months, the concentra-tions of SO4

2－ were higher in the riffle, pool, and side-pool than in the spring, which is thought to be attributedto leaching from the black shale substrates in the gravel,pebbles, and cobble of the stream bottom.

Based on these results, changes in water compositiondue to the natural input and subsequent leaching fromleaf litter are predicted. This hypothesis is supported bythe field measurements in the present study, i.e. that theEC value is lower in the spring than in the other environ-ments and that the K+, Mg2+, and Ca22+ concentrationsare considerably higher in the riffle, pool, and side-pool,in which litter had been supplied, than in the springwhere leaf litter is not deposited.

The previous study 20) clarified that leaves from riparianforests are input from May to July and tend to deposit in

Fig. 3. Corrlation between SO42- concentration and the

major ion concentrations at four different environments.


the side-pools of the reach for more than one month.Concentrations of the most abundant ionic leachate fromleaves, K+, are higher in the side-pool than in the otherstream environments in the present study. These findingsindicate that a natural ionic component is found in streamwater with high plant litter content. The relatively highconcentration of K+ in the pool during the defoliationperiod of the evergreen trees, therefore, should also cor-respond with the litter leaching process. The possiblealteration of K+, Mg2+, and Ca2+ concentrations in streamwater due to leachate from leaf litter during litter decom-position in the stream is supported by the previous exper-imention 3) 10) 16).

Generally, new leaf litter readily leaches to streamwater, and depending upon the species, may release from5 to 30% of its organic dry weight within 24 hours asdissolved organic matter 1) 25). Previous experiments havealso shown that anions and cations continue to leachfrom leaf litter for more than four weeks 19). In a naturalstream, allochthonous material, such as leaves, suppliedfrom terrestrial riparian forests into water is decomposedby a variety of physical, chemical, and biologicalprocesses. In particular, seasonal fluctuations occur inthe densities of invertebrates in the investigated reaches(Kitamura et al. 2003, unpublished). These invertebrates,such as Neoperla Needham, Goerodes japonicus andAnisocentropus immunis McLachlan, shatter leaves intofragments by their feeding and nesting activities, acceler-ating litter decomposition. These seasonal biological fac-tors may affect the leaching rates and the leaching ratesof leaves in natural streams should be higher than thoseof the experimental results in the present study.

6. CONCLUSIONS

The present study demonstrated that anion and cationconcentrations throughout the stream reach are notalways uniform due to heterogeneous distribution ofionic material caused by leachate from leaves, althoughstream water has been regarded as providing a uniformcontinuous body for the ionic materials.

For significant leaching to occur, leaves must reside ata site for a sufficient period of time. In addition, otherfactors are also important for the retention of leaves,including channel configuration and the area of the ripar-ian zone 5); the hydrological and substrate characteristicsalong the stream margin 34); the seasonal patterns of litterfall and the discharge characteristics of the stream 39); theamount of leaf litter entering a stream 28); and, biologicalconditions such as shredder density14) 17) 28) 30). Consequently,marked leaching in the side-pools should occur becausethese sampling points satisfy the above conditions forpromoting litter retention.

However, not all of the litter supplied at a sample sitein the present study was retained at the site, as most litterinput from riparian forests into lotic systems is transport-ed downstream. Accordingly, the litter characteristics ata site do not always reflect the water quality characteris-tics of that site. Litter decomposition and leaching ofionic materials from litter may be accelerated in naturalstreams because litter supplied into stream water isdecomposed via multiple processes, including mechani-cal decomposition by invertebrates and chemical decom-position based on residence time in the various streamenvironments.

The stream environments reflect the geomorphological

characteristics of the stream reach with sabo dams andshow uniform substrates consisting of similar sized parti-cles. Side-pools along the stream margins are promotedby the uniform cross-sectional distribution of low currentvelocity and shallow water depth. Substrates along thestream margins are more likely to trap leaves than thesame substrates in the main stream channel. The pres-ence of the sabo dams combined with the interactionsbetween the hydrological and substrate characteristicsalong the stream margin further increase the potential forretention 34). Furthermore, given that the increased reten-tion times attributed to the sabo dams also contribute tocreating an environment suitable for colonization byshredder fauna 30), litter decomposition in reaches withsabo dams is likely to be promoted further. Thus, com-parative studies of litter decomposition rates in reacheswith and without sabo dams are needed in the future.

Further experiments are also required to determinewhether substances other than anions and cations areleached from plants, including lignin, cellulose, hemicel-lulose 35), iron and manganese15). The nature of the leachedsubstances and the rates at which this occurs vary amongplant species and the parts of the plant, as well as accord-ing to the stage of plant decomposition9) 23) 29). Future studiesshould evaluate input of these substances to stream watercomposition.

REFERENCES

1 ) Allan J.D.: Stream ecology, Chapman and Hall, p.388(1995).

2 ) Boyd C.E.: Losses of mineral nutrients during decomposi-tion of Typha latifolia, Arch.Hydrobiol., 66(4), 511-517(1970).

3 )Brinson M.M.: Decomposition and nutrient exchange oflitter in an alluvial swamp forest, Ecology, 58, 601-609(1977).

4 ) Boyd C.E.: Losses of mineral nutrients during decomposi-tion of Typha latifolia, Arch. Hydrobiol., 64(4), 511-517(1970).

5 ) Cummins K.W., Wilzbach A.M., Gates D.M., Perry J.B.and Taliaferro W.B.: Shredders and riparian vegetation,Bioscience, 39(1), 24-30 (1989).

6 ) Dahm C.N.: Pathway and mechanisms for removal of dis-solved organic carbon from leaf leachate in stream,Can.J.Fish.Aquat.Sci., 38, 68-77 (1981).

7 ) Egglishaw H.J.: The quantitative relationship between bot-tom fauna and plant detritus in streams of different calciumconcentrations, Journal of Applied Ecology, 5, 731-740(1968).

8) Elizabeth E.L. and Robert A.B.: The global water cycle,Prentice-Hall, Inc., pp.397 (1987).

9 ) Gene E.L.: Effects of forest cutting and herbicide treatmenton nutrient budget in the Hubber brook watershed-ecosys-tem, Ecological Monographs, 40(1), 23-47 (1969)

10) Hodkinson I.D.: Dry weight loss and chemical changes invascular plant litter of terrestrial origin, occurring in abeaver pond ecosystem, B.APP.E., pp 131-142 (1974).

11) Hofsten B.V. and Edberg N.: Estimating the rate of degra-dation of cellulose fibers in water, Oikos, 23, 29-34 (1973).

12) Hynes H.B.: The biology of pollited waters, LIverpoolUniv.press, 202p (1960).

13) Judyl M. and Carol J.: The influence of elevated nitrateconcentration on rate of leaf decomposition in a stream,Freshwater Biology, 13, 177-183 (1983).


14) Kaushik N.K. and Hynes H.B.: The fate of the deadleaves that fall into streams, Arch.Hydobiol.,68, 465-515(1971).

15) KeithV.S. and Washington D.C.: Effect of leaves onwater Quality in the Cacapon river, Art.161 in U.S. GEOL.Survey Prof. 475-D, pp.181-185 (1964).

16)Killingbeak K.T., Smith D.L., and Marzof G.R.:Chemical cahnges in the tree leaves during decompositionin a Tallgras Prairie Stream, Ecology, 63(2), 585-589(1982).

17) King J.M., Day J.A. and Davies B.R.: Leaf-pack dynam-ics in a southern African mountain stream, FreashwaterBiology, 18, 325-340 (1987).

18) Kitamura H.: Studies on the slope movement and counter-measures of landslides in the Southern Part of the KyusyuMountains, Journal of the Japan Society of RevegetationTechnology, 18(1), 12-18 (1992) (in Japanese).

19) Kitamura H.: Estimation of cation and anion leachingcharacteristics and dissolved BOD-adsorption capacity ofhomemade charcoal produced by community participation,Bulletin of Minamikyushu University, 36(A), 7-19 (2006).

20) Kitamura H.: An extreme decrease in dissolved oxygenconcentrations resulting from litter decomposition in amountain stream in the Southerm Kyusyu Mountain,Bulletin of Minamikyusyu University, 37(A), 57-68 (2007).

21) Likense G.E., Bormann F.H, Johnson M.N., and PierceR.S.: The Calcium, magnesium, potassium, and sodiumbudgets for a small forested ecosystem, Ecology, 48(5),772-785 (1967).

22) Lock M.A. and Hynes H.B.: The fate of "dissolved"organic carbon derived from autumn-shed maple leaves(Acer saccharum) in a temperate hard-water stream,Limnology and Oceanography, 21(3), 436-443 (1976).

23) Lousier J.D. and Parkinson D.: Litter decomposition in acool temperate deciduous forest, Can.J.Bot, 54, 419-436(1975).

24)Lousier J.D. and Parkinson D.: Chemical elementdynamics in decomposing leaf littetr, Can.J.Bot., 56, 2795-2812 (1978).

25) McDowell W.H. and S.G.Fisher: Autumnal processing ofdissolved organic matter in a samll woodland streamecosystem, Ecology, 57, 561-569 (1976).

26)Meyer J.L.: Dynamics of phosphorus and organic matterduring leaf decomposition in a forest stream., Oikos, 34,44-53 (1980).

27) Myers H.C.: Manganese deposits in western reservoirs anddistribution system, Water Works Assoc. Jour., 3(5), 579-588 (1961).

28) Prochazka K., Varbara A.S. and Davise B.R.: Leaf litterretention and its implication for shredder distribution in twoheadwater stream, Arch. Hydrobiol., 120(3), 315-325(1991).

29)Robert C.P.and Kenneth W.C.: Leaf processing in awoodland stream, Freshwater Biology, 4, 343-368 (1974).

30) Rounick J.S. and Winterbourn M.J.: Leaf processing intwo contrastiong beech forest streams, Effects of physicaland biotic factors on litter breakdown, Arch, Hydrobiol., 96(4),448-474 (1983).

31) Schneller M.V.: Oxygen depletion in Salt Creek, Indiana,in v.4 of in of investigation of Indiana Lakes and streams,Indiana Univ.Dept.Zoology, pp.163-175 (1955).

32) Slack K.V.: A study of the factors affecting stream produc-tivity by the comparative method, in v.4 of in of investigationof Indiana Lakes and streams, Indiana Univ.Dept.Zoology,pp.3-47 (1955).

33) Slack K.V.: Effect of tree leaves on water quality in thecacapon river, West Virginia, U.S.Geol. Surv. Profess.

Papers 475-D, pp181-185 (1964).34) Speaker R., Moore K. and Gregory G.: Analysis oh the

process of retention of organic matter in stream ecosys-tems, Verh.Internat.Verein.Limnol., 22, 1835-1841 (1984).

35) Suberkropp K., Godshalk G.L., and Klug M.J.; Changesin the chemical composision of leaves during processing ina woodland stream, Ecology, 57, 720-727 (1976).

36) Takaya S.: Experimental study on leachate from blackshales of Miyazaki Formation, Proceeding of KyushuBranch Japan Society of Engineering Geology, (2000) (inJapanese).

37) Thomas W.A.: Weight and calcium loss from decompos-ing tree loeave on land and water, B.APP.E., pp.237-241(1969).

38)Webster J.R. and Benfield E.F.: Vascular plant break-down in freshwater ecosystems, Ann.Rev.Ecol.Syst., 17,567-594 (1986).

39)Winterbourn M.J.: Fluxes of litter falling into a smallbeech forest stream, N.Z.Journal of Marine and FreshwaterReserch, 10(3), 399-416 (1976).

南九州地域の山地渓流河畔域に生育する常緑広葉樹 leaf litterの溶出特性と渓流水質に及ぼす影響について

北村泰一南九州大学環境造園学部地域環境学科

緑地保全学研究室

要　約

河畔域から渓流に供給されるリターは膨大な量にのぼると推測され，リターからの溶出物は渓流水質の形成に重要な役割を果たすものと思われるが，渓流生態系は物質の一方向への移入が卓越する開放系であるため溶出成分はごく短期間に流出するであろうという先入観から，河畔域より渓流に供給されるリターの渓流水質への影響は注目されてはこなかった．そこで本研究は，室内溶出実験と現地調査により，南九州地域の山地渓流河畔域に広く生育する常緑広葉樹リターの溶出特性の概要を明らかにし，それが渓流水質の形成に及ぼす影響を類推することを目的として行った．対象流域の河畔域に優占する常緑広葉樹（10種）の落葉

（落葉直後のものと落葉後数日を経たもの）を採取し，十分な洗浄と乾燥の後，各5枚ずつ500ml の蒸留水中に投入し，投入から1日および30日後の水中の電気伝導度（ＥＣ），ｐＨ，陽イオン（Li+, Na+, NH4+, K+, Ca2+ , Mg2+）と陰イオン（F－,Cl－, NO2－, Br－, NO3－, PO43－, SO42－）濃度を測定した．さらに，2006年4月～2007年3月の毎月1回，対象渓流のspring，riffle，pool，side-pool各3箇所から採水し同様の項目を測定した．室内実験においては，リターからの溶出に起因して，ＥＣ，ｐＨ，陽イオン，陰イオン濃度は実験開始直後からすべてのサンプルで増加し，実験から30日後ではK+，Mg2+，Ca2+，Na+ が主要陽イオンを構成しこのうちK+ が70％以上を占めた．これに対して現地渓流では生物体への吸収や底質粘土粒子への吸着などによる消失のため，K+ は検出されたイオンのなかでは低い濃度を示した．さらに現地渓流では，リターが供給され滞留する傾向にあるpool，side-poolにおいてK+，Mg2+，Ca2+ 濃度がspringより高く，この傾向はK+ で特に著しいものであった．


以上の結果から，常緑広葉樹が優占する山地渓流では，河畔域から渓流に供給されるリターの初期分解過程である溶出段階において放出されるK+，Mg2+，Ca2+ が渓流生態系にお

ける重要な供給であり，渓流水中のイオン組成を規定する重要な因子となりうることが示唆された．

Documents

Leaching characteristics of anions and cations from …Bull. Minamikyushu Univ. 39A:57-66 (2009) Leaching characteristics of anions and cations from evergreen leaves supplied to the