Urbanization affects food webs and leaf-litter decomposition in a tropical stream in Malaysia

Urbanization affects food webs and leaf-litter decomposition in a tropical stream in MalaysiaAuthor(s): Catherine M. Yule, Jing Ye Gan, Tajang Jinggut and Kong Ving LeeSource: Freshwater Science, (-Not available-), p. 000Published by: The University of Chicago Press on behalf of Society for Freshwater ScienceStable URL: http://www.jstor.org/stable/10.1086/681252 .

Accessed: 19/03/2015 06:31

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

The University of Chicago Press and Society for Freshwater Science are collaborating with JSTOR to digitize,preserve and extend access to Freshwater Science.

http://www.jstor.org

This content downloaded from 175.142.179.86 on Thu, 19 Mar 2015 06:31:17 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=ucpress

http://www.jstor.org/action/showPublisher?publisherCode=nabs

http://www.jstor.org/stable/10.1086/681252?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


TROPICAL STREAMS

Urbanization affects food webs and leaf-litterdecomposition in a tropical stream in Malaysia

Catherine M. Yule1,2, Jing Ye Gan1,3, Tajang Jinggut1,4, and Kong Ving Lee1,5

1School of Science, Monash University, Jalan Lagoon Selatan, 47500 Bandar Sunway, Selangor, Malaysia

Abstract: Urbanization is occurring rapidly in southeastern Asia where streams are increasingly pressured. Weassessed the ecological integrity of Ampang River in Kuala Lumpur by comparing structural and functionalmeasures between forested and urban sites. We assessed 4 forested, 1 intermediate (deforested, not channelized),and 5 urban, channelized sites. Urbanization altered substrate (concrete at urban sites), riparian vegetation, light,temperature, O2, conductivity, nutrients, and major ion levels, and simplified food webs. Invertebrate composi-tion shifted from pollution-intolerant taxa at forested sites to tolerant taxa at urban sites. Richness in Surbersamples was 56 species at forested sites and 27 taxa at urban sites. Basal food sources at forested sites were leaflitter and biofilm, whereas at urban sites they were a sediment mat comprising organic matter, cyanobacteria,iron bacteria, and algae. Organic matter biomass on ceramic tiles was greater at urban than forested sites, butchlorophyll a was not. Faunal abundance was greater in litter bags at urban (mean = 72.6 animals/bag, 68%chironomids) than at forested sites (27.9 animals/bag; 33% chironomids), but species richness was higher inforested sites. Food webs at forested sites were more complex and differed from those at urban sites, whichlacked shredders, and grazers differed in taxonomic composition and diet between urban and forested sites. Atforested sites, insect larvae grazed biofilm, whereas at urban sites, snails consumed the organic mat. Leaf decom-position was fastest at the most disturbed site indicating rapid microbial decomposition caused by increasedtemperatures (+4.5°C) and nutrients, and possibly increased physical breakdown caused by rapid flow carryingabrasive sediment and garbage. Urbanization has had severe impacts on ecosystem function in Ampang Riverthat could be alleviated by preventing pollution, restoring riparian vegetation, and providing natural substrate.Key words: urban streamsyndrome, ecological integrity, macroinvertebrates, structural and functional indicators,water quality, pollution, algae

Humans are an increasingly urban species with >50%of the world’s population now living in urban areas com-pared with only 10% in 1900 (Meyer et al. 2005, Chin2006, Pickett et al. 2011). Over the next 50 y, the urbanpopulation is expected to increase to >70%, with most ofthis growth occurring in developing countries in tropicalAsia (UN 2012). Therefore, studies on urban stream ecol-ogy have become increasingly important (Paul and Meyer2001, Meyer et al. 2005, Walsh et al. 2005b, Pickett et al.2011).

The effect of urbanization on stream ecosystems is be-coming well understood in temperate regions where the‘urban stream syndrome’ has been extensively documented(e.g., Paul and Meyer 2001, Meyer et al. 2005, Walsh et al.2005b, Davies et al. 2010, Pickett et al. 2011, Wright et al.2011, Komínková 2012, Tippler et al. 2014). Urban water-sheds usually have high levels of impermeable surface cover,resulting in high surface runoff and low infiltration after

storms (Walsh et al. 2005a, Chin 2006). In response, urbanstreams experience flashy storm flows, reduced base flows,bank erosion, widened channels, and sedimentation. Urbanstreams typically are channelized, lack native riparian vege-tation, and are polluted with high concentrations of nutri-ents and contaminants. This harsh physical and chemicalenvironment tends to produce biotic assemblages of lowdiversity dominated by tolerant and invasive species (Walshet al. 2005b, Pickett et al. 2011). Ecosystem processes intemperate urban streams differ from in pristine streams,with faster rates of leaf decomposition, reduced nutrientuptake, and higher respiration rates (Roy et al. 2010).

Relatively little attention has been given to the ecologi-cal effects of urbanization on stream ecosystems in thetropics (Table 1; Ramírez et al. 2008, 2009) where higherrainfall and temperatures, frequent intense spates, rapidmicrobial activity, and different taxa of aquatic flora andfauna could potentially create different responses to ur-

E-mail addresses: [email protected]; [email protected]; [email protected]; [email protected]

DOI: 10.1086/681252. Received 20 May 2014; Accepted 23 January 2015; Published online 13 March 2015.Freshwater Science. 2015. 34(2):000–000. © 2015 by The Society for Freshwater Science. 000



banization than those observed in temperate streams. Fur-thermore, the problems of urbanization are exacerbatedin developing regions by poor infrastructure (e.g., lack ofsanitation and rubbish collection), poor public awareness,lack of funds for stream protection, and a common beliefthat environmental destruction is the inevitable price ofdevelopment.

In Malaysia, ∼95% of water for human use originatesfrom rivers (Othman et al. 2012), but most stream ecosys-tems in Malaysia have been severely compromised sincethe 1950s, mainly because deforestation and agriculturalconversion (Abdullah and Nakagoshi 2007) have causedsurface erosion and severe sedimentation (Azrina et al.2006, Abdullah and Nakagoshi 2007, Abdullah and Hezri2008). Coastal rivers are increasingly affected by urbaniza-tion, industrial activity, and associated channelization andpoint- and nonpoint-source pollution.

In Malaysia, as in much of southeastern Asia, all urbanstreams are polluted with rubbish because of the cultureof waste disposal. Inadequate garbage-collection services,lack of recycling, and the regular illegal dumping of do-mestic and industrial waste directly into rivers, along road-sides, and in forests are enormous problems. Most rubbishends up washed into rivers and out to sea, so beachesand mangrove swamps have become repositories for vastamounts of urban waste. Lack of government and publicconcern, inadequate, aging infrastructure (e.g., leaking sewerpipes), and untreated household discharges add to the chal-lenges.

A major effect of urbanization is removal of forest coverand riparian vegetation, which results in a decrease in

allochthonous inputs into streams. Urbanization affectshow leaves are broken down by microbes and fauna intemperate streams (Castela et al. 2008, Young et al. 2008),but less is known about leaf decomposition processes intropical streams (Yule et al. 2009, Boyero et al. 2011), par-ticularly regarding the effects of urbanization. Rates of leafdecomposition often have been reported to be faster intropical than in temperate streams, largely because of in-creased microbial activity consequent to higher tempera-tures and lack of seasonality. Thus, microbial breakdownof leaf litter may be more important in tropical streamsthan in their temperate counterparts (Wantzen et al. 2008,Boyero et al. 2011), and this difference may extend intourbanized streams (Ribas et al. 2006). Loss of submergedleaf litter can be accelerated by physical processes, such asmore frequent fast flow rates in urbanized streams, andthis effect could be greater in the tropics where rainfall ishigher and spates are frequent.

Relatively few studies have been done in tropical urbanstreams, and most of them were focused on water qualityand stream hydrology (Table 1). As in temperate areas,tropical urban streams generally exhibit highly depleted andsimplified macroinvertebrate communities (Ndaruga et al.2004, Azrina et al. 2006, Ramírez et al. 2009, Al-Shami et al.2011), further exacerbated in developing regions by thedischarge of untreated industrial wastewater and sewage.However, an understanding of the effect of this simplifi-cation of invertebrate communities in tropical streams islargely lacking, and no studies of the effects of urbanizationon stream ecosystem function in Malaysia have been pub-lished.

Table 1. Studies of the impact of urbanization on tropical rivers.

Location Topic Authors

Malaysia Hydrology and sedimentation Gupta (1984)

Malaysia Hydrology and sedimentation Balamurugan (1991)

Malaysia Water quality Chop et al. (2002)

Malaysia Macroinvertebrate distribution, biodiversity and water quality Azrina et al. (2006)

Malaysia Macroinvertebrate distribution and biodiversity Al-Shami et al. (2011)

Malaysia Water quality Othman et al. (2012)

Ghana Pollution and water quality Keraita et al. (2003)

Kenya Macroinvertebrate distribution and biodiversity Ndaruga et al. (2004)

Kenya Diatoms Ndiritu et al. (2006)

Nigeria River hydrology Ebisemiju (1989)

Zimbabwe River hydrology Whitlow and Gregory (1989)

Brazil Water quality Cunha et al. (2011)

Puerto Rico Hydrology, water quality, macroinvertebrates, fish Ramírez et al. (2009)

Venezuela Water quality Castillo (2010)

Southeastern Asia Review of ecology (tropical rivers in general) Dudgeon (2000)

Southeastern Asia Review of urban water problems Low and Balamurugan (1991)

Tropics in general Conservation Ramírez et al. (2008)

000 | Urbanization affects a tropical river C. M. Yule et al.



Studies of urban stream ecology are vital for a betterunderstanding of the various urbanization processes thatalter stream ecosystem function and to identify potentialconservation, remediation, and recovery strategies. There-fore, we chose to investigate the effect of urbanization onthe ecology of Ampang River, on the eastern edge of KualaLumpur, the capital of Malaysia.

We predicted that, as in temperate streams, urbaniza-tion would cause: 1) a decrease in pollution-sensitivetaxa, particularly Ephemeroptera, Trichoptera, and Plecop-tera (EPT), 2) an increase in abundance of tolerant taxa(e.g., snails, chironomids, and algae), and 3) a decrease inshredder-mediated leaf-litter decomposition, but an in-crease in microbe-mediated litter decomposition. We testedthese hypotheses by comparing the structure of macroin-vertebrate communities at 4 pristine and 6 urban sitesbased on Surber sampling, colonization of leaf litter bags(species composition, richness, and food webs), coloniza-tion of ceramic tiles by algae (chlorophyll a and biomass),and ecosystem function (rates of leaf-litter decompositionin litter bags). We predicted that 4) algae would increasein urban sites because of lack of riparian shading andnutrient inputs and that 5) food webs would be simplifiedbecause of a decrease in macroinvertebrate diversity inurban sites.

METHODSStudy sites

We studied Ampang River, a tributary of the KlangRiver, the major river flowing through Kuala Lumpur.Ampang River is a small headwater stream that rises invirtually pristine, primary lowland dipterocarp forest inthe Ampang Forest Reserve only 9 km from the city cen-ter. The forest ends abruptly in a housing estate, fromwhere the river flows through a concrete channel to meet

the Klang River. It becomes increasingly polluted by do-mestic and industrial waste, road runoff, and garbage as itflows downstream, a situation further exacerbated in theKlang River (Chop et al. 2002). The climate is tropical.Daily temperatures range from 23 to 33°C and mean an-nual rainfall is 3077 mm with no strong seasonal variation.Heavy rainfall causing spates occurs regularly at any timeof year and results in rapid rises and falls in stream veloc-ity and discharge.

We chose 10 sites along a gradient of increased urbani-zation (increased housing density, catchment impervious-ness, road runoff, garbage disposal, etc.; Fig. 1, Table 2). Westudied the macroinvertebrate community at all 10 sites,and we studied litter decomposition and algal colonizationof tiles at sites 1, 3, 5, 7, and 9. Sites 1 to 4 are relativelypristine sites in the forest reserve. They had dense canopycover and highly diverse riparian vegetation (>50 tree spe-cies along an ∼50-m stretch of forest at each site). Thestream consisted of pools, riffles, and waterfalls with a sub-strate of granite bedrock, boulders, cobbles (with some cal-cite), and coarse sand. After exiting the forest, the streamenters a housing estate where it flows for ∼100 m overnatural substrate (site 5) before passing through a culvertunder a road and into a broad concrete drain with depositsof sand. Sites 6 to 10 were in the concrete drain (partiallycovered with shifting deposits of sand eroded from the for-est catchment) along the urbanized part of the stream bor-dered by houses and a golf course.

Only native flora occurred in the forest. Moss, biofilm,and the red alga Balliopsis sp. (Rhodophyta, Batrachos-permales, Florideophyceae) were observed in the stream atsites 1 to 4; moss, biofilm, and the green alga Spirogyra(Charophyta, Zygnematales, Zygnemataceae) were observedat site 5; and a sediment mat comprising a variety of unicel-lular and filamentous Chlorophyta, cyanobacteria, and dia-toms occurred at sites 6 to 10. Sewage fungus was observed

Figure 1. Study area and locations of study sites on the Ampang River on the northeastern edge of Kuala Lumpur.

Volume 34 June 2015 | 000



Table

2.Locationandmean(±

SE)site

characteristicsin

theAmpang

River,K

uala

Lum

pur.

Site

Forest

Interm

ediate

Urban

12

34

56

78

910

Latitud

e(N

)03°10′09.8″

03°10′09.2″

03°10′03.4″

03°10′02.2″

03°10′00.1″

03°09′59.2″

03°09′59.1″

03°09′57.7″

03°09′55.8″

03°09′53.7″

Lon

gitude

(E)

101°46

′41.7″

101°46′39.0″

101°46′37.5″

101°46′35.7″

101°46

′30.4″

101°46′23.5″

101°46′21.0″

101°46

′01.6″

101°

5′58.8″

101°45′55.4″

Elevation

(m)

210

208

197

176

146

9797

9490

86

Airtemperature

(°C)

25.0±0.70

27.2±0.93

25.7±0.67

25.3±0.71

27.0±1.16

28.2±0.88

28.1±0.47

27.4±0.57

28.5±0.45

29.1±0.43

Water

temperature

(°C)

23.3±0.94

23.4±0.88

24.7±0.68

23.5±0.80

24.0±0.63

26.5±0.72

27.0±1.05

26.7±0.96

27.8±0.91

27.2±0.98

Meanchannelw

idth

(cm)

224±28.39

214±27.86

303±56.03

183±14.11

341±31.87

456±15.03

332±5.15

133±3.39

133±7.35

396±25.42

Meanwater

depth(cm)

8.0±1.2

13±1.1

12.4±1.4

19.2±1.9

16.1±2.9

6.4±0.7

5.4±0.5

9.2±0.4

7.4±0.5

4.8±0.4

Meanvelocity

(m/s)

0.3±0.1

0.1±0.0

0.3±0.1

0.3±0.0

0.2±0.0

0.4±0.1

0.4±0.0

0.4±0.0

0.6±0.2

0.4±0.0

pH6.91

±0.16

6.15

±0.40

6.61

±0.18

6.97

±0.18

7.30

±0.32

6.97

±0.30

6.80

±0.30

6.72

±0.04

6.81

±0.04

6.62

±0.02

Con

ductivity(μS/cm

)27.28±2.13

27.40±1.63

23.08±0.38

21.91±1.11

21.98±1.31

88.1±20.04

87.8±18.86

105.8±25.21

101.5±34.31

109.5±75.72

Dissolved

O2(m

g/L)

7.48

±0.57

6.94

±0.27

7.06

±0.24

7.35

±0.16

7.26

±0.19

6.15

±0.21

4.83

±0.27

4.96

±0.72

6.73

±0.88

5.93

±0.23

%bo

ulders

2060

550

200

00

00

%cobb

les

320

4520

105

55

510

%sand

5020

5030

7060

6060

5090

%concrete

00

00

035

3535

450

%cano

pycover

6080

7070

605

50

00

Light

(Lux

)6190

490

1467

1753

11,047

22,100

113,700

152,000

135,000

137,000

Ripariantree

species

richness

>50

>50

>50

>50

15–50

15–50

10–14

10–14

10–14

<10



in a drain entering the stream at site 7, where the streamhad a distinct odor and color. Iron bacteria were evident atsites 8 to 10. Large amounts of domestic waste, particularlyplastic bags and bottles, were dumped in the river or werewashed into it at sites 5 to 10. Riparian vegetation graduallydecreased downstream leaving the river increasingly ex-posed, and most of the riparian trees were introduced orsecondary-forest species. Stormwater and household dis-charges and runoff from the golf course entered the streamat sites 5 to 10.

We measured water and air temperature, conductivity(WP-81 meter; TPS Pty Ltd., Brisbane, Australia), pH, dis-solved O2 (CyberScan PD650 meter; EUTECH, ThermoFisher Scientific Inc., Waltham, Massachusetts), water ve-locity (Switzerland Flowatch® meter; JDC Electronic SA,Yverdon-les-Bains, Switzerland), and light intensity (LX-103 light meter; Lutron, Taipei, Taiwan) at each site. Wetook spot measurements on each of 4 sampling occasionsduring the morning between 0900 and 1200 h and aver-aged the results. We measured channel width and watervelocity at 5 transects (2–3 m apart) along the stream ateach site. At each transect, we measured water depth at5 intervals across the stream and noted the presence orabsence of leaf litter to obtain % leaf-litter cover. We es-timated % canopy cover and substrate composition withreference to a grid depicting % cover. We assessed thenumber of different riparian species (<10, 10–14, 15–50,or >50) along a 50-m stretch of river and noted substratetypes and aquatic vegetation. We collected 4 sets of watersamples (March, May, July, and December 2014) fromeach site with 1-L polyethylene bottles that were pre-washed with 3% HCl to remove deposits and bonded ionsfrom the inner surface of the bottle. Samples were filteredthrough 0.45-μm cellulose acetate filter paper and storedat 4°C for analysis. Nutrients were measured using high-performance liquid chromatography (515 HPLC pumpand 432 conductivity detector [detection limit = 0.1 mg/L];Waters, Milford, Massachusetts). We measured concen-trations of major ions by atomic absorption spectroscopy(3100; Perkin–Elmer, Waltham, Massachusetts).

Macroinvertebrate communities and dietsWe collected 5 macroinvertebrate samples (Surber sam-

pler, 20 × 20 cm, 300-μm mesh) from each of the 10 sitesalong the stream in March 2013. We removed large debrisand stones from the samples, which we preserved in 70%ethanol for laboratory analyses. We identified macroinver-tebrates to the lowest taxonomic level possible (separatedto morphospecies) with the aid of keys published by Yuleand Yong (2004). We assessed the diets of up to 10 individ-uals (depending on availability) of all macroinvertebratespecies represented by specimens with at least partly filledguts. We dissected the foreguts and squashed the contents(for very small specimens, the entire animal was squashed)

on a microscope slide before mounting them in polyvinylalcohol lactophenol mountant. We examined the slides un-der a microscope at magnifications of 200 and 400×. Wecategorized gut contents as: 1) fine particulate organic mat-ter (FPOM, particles < 500 μm), 2) coarse particulate or-ganic matter (CPOM, particles 500 μm–1 mm), 3) fungalhyphae, 4) leaf-litter fragments, 5) algae (diatoms, greenalgae, blue-green algae), and 6) animal tissue.We estimatedthe percentage of material in each food category for eachgut (this assessment was subjective but consistent) and as-signed taxa to feeding groups, i.e., collectors, grazers, shred-ders, or predators (Merritt and Cummins 1996), dependingon their diet. Collector–gatherers and collector–filtererscould not be distinguished by their diet, but field and labo-ratory observations on their mode of feeding and, whennecessary, examination of mouthparts under the micro-scope, allowed their separation for the construction of foodwebs.

Litter decompositionWe carried out a leaf-litter decomposition experiment

over 60 d from mid-April 2013 to investigate the effect ofurbanization on leaf-litter breakdown at sites 1, 3, 5, 7,and 9 along Ampang River. We collected recently abscisedleaves of a common N-fixing legume, Koompassia malac-censis Magingay ex Benth. (subfamily Caesalpinioideae,family Fabaceae), from Rimba Ilmu Botanic Gardens inthe University of Malaya, Kuala Lumpur, and air-driedthem.We weighed leaves (6.00 ± 0.01g) and placed them incoarse- (5 mm) and fine- (<0.5 mm) mesh bags, whichwere ∼20 × 20 cm in size. We used a total of 128 bags(equal numbers of coarse and fine) for the experiment.Eight bags were used to estimate handling losses. We de-ployed the remaining 120 bags at the 5 study sites(12 coarse and 12 fine bags/site). We removed 4 coarseand 4 fine bags (8 total) from each site on days 10, 30, and60 with a dip net (250-μm mesh) placed immediatelydownstream to avoid loss of invertebrates. In the labora-tory, we rinsed leaves from coarse-mesh bags, and re-moved and identified aquatic invertebrates. We oven-driedthe leaves remaining in the litter bags for 48 h at 60°C andweighed them to the nearest 1 mg before combustingthem at 550°C for 4 h. We cooled the ash in a desiccatorand weighed it the nearest 1 mg to obtain the ash-freedry mass (AFDM). We calculated decomposition rates withan exponential decay model Mt = Mie

–kt, where Mt isthe mass remaining at time t, Mi is initial mass, –k is theexponential decay coefficient expressed in mass loss/d, andt is the elapsed time in days.

Algal colonizationWe deployed forty 20- × 20-cm white ceramic tiles at

sites 1, 3, 5, 7, and 9 along the stream to allow coloniza-tion by periphyton. At each site, we tied 2 sets of 4 tiles




together on the substrate beneath the surface of the wa-ter. After 30 d, we retrieved the tiles and placed each oneon ice in a labeled Ziploc® bag. We used 4 tiles from eachsite to measure chlorophyll a and the other tiles to mea-sure AFDM. We removed algae growing on the surfaceof each tile carefully with a brush and rinsed the tileswith a minimum amount of distilled water into a 15-mLFalcon tube. We filtered the samples through prewashedWhatman GF/C membrane filter papers (47 mm) with50 mL of deionized water and stored the filters in thedark at –80°C for later analysis of chlorophyll a andAFDM (described in Hill et al. 2011).

Statistical analysisWe used 1-way analysis of variance (ANOVA) to com-

pare invertebrate abundance, species richness, and EPTrichness between study sites and 2-way ANOVA to com-pare the number of invertebrates colonizing leaf bags, withsites and time (days) as fixed factors (α = 0.05). Whenevertests were significant between sites, we used Tukey’s posthoc multiple comparisons to show the differences betweenthe sites and time. We also used 1-way ANOVA andTukey’s post hoc tests to compare dissolved metals, dis-solved nutrients, N, and P data to test whether these vari-ables differed among sites.

We used nonmetric multidimensional scaling (NMDS)to assess the similarity of macroinvertebrate communitiessampled with Surber samples between sites. We 4√(x)-transformed species abundance data to include rare spe-cies. We used principal components analysis (PCA) to sum-marize all measured chemical variables among sites. Wecompared leaf-mass remaining over time among sites withanalysis of covariance (ANCOVA), with sites as fixed fac-tors, time (days) as covariates, and mass remaining as theresponse variable. Prior to the ANCOVA, we tested forsignificant mass loss over time using linearized exponen-tial decay models and simple regressions. Statistical analy-ses were performed using SPSS 13.0 (SPSS Inc., Chicago,Illinois).

We described and compared the food webs based on thenumber of elements (S; species + food categories), numberof trophic links (L) between elements, and linkage com-plexity (S/connectance, where connectance = 2L/S[S – 1]),and connectance is the proportion of all possible trophiclinks that are realized.

RESULTSImpact of urbanization on stream structureand water chemistry

Urbanization resulted in major changes in substrate andsinuosity (the stream was confined within a concrete chan-nel), riparian vegetation (secondary and introduced ripar-ian species), canopy cover (from 60–80% at sites 1–5 to 0%at sites 8–10), light (490 Lux at site 2 to 152,000 Lux at

site 8), air and water temperature (diurnal increase of wa-ter temperature of 4.5°C between sites 1 and 9), leaf litter(average 70% cover in sites 1–4, 60% at site 5 and 0–5% atsites 6–10), and O2 (7.5 mg/L at site 1, 4.8 mg/L at site 7)(Table 2).

Pollution from stormwater drainage, household dis-charges, and runoff from the golf course (nutrients, herbi-cides, pesticides) was reflected in high conductivity (300–400% increase at sites 6–10) and Ca, Mg, Na and CO3

–,which all increased significantly at urban sites, particularlyat sites 8 to 10 (ANOVA, F9,20 Ca = 5.6, Mg = 7.3, Na = 5.6,CO3

– = 6.0; all p < 0.05; Table 3). High levels of Ca andCO3

– probably are the result of dissolution from the con-crete drainage system (Davies et al. 2010). Sewage fungihave been evident in a drain entering the stream at site 7for several years, probably because of a leaking sewer pipe.Higher temperatures, a lack of waterfalls and riffles, andincreased microbial activity caused lower O2 levels despitethe abundant algae and cyanobacteria at urban sites. Lowlevels of Cu and Zn were detected at site 10, but urbaniza-tion did not appear to result in significant heavy metal pol-lution. However, Cu and Zn levels were elevated at site 1 inMarch (but not in subsequent samples) (Table 3). Levelsof both Cu and Zn were above those recommended forthe protection of aquatic ecosystems (0.005 mg/L for Cuand 0.0005 mg/L for Zn; Garman 1983). Presumably thiselevated concentration was a consequence of underlyinggeology, and such natural increases are not uncommon(Garman 1983). Mean values of N, Fe, and Cl– were muchhigher at the urbanized than pristine sites, but the levelswere variable mainly because of occasional very high levelsand the differences among sites were not significant. How-ever, the presence of Fe bacteria at sites 8 to 10 indicatesthat high Fe levels are typical of these sites. NO2

– was vari-able but was high at sites 5, 9, and 10. Site 1 had high levelsof P in December, possibly from runoff from the forest be-cause heavy rainfall events are common at this time of year,but P did not differ significantly among sites overall. PO4

3–,Mn, and Ni were not detected in any of the water samples.

PCA axes 1 and 2 explained 84.5% of the variation inwater chemistry among sites. Forested and urban sites wereclearly separated along PC1 (Fig. 2). Intermediate site 5 wasmost similar to the 4 forested sites. The urban sites clus-tered together, as did sites 2 to 5, whereas site 1 was clearlyan outlier based on PC2, reflecting the very high levels ofCu and Zn in March.

AlgaeAt urban sites, filamentous and unicellular cyanobac-

teria occurred in a thick mat of fine organic sediment in-cluding bacteria, fungi, abundant protozoa and rotifers, andunicellular green algae and diatoms. This mat led to signifi-cantly higher AFDM deposits on the tiles with intensifyingurbanization downstream (Table 4). Despite the obvious




Table

3.Mean(±

SE)annu

al(M

arch,M

ay,July,Decem

ber)

concentrations

(mg/L)of

metalsandmajor

ions

ateach

site

intheAmpang

River,K

uala

Lum

pur.ValuesforMn,

Ni,andPO

4were0at

alltimes

andsites.

Site

Ca

Mg

Cu

FeZn

Pb

KNa

CO

32−

Cl

NO

2−

NO

3−

SO42−

NP

10.08

±0.03

0.02

±0.00

0.25

±0.24

0.1±0.02

0.37

±0.35

0.03

±0.03

0.3±0.10

0.11

±0.01

0.08

±0.05

0.08

±0.02

00.46

±0.11

1.22

±0.58

0.35

±0.27

0.5±0.29

20.05

±0.03

0.02

±0.00

0.02

±0.02

0.08

±0.03

0.04

±0.04

0.02

±0.02

0.3±0.11

0.09

±0.01

0.06

±0.04

0.07

±0.01

00.54

±0.15

1.1±0.67

0.29

±0.29

0.16

±0.06

30.07

±0.05

0.02

±0.01

0.02

±0.02

0.09

±0.03

0.04

±0.04

0.02

±0.02

0.29

±0.11

0.09

±0.01

0.08

±0.03

0.04

±0.01

00.38

±0.11

1.22

±0.64

0.29

±0.29

0.37

±0.14

40.09

±0.06

0.02

±0.00

00.1±0.04

0.01

±0.01

0.02

±0.02

0.27

±0.10

0.08

±0.00

00.1±0.04

00.41

±0.08

0.31

±0.15

0.31

±0.31

0.13

±0.05

50.07

±0.04

0.03

±0.02

0.01

±0.00

0.31

±0.21

00.01

±0.01

0.27

±0.11

0.08

±0.01

0.25

±0.10

0.1±0.03

0.04

±0.02

0.16

±0.01

0.18

±0.07

0.28

±0.28

0.31

±0.15

60.37

±0.13

0.08

±0.02

00.35

±0.25

00.01

±0.01

0.29

±0.10

0.21

±0.05

2.57

±0.62

0.52

±0.32

00.55

±0.19

0.38

±0.18

0.93

±0.49

0.26

±0.12

70.85

±0.41

0.07

±0.01

00.36

±0.25

00.01

±0.01

0.35

±0.13

0.19

±0.04

2.9±0.67

0.72

±0.35

00.43

±0.11

0.43

±0.10

0.8±0.24

0.3±0.17

81.5±0.73

0.12

±0.02

00.41

±0.28

00.01

±0.01

0.35

±0.11

0.2±0.02

2.78

±0.73

0.43

±0.15

00.51

±0.07

0.4±0.09

0.91

±0.47

0.21

±0.08

92.18

±1.43

0.12

±0.02

00.41

±0.28

00.01

±0.01

0.4±0.14

0.18

±0.01

2.72

±0.32

0.25

±0.08

0.03

±0.02

0.29

±0.05

0.41

±0.07

0.81

±0.41

0.1±0.05

102.9±1.97

0.13

±0.02

0.02

±0.01

0.42

±0.28

0.03

±0.03

0.02

±0.02

0.39

±0.13

0.17

±0.01

2.32

±0.96

0.35

±0.10

0.05

±0.03

0.43

±0.03

0.63

±0.22

0.7±0.38

0.31

±0.17



algal growth at the urban sites and the minimal algae ob-served at forested sites (only a thin layer of biofilm on largeboulders and bedrock and occasionally the red alga Bal-liopsis), chlorophyll a did not differ between urban and for-ested sites. The highest levels of chlorophyll a were at site 5(0.41 ± 0.11 μg/cm2; Table 4) where Spirogyra was naturallyabundant, and light levels were high because the multilay-ered canopy had been removed. Deposits of organic matterin an algae–sediment mat did not occur at site 5, and AFDMat site 5 did not differ significantly from the forested sites.In the urban sites, tiles were abraded by sand and objects,such as rubbish and branches, which scraped some of thealgae and sediment from the surface of the tiles.

Macroinvertebrate faunaVariation among sampling sites A total of 2124 macro-invertebrates belonging to 67 different taxa were collectedin the Surber samples. Fifty-six taxa were recorded from theforested and intermediate sites 1 to 5, and 27 taxa wererecorded from the urban sites 6 to 10 (Fig. 3A). Taxon rich-ness differed significantly between forested and urban sites(ANOVA, F9,40 = 14.67, p < 0.001; Fig. 3A). Average speciesabundance also differed significantly (ANOVA, F9,40 =3.381, p < 0.005; Fig. 3B). EPT richness differed markedlybetween sites 1 to 5 and sites 6 to 10 (ANOVA, F9,40 = 27.5,p < 0.001; Fig. 3C). Site 5, with natural substrate similar toupstream conditions, had faunal composition and richnesssimilar to the forested sites (Fig. 3A–C). Taxon richness atsite 1 was lower than at the other forested sites, possiblybecause of the effects of natural seepage of heavy metals(Cu and Zn) into the stream at site 1.

Variation in taxonomic composition with disturbanceNMDS produced a 2-dimensional solution that clearlygrouped the urban and forested sites by taxonomic compo-sition (Fig. 4). Site 5 was grouped with the forested sites, buta variety of taxa tolerant to the changed conditions (lack offorest cover but natural substrate) also were present. Theurban sites showed increasing dissimilarity with increasingurbanization.

Urbanization caused elimination of EPT taxa and adecrease in Blattodea, Hemiptera, Odonata, Coleoptera,Tipulidae, and Decapoda. Tolerant taxa increased, particu-larly collector–gatherer chironomids Polypedilum, Micro-tendipes, and Dicrotendipes and the predatory tanypodThienemannimyia. Chironomid larvae made up 39% of thefauna at the urban sites, but only 20% of the fauna at sites 1to 5. Chironomids made up 68% of the fauna in the litterbags at urban sites, but only 34% of the fauna in the litterbags at forested sites. Eleven taxa were found only at urbansites: 6 Gastropoda species, the predatory leech Barbroniaweberi, and sanguivorous Placobdelloides stellapapillium,an enchytraeid oligochaete, and the chironomids Micro-tendipes sp. and Monopelopia sp. Introduced macroin-vertebrate species occurred only at the urban sites (Gas-tropoda:Physa sp.). The red-eared terrapin (Trachemysscripta elegans) introduced from USA has been observedin the stream, and some of the fish species that were abun-dant at sites 6 to 10 probably were not native.

Food webs The food webs for the forested sites (1–4) andthe urban sites (6–10) were quite distinct with respect to theabundance and diversity of functional feeding groups pre-sent and basal food resources, whereas the food web for theintermediate site 5 most closely resembled the forested foodwebs (food webs are provided in Fig. S1A–C). The urbansites lacked shredders. At forested sites, larvae of psephe-nids, mayflies, and glossosomatids grazed biofilm, whereasat urban sites, snails grazed the algae–sediment mat. For-ested sites had diverse predators including Odonata, Ple-coptera, Veliidae, and Gerridae. The tanypod Thieneman-nimyia was found down to site 6, whereas at urban sites 7

Figure 2. Results of the principal components analysis (PCA)of water-chemistry variables at the different sampling sites. Sites1 to 4 are forested, 5 is intermediate, and 6 to 10 are urban.

Table 4. Mean (± SE, n = 4) ash-free dry mass (AFDM) andchlorophyll a of biofilm at each site. Values with the sameletter are not significantly different (analysis of variance andTukey’s pairwise comparison, α = 0.05).

Site AFDM (mg/cm2) Chlorophyll a (μg/cm2)

1 0.00 ± 0.00a 0.04 ± 0.02a

3 0.32 ± 0.06a 0.13 ± 0.01ab

5 0.87 ± 0.34a 0.41 ± 0.11b

7 9.52 ± 1.99b 0.17 ± 0.10ab

9 16.84 ± 2.02c 0.17 ± 0.08ab




to 10, the only predators were leeches and 1 corduliid nymphcollected at site 7. Foodweb statistics (Fig. 5) showed theimpact of urbanization with a decrease in complexity re-flected by a decrease in number of elements, number oflinks, and linkage complexity.

Leaf-litter decompositionFunctional changes with disturbance All regressions formass loss against time were significant, and we compareddecomposition at 2 scales, between mesh sizes at each siteand among sites for each mesh size (Fig. 6A, B). Only leavesincubated in site 7 showed significant effects of mesh size ondecomposition (ANCOVA, p < 0.05), with faster decompo-sition rates in coarse- than in fine-mesh bags. Across thesites, coarse-mesh decomposition rates were faster at sites 7and 9 than at sites 1 to 5, whereas fine-mesh decompositionrates were higher at site 9 than at sites 1 to 7. Feeding markscould be seen on the K. malaccensis leaves in coarse-meshbags at forested sites, whereas leaves from urban sites hadno noticeable feeding marks and were covered with ironbacteria.

Invertebrates in litter bags A total of 2751 inverte-brates, representing 93 taxa from 54 families, were col-lected from the coarse-mesh litter bags. Chironomids werenumerically dominant at urban sites and were particularlyabundant at the most-disturbed site 9 (345 Polypedilum,151 Microtendipes, and 77 Dicrotendipes). Litter bags atsite 5 had significantly more invertebrates than bags at theother sites (ANOVA, F4,60 = 9.32, p < 0.0001; Table S1),whereas litter bags from urban sites had the lowest EPTrichness (ANOVA, F4,60 = 13.71, p < 0.0001). Thirty-sixtaxa recorded from the litter bags were not collected in theSurber samples. This number included 12 species at the

Figure 4. Nonmetric multidimensional scaling (NMDS) plotof taxonomic variation between the sites using 4√(species abun-dance) data. Sites 1 to 4 are forested, 5 is intermediate, and 6to 10 are urban (stress = 0.18).

Figure 3. Mean (±1 SD) taxon richness/sample (A), inverte-brate abundance (individuals/sample) (B), and Ephemeroptera,Plecoptera, Trichoptera (EPT) taxon richness at each site (C).Bars with the same letter are not significantly different (Tukey’spairwise comparison, p < 0.05).




urban sites (e.g., low numbers of Ephemeroptera, Coleop-tera, and Hydropsychidae). Litter bags were favorable habi-tats for chironomids in the urbanized concrete channelwhere they provided both food and shelter.

DISCUSSIONAmpang River exhibits all of the major ecological char-

acteristics of urban stream syndrome that have been de-scribed for temperate urban streams (e.g., Paul and Meyer2001, Meyer et al. 2005, Walsh et al. 2005b, Pickett et al.2011, Komínková 2012, Tippler et al. 2012) that arise fromchannelization, decreased riparian vegetation, altered hy-drology, increasingly impervious catchment, and elevatedlevels of nutrients and contaminants. Urbanization clearlyhad a significant impact on the ecological integrity of Am-pang River with respect to structural (species richness, elim-ination of sensitive taxa and introduction of tolerant taxa,algal colonization, sedimentation and creation of an algae–sediment mat, and food webs) and functional measures (leafdecomposition).

Average water velocity measured during the study wassimilar among sites, but channelization caused uniformflow confined within widely spaced concrete walls, andthe urban sites lacked the pools, riffles, and waterfalls ofthe upstream sites, which provide a range of habitats forstream invertebrates. The region experiences high rainfall,which causes frequent spates and a more flashy hydrology

at the urban than at the forested sites because the impervi-ous surfaces cause rapid runoff and substrate variability orriparian vegetation are not present to impede flow or pro-vide refuges for the fauna. At forested sites, the streamchannel often changes dramatically because spates causefallen trees, dislodged boulders, and landslides (CMY, per-sonal observation). Fallen trees supply large woody debris,

Figure 6. Mean (±1 SE) leaf decomposition rates in coarse-(A) and fine- (B) mesh bags across study sites. Values shownare the regression slopes for linearized regression model of de-composition rates (/d). Bars with the same letter are not signifi-cantly different (analysis of covariance, p > 0.05).

Figure 5. Comparison of foodweb characteristics betweenforested, intermediate, and urban sites. Number (No.) of ele-ments: S = number of species plus food categories, L = numberof trophic links between taxa, S/connectance = linkage com-plexity where connectance = 2L/S(S − 1). Connectance = pro-portion of all possible trophic links that are realized.




which contributes to substrate heterogeneity and increaseshabitat complexity (Cushing and Allan 2001). The health-ier conditions at intermediate site 5 than at sites 6 to 10(Table 3, Figs 2–4) clearly showed the importance of anatural substrate to stream ecology. A similar result wasseen at another urban stream in Kuala Lumpur. Thisstream flowed through a concrete drain for most of itslength, but showed marked improvement in species diver-sity in a small downstream section that had cobbles andgravel substrate (CMY, personal observation).

Diurnal water temperatures measured for the AmpangRiver were typical of streams in tropical regions (23.3–27.8°C; Cushing and Allan 2001). Water temperature isprimarily affected by air temperature, but the 4.5°C in-crease in water temperature from site 1 to site 9 also is aconsequence of household discharges, runoff warmed byroads and houses, and the reduction in canopy cover (from60–80% to 0%) (Table 2). Climate warming is likely to ex-acerbate problems caused by increased temperatures inurban streams (Cushing and Allan 2001).

Our hypotheses were all supported. As we predicted, ur-banization caused a significant decrease in species richnessby eliminating pollution-sensitive taxa. Some tropical in-vertebrates clearly differ from temperate species in termsof pollution tolerance. For example, Ephemeroptera andChironomidae in the region (Yule et al. 2010, Jinggut et al.2012) seem to be more pollution tolerant than their tem-perate counterparts, in part because they are adapted toelevated temperatures, and are small with shorter life cy-cles (Marchant 1982, Schowalter 2006).

AFDM was significantly higher at urban than at for-ested sites (Table 4) because of the growth of a thick algae–sediment mat on the concrete substrate. This mat com-prised cyanobacteria and iron bacteria, which trapped fineorganic and inorganic sediment and supported a commu-nity of microbes, protozoa, and rotifers, together with chi-ronomids, predatory and sanguivorous leeches, and worms.The mat was grazed by 6 species of snails and fish thatwere observed at sites 6 to 10. A similar algae–sedimentmat was observed on boulders and cobbles in a forestedriver in Borneo that was polluted by sediment and Mnfrom mining activities (Yule et al. 2010). Cyanobacteriacan be indicators of N pollution, and some species pro-duce neurotoxins and hepatotoxins that have negative ef-fects on stream organisms, and particularly on small mac-roinvertebrates (Jafari and Gunale 2006, Sen et al. 2013).In temperate streams, urbanization can have varied effectson algae. For example, Munn et al. (2002) found that algalcommunities in forested streams in Washington (USA)were dominated by cyanobacteria, whereas urban streamswere diatom-dominated. In contrast, Taylor et al. (2004)reported that forested streams in Victoria (Australia) werediatom-dominated, whereas urban streams were character-ized by filamentous algae.

Urbanization decreased the complexity of aquatic foodwebs (Fig. 5). Food webs at forested and urban sites differedwith respect to the abundance and diversity of functionalfeeding groups present and basal food resources, whereasthe food web for the intermediate site 5 most closely resem-bled forested food webs. The urban sites lacked shreddersbecause litter was not retained in the smooth concretechannel. The provision of litter bags showed that more spe-cies could tolerate the urban conditions were leaf packsavailable. Grazers showed a complete shift in taxa and diet.At forested sites, psephenids, glossosomatids, and mayflylarvae grazed biofilm, whereas at urban sites, snails grazedthe algae–sediment mat. Forested sites had diverse insectpredators, whereas at the most-disturbed urban sites (sites7–10), the only predators were leeches (and 1 corduliidnymph collected at site 7).

We predicted a decrease in shredder-mediated leaf-litter decomposition and an increase in microbe-mediatedlitter breakdown with urbanization because of higher tem-peratures and nutrients. This prediction was supported.Gessner and Chauvet (2002) proposed a functional frame-work in which the ratio of breakdown rates from impairedand reference sites (with kimpacted/kreference) is used as ameasure of stream integrity. If this framework is appliedin Ampang River (where sites 1 and 3 are reference and 7and 9 are impaired sites), decomposition rates in coarse-mesh bags are 2× higher at impaired than at referencesites. No shredders were found in the urban litter bags,so these results must have been caused by faster microbialand physical decomposition at sites 7 and 9. In contrast,breakdown rates were 50% lower at intermediate site 5than at reference sites. At Site 5, shredders were less di-verse and abundant than at the forested sites, which causedslower decomposition. In fine-mesh bags, decompositionrates were similar between sites 5 and 7 and sites 1 and 3,but rates were 250% higher at site 9 (indicating faster mi-crobial decomposition). Based on the framework proposedby Gessner and Chauvet (2002), shifts of 50 to 200% fromreference rates indicate a heavily compromised ecosystem,demonstrated in our study along a gradient of urbanization.Forested sites 1 and 3 were undisturbed (0.75–1.33), inter-mediate site 5 was altered (0.5–0.75, 1.33–2.0), and urbansites 7 and 9 were compromised (<0.5, >2.0).

The fastest litter decomposition for both fine- andcoarse-mesh bags occurred at site 9, the most severelyimpaired site (Fig. 6), where the water temperatures were4.5°C higher than at site 1. Authors of several studies intemperate urban streams have reported an increased rateof litter decomposition related to an increase in microbialactivity and in physical abrasion resulting from flash flood-ing (Meyer et al. 2005, Chadwick et al. 2006, Imberger et al.2008). However, Del Arco et al. (2012) reported that tem-perate urban streams in Portugal exhibited slower litter de-composition than forested streams. Jinggut et al. (2012)




reported a gradient of decomposition rates in Borneanstreams flowing through undisturbed, farmed, and loggedforests, and attributed slower decomposition rates to highrates of sedimentation in the logged stream and increasedmicrobe-mediated decomposition consequent to externalnutrient inputs from agricultural practices. Chadwick et al.(2006) reported that high levels of toxic compounds canimpede microbial activity and retard litter breakdown, butsuch conditions would be more likely to arise from indus-trial pollution or mining activities than the lower levels ofpollutants in household and stormwater discharge in Am-pang River.

Decomposition rates did not differ between coarse- andfine-mesh bags at forested sites, but feeding marks wereseen on the leaves from coarse-mesh bags. In contrast,leaves from bags at urban sites had no noticeable feedingmarks and were, instead, covered in iron bacteria and or-ganic matter. No shredders were collected in the Surbersamples, and no settled leaf litter was recorded on theconcrete channelized stream bottom of the urban sites be-cause of the smooth concrete surface, constant rapid flow,and frequent spates (typical of streams in the wet tropics),all of which precluded litter retention. Consequently, un-der natural conditions, most leaves falling into the streamwould be broken down by microbes and physical abrasionwhile being carried downstream in the water column, un-less they were trapped against large debris, such as fallentrees or discarded items, such as furniture. Three shred-ding species (a cockroach, an elmid larva, and 3 adultBerosus sp.) were collected from urban litter bags (butwere not found in the Surber samples) indicating theycould survive in the stream despite the pollution if theyhad appropriate habitat available. Trichoptera and variousmayfly species also were absent from Surber samples atthe urban sites, but appeared in low numbers in the litterbags, a result showing that they too were absent largelybecause of lack of leaf retention in the channelized streamrather than pollution. Presumably they (together with theshredding elmid found in a litter bag at site 7) colonizedthe litter bags via drift from forested sites. The shreddersBlattodea sp. and Berosus were adults with wings andcould have flown downstream.

In conclusion, urbanization has affected all aspects ofthe ecology of Ampang River. Tropical conditions (highrainfall and temperatures, frequent intense spates, andrapid microbial activity) combined with inadequate streamprotection (particularly illegal dumping of waste) to exac-erbate the urban stream syndrome. The impact of frequentspates is worthy of further investigation. Spates flush outpollutants and garbage, but they bring more in from thecatchment and they cause extensive marine pollution. Onemight ask whether the enhanced microbial activity causedby high tropical temperatures is beneficial via rapid break-down of organic pollutants, such as sewage. In Malaysiaand throughout the tropics, rapid urbanization, indus-

trialization, and poor infrastructure are stressing urbanstreams. Apart from preventing sources of pollution, reha-bilitation of Ampang River and other urban streams inMalaysia could be enhanced by replanting native riparianvegetation, re-establishing variable flow conditions by in-troducing riffles and pools, and reintroducing bed sub-strata (e.g., large woody debris, cobbles, boulders), al-though this step would be difficult because of the frequentspates. The most important rehabilitation measure neededin Malaysia and most of southeastern Asia is that the cul-ture of waste disposal must be changed through educationand improved urban waste management, particularly be-cause the final destination of most of the waste is the ma-rine environment with disastrous consequences.

ACKNOWLEDGEMENTSThis work was part of the research for BSc Honours by

KVL and JYG funded by Monash University Malaysia Campus.Ng Ting Hui from the National University of Singapore identi-fied the gastropods. Chou Lee Yiung provided assistance withstatistical analyses. We thank Manuel Graça and 2 anonymousreferees whose comments greatly improved an earlier versionof this manuscript, and Editor Pamela Silver for her carefulediting.

LITERATURE CITEDAbdullah, S. A., and A. A. Hezri. 2008. From forest landscape

to agricultural landscape in the developing tropical countryof Malaysia: pattern, process, and their significance on pol-icy. Environmental Management 42:907–917.

Abdullah, S. A., and N. Nakagoshi. 2007. Forest fragmentationand its correlation to human land use change in the state ofSelangor, Peninsular Malaysia. Forest Ecology and Manage-ment 241:39–48.

Al-Shami, S. A., C. S. M. Rawi, A. H. Ahmad, S. A. Hamid, andS. A. M. Nor. 2011. Influence of agricultural, industrial, andanthropogenic stresses on the distribution and diversity ofmacroinvertebrates in Juru River Basin, Penang, Malaysia.Ecotoxicology and Environmental Safety 74:1195–1202.

Azrina, M. Z., C. K. Yap, A. R. Ismail, A. Ismail, and S. G. Tan.2006. Anthropogenic impacts on the distribution and biodi-versity of benthic macroinvertebrates and water quality ofthe Langat River, Peninsular Malaysia. Ecotoxicology and En-vironmental Safety 64:337–347.

Balamurugan, G. 1991. Sediment balance and delivery in a hu-mid tropical urban river basin: the Kelang River, Malaysia.Catena 18:271–287.

Boyero, L., R. G. Pearson, D. Dudgeon, V. Ferreira, M. A. S. Graça,M. O. Gessner, A. J. Boulton, E. Chauvet, C. M. Yule, R. J.Albariño, A. Ramírez, J. E. Helson, M. Callisto, M. Arun-achalam, J. Chará, R. Figueroa, J. M. Mathooko, J. F. Gonçal-ves, M. S. Moretti, A. M. Chará-Serná, J. N. Davies, A. En-calada, S. Lamothe, L. M. Buria, J. Castela, A. Cornejo, A. O. Y.Li, C. M’Erimba, V. Diaz Villanueva, M. C. Zúñiga, C. M.Swan, and L. A. Barmuta. 2011. Global patterns of streamdetritivore distribution: implications for biodiversity loss inchanging climates. Global Ecology and Biogeography 21:134–141.




Castela, J., V. Ferreira, and M. A. S. Graça. 2008. Evaluationof stream ecological integrity using litter decomposition andbenthic invertebrates. Environmental Pollution 153:440–449.

Castillo, M. M. 2010. Land use and topography as predictorsof nutrient levels in a tropical catchment. Limnologica 40:322–329.

Chadwick, M. A., D. R. Dobberfuhl, A. C. Benke, A. D. Huryn,K. Suberkropp, and J. E. Thiele. 2006. Urbanization affectsstream ecosystem function by altering hydrology, chemistryand biotic richness. Ecological Applications 16:1796–1807.

Chin, A. 2006. Urban transformation of river landscapes in aglobal context. Geomorphology 79:460–487.

Chop, A., J. Jusoh, and N. W. Chan. 2002. The Klang Rivercleanup programme. Pages 378–379 in N. W. Chan (editor).Rivers: towards sustainable development. Penerbit Universiti,Sains, Malaysia.

Cunha, D. G. F., W. K. Dodds, and M. D. C. Calijuri. 2011.Defining nutrient and biochemical oxygen demand baselinesfor tropical rivers and streams in Sao Paulo State (Brazil): acomparison between reference and impacted sites. Environ-mental Management 48:945–956.

Cushing, C. E., and J. D. Allan. 2001. Streams: their ecologyand life. Academic Press, London, UK.

Davies, P. J., I. A. Wright, O. J. Jonasson, and S. J. Findlay.2010. Impact of concrete and PVC pipes on urban waterchemistry. Urban Water Journal 7:233–241.

Del Arco, A. I., V. Ferreira, and M. A. S. Graça. 2012. Theperformance of biological indicators in assessing the ecolog-ical state of streams with varying catchment urbanizationlevels in Coimbra, Portugal. Limnética 31:141–154.

Dudgeon, D. 2000. The ecology of tropical Asian rivers andstreams in relation to biodiversity conservation. Annual Re-view of Ecology and Systematics 31:239–263.

Ebisemiju, F. S. 1989. The response of headwater stream chan-nels to urbanization in the humid tropics. Hydrological Pro-cesses 3:237–253.

Garman, E. E. J. 1983. Part I: water quality issues in Australia,Water 2000. In Department of Resources and Energy (edi-tor). Consultants Report No. 7. Water quality issues. Aus-tralian Government Publishing Service, Canberra, Australia.

Gessner, M. O., and E. Chauvet. 2002. A case for using litterbreakdown to assess functional stream integrity. EcologicalApplications 12:498–510.

Gupta, A. 1984. Urban hydrology and sedimentation in the hu-mid tropics. Pages 240–267 in J. E. Costa and P. J. Fleisher(editors). Developments and applications of geomorphology.Springer–Verlag, Berlin, Germany.

Hill, W. R., B. J. Roberts, S. N. Francoeur, and S. E. Fanta. 2011.Resource synergy in stream periphyton communities. Jour-nal of Ecology 99:454–463.

Imberger, S. J., C. J. Walsh, and M. R. Grace. 2008. More mi-crobial activity, not abrasive flow or shredder abundance,accelerates breakdown of labile leaf litter in urban streams.Journal of the North American Benthological Society 27:549–561.

Jafari, N. G., and V. R. Gunale. 2006. Hydrobiological study ofalgae of an urban freshwater river. Journal of Applied Sci-ences and Environmental Management 10:153–158.

Jinggut, T., C. M. Yule, and L. Boyero. 2012. Stream ecosystemintegrity is impaired by logging and shifting agriculture in a

global megadiversity center (Sarawak, Borneo). Science ofthe Total Environment 437:83–90.

Keraita, B., P. Drechsel, and P. Amoah. 2003. Influence of urbanwastewater on stream water quality and agriculture in andaround Kumasi, Ghana. Environment and Urbanization 15:171–178.

Komínková, D. 2012. The urban stream syndrome—a mini-review. Open Environmental and Biological Monitoring Jour-nal 5:24–29.

Low, K. S., and G. Balamurugan. 1991. Urbanization and urban wa-ter problems in southeast Asia: a case of unsustainable devel-opment. Journal of Environmental Management 32:195–209.

Marchant, R. 1982. Life spans of two species of tropical mayflynymph (Ephemeroptera) from Magela Creek, Northern Ter-ritory. Australian Journal of Marine and Freshwater Research33:173–179.

Merritt, R. W., and K. W. Cummins (editors). 1996. An intro-duction to the aquatic insects of North America. 3rd edition.Kendall/Hunt Publishing Company, Dubuque, Iowa.

Meyer, J. L., M. J. Paul, and W. K. Taulbee. 2005. Stream eco-system function in urbanizing landscapes. Journal of theNorth American Benthological Society 24:602–612.

Munn, M. D., R. W. Black, and S. J. Gruber. 2002. Response ofbenthic algae to environmental gradients in an agriculturallydominated landscape. Journal of the North American Ben-thological Society 21:221–237.

Ndaruga, A. M., G. G. Ndiritu, N. N. Gichuki, and W. N.Wamicha. 2004. Impact of water quality on macroinver-tebrate assemblages along a tropical stream in Kenya. AfricanJournal of Ecology 42:208–216.

Ndiritu, G., N. Gichuki, and L. Triest. 2006. Distribution ofepilithic diatoms in response to environmental conditions inan urban tropical stream, Central Kenya. Biodiversity & Con-servation 15:3267–3293.

Othman, F., M. E. Alaa Eldin, and M. Ibrahim. 2012. Trendanalysis of a tropical urban river water quality in Malaysia.Journal of Environmental Monitoring 14:3164–3173.

Paul, M. J., and J. L. Meyer. 2001. Streams in the urban land-scape. Annual Review of Ecology, Evolution, and Systemat-ics 32:333–365.

Pickett, S. T. A., M. L. Cadenasso, J. M. Grove, C. G. Boone,P. M. Groffman, E. Irwin, S. S. Kaushal, V. Marshall, B. P.McGrath, C. H. Nilon, R. V. Pouyat, K. Szlavecz, A. Troy,and P. Warren. 2011. Urban ecological systems: scientificfoundations and a decade of progress. Journal of Environ-mental Management 92:331–362.

Ramírez, A., R. de Jesús-Crespo, D. M. Martinó-Cardona,N. Martínez-Rivera, and S. Burgos-Caraballo. 2009. Urbanstreams in Puerto Rico: what can we learn from the tropics?Journal of the North American Benthological Society 28:1070–1079.

Ramírez, A., C. M. Pringle, and K. M. Wantzen. 2008. Tropicalriver conservation. Pages 285–304 in D. Dudgeon (editor).Tropical stream ecology. Elsevier Science, London, UK.

Ribas, A. C. D. A., M. O. Tanaka, and A. L. T. De Souza. 2006.Evaluation of macrofaunal effects on leaf litter breakdownrates in aquatic and terrestrial habitats. Australian Ecology31:783–790.

Roy, A. H., M. J. Paul, and S. J. Wenger. 2010. Urban streamecology. Pages 341–352 in J. Aitkenhead-Peterson and A.




Volder (editors). Urban ecosystem ecology. American Soci-ety of Agronomy, Crop Science Society of America, Soil Sci-ence Society of America, Madison, Wisconsin.

Schowalter, T. D. 2006. Insect ecology: an ecosystem approach.3rd edition. Elsevier/Academic, San Diego, California.

Sen, B., M. T. Alp, F. Sonmez, M. A. E. Kocer, and O. Canpolat.2013. Relationship of algae to water pollution and wastewater treatment. Pages 335–354 in W. El-Shorbagy andR. K. Chowdhury (editors). Water Treatment. InTech, Rijeka,Croatia. (Available from: http://www.intechopen.com/books/water-treatment/relationship-of-algae-to-water-pollution-and-waste-water-treatment)

Taylor, S. L., S. C. Roberts, C. J. Walsh, and B. E. Hatt. 2004.Catchment urbanisation and increased benthic algal biomassin streams: linking mechanisms to management. FreshwaterBiology 49:835–851.

Tippler, C., I. A. Wright, P. J. Davies, and A. Hanlon. 2014. Theinfluence of concrete on the geochemical qualities of urbanstreams. Marine and Freshwater Research 65:1009–1017.

Tippler, C., I. A. Wright, and A. Hanlon. 2012. Is catchment im-perviousness a keystone factor degrading urban waterways? Acase study from a partly urbanised catchment (Georges River,South-Eastern Australia). Water, Air, and Soil Pollution 223:5331–5344.

UN (United Nations). 2012. World urbanization prospects, the2011 revision: highlights. Page 33. United Nations PopulationDivision, Department of Economic and Social Affairs, NewYork. (Available from: http://esa.un.org/wpp/Documentation/pdf/WPP2012_HIGHLIGHTS.pdf)

Walsh, C. J., T. D. Fletcher, and A. R. Ladson. 2005a. Streamrestoration in urban catchments through redesigning storm-water systems: looking to the catchment to save the stream.

Journal of the North American Benthological Society 24:690–705.

Walsh, C. J., A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M.Groffman, and R. P. Morgan. 2005b. The urban stream syn-drome: current knowledge and the search for a cure. Journalof the North American Benthological Society 24:706–723.

Wantzen, K. M., C. M. Yule, J. M. Mathooko, and C. M. Pringle.2008. Organic matter processing in tropical streams. Pages43–64 in D. Dudgeon (editor). Tropical stream ecology. Aca-demic Press, London, UK.

Whitlow, J. R., and K. J. Gregory. 1989. Changes in urbanstream channels in Zimbabwe. Regulated Rivers: Researchand Management 4:27–42.

Wright, I. A., P. J. Davies, S. J. Findlay, and O. J. Jonasson.2011. A new type of water pollution: concrete drainage in-frastructure and geochemical contamination of urban wa-ters. Marine and Freshwater Research 62:1355–1361.

Young, R. G., C. D. Matthaei, and C. R. Townsend. 2008. Organicmatter breakdown and ecosystem metabolism: functional in-dicators for assessing river ecosystem health. Journal of theNorth American Benthological Society 27:605–625.

Yule, C. M., L. Boyero, and R. Marchant. 2010. Effects of sedi-ment pollution on food webs in a tropical river (Borneo, In-donesia). Marine and Freshwater Research 61:204–213.

Yule, C. M., M. Y. Leong, K. C. Liew, L. Ratnarajah, K. Schmidt,H. M. Wong, R. G. Pearson, and L. Boyero. 2009. Shredders inMalaysia: abundance and richness are higher in cool uplandtropical streams. Journal of the North American Bentho-logical Society 28:404–415.

Yule, C. M., and H. S. Yong. 2004. Freshwater invertebrates ofthe Malaysian region. Academy of Sciences Malaysia, KualaLumpur, Malaysia.




Urbanization affects food webs and leaf-litter decomposition in a tropical stream in Malaysia

Documents