Intensification of aquaculture may result in more fish culture waste being discharged into adjacent rivers and streams. Due to composition of such wastes, ecological conditions in waterbodies may be adversely affected. We determined the ecological consequences of freshwater land-based Tilapia farms on headwater streams using macroinvertebrate community attributes and functional feeding response in an upstream tributaries of a highland stream in Kenya. Nine aquaculture sites adjacent to tributaries of three headwater streams with different fish production volumes were sampled and monitored for macroinvertebrate abundance, richness, composition of Ephemeroptera, Plecoptera and Trichoptera, Oligochaetes and Chironomids (percentage Oligochaetes and Chironomids), species diversity as well as the functional feeding group responses. The total abundance of benthic macroinvertebrate consistently increased near discharge points and immediately downstream of the effluent outlets near the aquaculture farms. We observed positive correlations between macroinvertebrate attributes (except Ephemeroptera, Plecoptera and Trichoptera) with fish production at aquaculture facilities adjacent to the tributaries of the headwater streams. The proportion of Oligochaetes and Chironomids (percentage Oligochaetes and Chironomids) increased while that of Ephemeroptera, Plecoptera and Trichoptera at discharge points and downstream of the farms decreased. Also, relative abundance of scrapers and shredders decreased significantly, while significant increase of abundance was observed for deposit feeders, filter feeders and parasites with low predator population at discharge and downstream points. These consistent patterns indicated changes in ecosystem integrity and functioning, due to aquaculture effluents with particulate organic matter from fish food-derived wastes becoming a central source of energy in river benthic food webs.

Introduction

Aquaculture, the farming of aquatic organisms including: fish, molluscs, crustaceans and aquatic plants, has increased by 7% per annum between 1970 and 2008 (FAO, 2012). A key challenge facing aquaculture today is sustaining a continued increase in production whilst minimizing impact of the aquaculture activities on the environment (Navarrete-Mier et al., 2010). In most countries, aquaculture is practiced in fish ponds or fish cages located adjacent to surface waterbodies (Naylor et al., 2000) where water is diverted from a stream into aquaculture facility, and subsequently effluents from the aquaculture facility discharged back to the rivers/streams, without proper water treatment (Camargo and Gonzalo, 2007; Tello et al., 2010).

Fish farm effluents contain four different types of pollutants: (1) suspended solids; (2) pathogenic bacteria, viruses and parasites; (3) drugs and disinfectants used for disease and parasite control; (4) residual food and fecal materials (McVicar, 1997; Pérez et al., 2003; Gravningen, 2007). These effluents may lead to substantial amounts of organic waste material being released to the waters close to a fish farm (Dempster et al., 2009), and may be available to the benthic organisms (Boaventura et al., 1997; MacMillan et al., 2003). Residual food and fecal materials generate physicochemical and biological changes downstream from fish farm outlets. Therefore, there is need to establish environmental impacts of fish farm discharge.

Biological monitoring has been recommended for assessment of long term pollution impacts in freshwater. The biological monitoring is usually based on the numerical values of several indices that integrate the ecological responses of aquatic communities (e.g. macrophytes, epilithic diatoms, benthic macro-invertebrates, fish) to pollutants (dos Santos Simões et al., 2008). Macro-invertebrate communities respond to a wide range of anthropogenic impacts and are considered good biological indicator of stream ecological condition (Camargo et al., 2011; Tsuzuki, 2015; Dedieu et al., 2015). These species vary in their sensitivity to organic pollution and, thus, their relative abundances can be used to make inferences about pollution loads. Discharge of aquaculture effluents into stream represents a source of pollution load into the waterbody that may induce ecological impacts. The aim of this study was to assess influence of freshwater Nile Tilapia (Oreochromis niloticus) aquaculture effluents on benthic macro-invertebrates populations and functional feeding responses in a tropical headwater stream.

Materials and methods

Study locations and sampling points

The study was conducted in tributaries of a lower order highland stream in Gatundu in Central Kenya (Figure 1) between January and September 2014. It lies at an average altitude of 1759 m above the sea level. Temperatures range from 12.8°C to 25.4°C with an average of 18.7°C. The mean rainfall ranges between 1000 to 1300 mm annually. Human activities in the catchment include agriculture (crop production and ranching) in areas of rich volcanic soils. Nine flow-through fish farms producing Nile Tilapia were studied. They were coded from S1 to S9 to preserve the confidentiality of farmers. The study locations were based along the profiles of three rivers differing in number (and production) of fish ponds. Their annual production ranged from 29 to 244 mt year−1 (Table 1). Streams were riffle-run dominated with coarse bottom substrate (mostly pebble and stones); their mean annual discharge varied from 0.88 to 5.36 m3 s−1 (Table 1). Sampling points were upstream of effluent discharge points (UP), the effluent discharge points (ED) and 1000 m downstream of the effluent discharge points (DP).

Figure 1.

Location of nine Nile Tilapia farms studied at the tributaries of the Kenyan Highland streams.

Figure 1.

Location of nine Nile Tilapia farms studied at the tributaries of the Kenyan Highland streams.

Table 1.

Production of O. niloticus at nine locations studied, and some features of hydrological contexts of rivers at the vicinity of the farms.

Farm sites Total pond areas (ha) Annual production (mt year−1Mean width (m) Mean depth (m) Mean annual river discharge (m3 s−1
6.7 209 1.4 0.35 0.88 
2.7 42 3.1 0.42 2.33 
6.9 54 7.0 0.14 1.75 
6.4 204 8.1 0.11 1.59 
7.7 121 6.1 0.12 1.31 
6.6 130 5.1 0.24 2.19 
8.4 25 4.2 0.44 3.31 
6.8 13 3.7 0.81 5.36 
8.1 57 3.1 0.39 2.16 
Farm sites Total pond areas (ha) Annual production (mt year−1Mean width (m) Mean depth (m) Mean annual river discharge (m3 s−1
6.7 209 1.4 0.35 0.88 
2.7 42 3.1 0.42 2.33 
6.9 54 7.0 0.14 1.75 
6.4 204 8.1 0.11 1.59 
7.7 121 6.1 0.12 1.31 
6.6 130 5.1 0.24 2.19 
8.4 25 4.2 0.44 3.31 
6.8 13 3.7 0.81 5.36 
8.1 57 3.1 0.39 2.16 

Benthic macroinvertebrates sampling

A total of 5 benthic samples were collected at random locations in each site with a Surber sampler (0.09 m2, 250 μm mesh size). Approximately 0.3 × 0.5 m area was disturbed vigorously for 20 s, so as to avoid escape of large macro-invertebrates. All contents of the net were emptied into polythene bags. Macro-invertebrates collected were preserved using 70% ethanol, and transported to the laboratory for further processing.

Macroinvertebrates processing

In the laboratory, samples were washed through a 250 μm mesh size sieve to remove mud and sand and macro-invertebrates sorted, identified to the lowest-possible taxonomic level with an aid of several keys (Day et al., 2002; de Moor et al., 2003a,b; Stals and de Moor, 2007; Merritt et al., 2008) and counted in a stereomicroscope. The (1) abundance of benthic macro-invertebrates; (2) taxonomic richness; (3) Shannon index (H’); (4) percentage of Oligochaetes and Chironomids (%OC); (5) percentage of Ephemeroptera, Plecoptera and Trichoptera (%EPT); (6) relative abundances of the different functional feeding groups (%FFG); (7) Indice biologique global normalize (IBGN) value were computed at each location and site. The IBGN biotic index of water quality was developed for French streams (see Association Française de Normalisation [AFNOR], 2004 for details) but can also be used for characterizing tropical rivers. The IBGN faunal list is restricted to 138 taxa of which 38, in nine indicator groups, are indicators of water quality. Group 9, including Plecoptera, is the most sensitive to pollution; group 1, including worms which can live in highly polluted, turbid waters with little oxygen, is the most tolerant to pollution. The invertebrate samples reflect biological water quality on a scale of 0−20, where 0 indicates pollution and 20 no pollution. The IBGN score for a site depends on both the number of taxa and indicator groups recorded from the samples. Functional traits affinities dataset proposed by Chevenet et al. (1994) was used to classify individuals into the following functional feeding groups (FFGs): scrappers, shredders, deposit feeder, filter feeders, predator and parasite.

Statistical analyses

Differences between UP and ED values (UP−ED) and between UP and DP (UP−DP) were calculated for each variable of macro-invertebrate community to determine differences relative to the reference point. At each site, Kruskall−Wallis test was applied for abundances and richness of benthic macro-invertebrates. Interactions between sampling location and effluent discharge point on biological attributes were analyzed using Friedman Two-Way ANOVA with sampling location (1−9 ponds) and sampling points (UP, ED and DP) as factors. Linear regression was used to determine the relationships between the macro-invertebrate attributes and fish farm biomass. Principal component analysis (PCA) was used to reveal groups/clusters within our data set, in order to determine if there are any influences of the sampling points on the macroinvertebrate attributes. All statistical analyses were performed with STATISTICA 10.0 Statistical Package.

Results

The benthic macroinvertebrate community values at the vicinity of the nine fish farms are provided in Figure 2. All benthic macro-invertebrate community attributes were affected by the interactions between sampling points and sampling points (p < 0.05). Highest abundance of benthic macro-invertebrates as well as %OC was reported in ED compared to reference points and DP. Species richness, diversity, and IBGN were higher at reference sampling points compared to impacted sampling points (ED and DP). The %EPT was highest at reference sampling points and appeared positively influenced by fish biomass (S1, S4, S5, and S6).

Figure 2.

Values of benthic invertebrate community values at the nine fish farms UP: immediately upstream, ED: discharge points, and DP: 1000 m downstream of the effluent outlet. Abundance and Richness are, respectively, total abundance of benthic invertebrates and taxonomic richness in samples collected in each sampling site. H’ = Shannon diversity index; %OC is the relative abundance of Oligochaetes and Chironomids; %EPT is the relative abundance of Ephemeroptera, Plecoptera and Trichoptera.

Figure 2.

Values of benthic invertebrate community values at the nine fish farms UP: immediately upstream, ED: discharge points, and DP: 1000 m downstream of the effluent outlet. Abundance and Richness are, respectively, total abundance of benthic invertebrates and taxonomic richness in samples collected in each sampling site. H’ = Shannon diversity index; %OC is the relative abundance of Oligochaetes and Chironomids; %EPT is the relative abundance of Ephemeroptera, Plecoptera and Trichoptera.

Changes of benthic macro-invertebrate community values at nine locations are shown in Figure 3. Higher macro-invertebrate abundance occurred at all impacted sampling locations. Species richness, diversity and IBGN values of macro-invertebrates decreased at impacted sampling points regardless of the fish biomass. The %OC increased at all sampling locations with largest increase reported at locations with large fish biomass. Nevertheless, ED and DP of location S2 and S7 recorded an overall decline in %OC. Largest changes in %EPT occurred at sampling locations with highest fish biomass (S1, S4, and S5).

Figure 3.

Changes of benthic invertebrate community values at the nine fish farms. Δ were calculated by comparing values immediately upstream (UP) to values at ED (shaded) and 1000 m DP (black).

Figure 3.

Changes of benthic invertebrate community values at the nine fish farms. Δ were calculated by comparing values immediately upstream (UP) to values at ED (shaded) and 1000 m DP (black).

Variations in macro-invertebrate community attributes at discharge points (ED) with respect to fish biomass are shown in Figure 4. Macro-invertebrate attributes (abundance, richness and %OC) showed significant (p < 0.05) positive relationships but %EPT showed significant (p < 0.05) negative correlation with fish biomass. Using benthic macro-invertebrate metric attributes (abundance, richness, %OC, %EPT and IBGN), three principle factors (Eigen values (>1)) were extracted to explain variability in the PCA and together, two main factors explained 72.89% of the total data variance (Figure 5). Principal Component Analysis (PCA) clustered most effluent discharge point and downstream points in one cluster which differed from those of upstream locations.

Figure 4.

Variations (Δ) in macroinvertebrate community attributes in the vicinity of the fish farms as a function of fish biomass. Δ were calculated by comparing samples collected upstream (UP) and with values at the discharge points (ED) and at the downstream of the fish farm outlet (DP).

Figure 4.

Variations (Δ) in macroinvertebrate community attributes in the vicinity of the fish farms as a function of fish biomass. Δ were calculated by comparing samples collected upstream (UP) and with values at the discharge points (ED) and at the downstream of the fish farm outlet (DP).

Figure 5.

Principal Component Analysis (PCA) showing clustering of sampling sites and locations based on benthic macro-invertebrate metric attributes (abundance, richness, %OC, %EPT and IBGN). The location were upstream (UP, circles), effluent discharge points (ED, rectangles) and 1000 m downstream of fish farm outlet (DP, triangles).

Figure 5.

Principal Component Analysis (PCA) showing clustering of sampling sites and locations based on benthic macro-invertebrate metric attributes (abundance, richness, %OC, %EPT and IBGN). The location were upstream (UP, circles), effluent discharge points (ED, rectangles) and 1000 m downstream of fish farm outlet (DP, triangles).

Scrappers, shredders, deposit feeders, and filter feeders dominated macroinveterbrate functional feeding groups (Figure 6). Deposit feeders, and filter feeders significantly (p < 0.05) increased in abundance while scrappers and shredders decreased at ED and DP at all sampling locations.

Figure 6.

Mean proportions (±SD) of each functional feeding group among invertebrate community samples collected immediately upstream (UP), at the discharge point (ED) and 1000 m downstream (DP) and 1000 m downstream (D2) of fish farm outlet in the studied rivers.

Figure 6.

Mean proportions (±SD) of each functional feeding group among invertebrate community samples collected immediately upstream (UP), at the discharge point (ED) and 1000 m downstream (DP) and 1000 m downstream (D2) of fish farm outlet in the studied rivers.

Discussion and conclusions

In this study, highest abundance of benthic macro-invertebrates and %OC was reported at effluent discharge point suggesting that fish farm effluents affect macroinvertebrates abundance as reported in other studies (Camargo, 1994; Boaventura et al., 1997; dos Santos Simões et al., 2008; Živić et al., 2009; Guilpart et al., 2012). The effect of aquaculture effluent is evidence since at the UP sampling points, species richness, diversity, %EPT and IBGN were highest compared to impacted sampling locations (ED and DP). The effluents are mainly fish feed which is transformed into fish biomass and partly released into the waterbody (Boyd, 2003; True et al., 2004) and drugs to control fish diseases that increase organic and inorganic matter to aquatic ecosystem (Stephen and Farris, 2004). Increase in abundance, richness, %EPT and %OC was obvious near locations with higher fish production and thus it appears that volume of production increased the organic matter loading in the stream that affected macroinverterbrate population attributes. The IBGN method relies on analysis of benthic invertebrate and it is expected to be sensitive to organic pollution and habitats homogenization (Guilpart et al., 2012). In our study, IBGN value was low in impacted sites. This method often indicates that more polluted areas have higher IBGN (AFNOR., 2004) therefore the method detected impacts of aquaculture effluents in our study.

Locations with high fish production showed lower abundance of macroinvertebrate communities downstream but at the locations with low production volumes, the abundance was somewhat similar between the effluent discharge point and downstream sampling points suggesting possibility of recovery of the macroinvertebrate community 200 to 1000 m downstream at low fish production level. Nevertheless, we also observed that even with reduced macro-invertebrate abundance downstream, a strong increase in Oligochaetes and Chironomids was observed simultaneously suggesting polluted environment due to the discharge of effluent from fish farms. The PCA indicated that sampling points upstream had somewhat similar characteristics while ED and DP aggregated together suggesting that sampling locations along the river did not have significant influence on the macroinvertebrate attributes. These results provide evidence that macroinvertebrate attributes were affected by sampling points at each location.

In terms of functional feeding diversity, scrapers and shredders populations reduced downstream of fish farm effluents. Scrapers and shredders feed mainly on epilithic and epiphytic biofilms and are an important component of algal-grazer pathway. Abundance of deposit feeders (Oligochaetes and Chironomids) and filter feeders (Hydropsychidae, Simuliidae and Sphaeridae) significantly increased in macroinvertebrate samples collected downstream suggesting that effluent from fish farm contained organic matter that support such feeding traits corresponding to generalist taxa (Barbour et al., 1999). Presence of these macroinveterbrates also suggests that effluent discharge points are polluted due to their persistence in polluted environments (dos Santos Simões et al., 2011). Parasite feeders were almost absent in macroinvertebrate samples collected upstream of fish farms. Parasite taxa could have been promoted because of their good tolerance to organic pollution and because of an increase in abundance of their hosts; their abundance could therefore be an indicator of strength of ecological disturbance linked to fish farms. Therefore, discharge of aquaculture effluents appears to affect functional feeding traits of macroinveterbrates.

While analysis of macroinvertebrate populations and functional feeding groups revealed effects of aquaculture effluents in the upland streams, ecological disturbances induced by freshwater land-based Tilapia farms appear to be emphasized when using taxonomy-based and functional feeding group indicators species in streams with high fish biomass. Therefore, a future concern involve aquaculture facilities with high production volumes that need to balance aquaculture farming and maintenance of downstream water quality since the stream community might not be able to recover from large aquaculture farm. Particularly, this study recommends further study to determine the relationship between size of the aquaculture ponds and ability for downstream recovery to protect the integrity of the streams from intensive aquaculture activities.

Funding

We would like to acknowledge the financial support granted by Aquafish Innovation Lab, formerly Aquafish Collaborative Research Support Program (ACRSP), partially funded by the United States Agency for International Development (USAID) under Grant No. LAG-G-00-96-90015-00.

References

AFNOR
,
2004
.
Determination de l'Indice Biologique Global Normalise (IBGN) (Global Biological Index Normalized)
.
Assoc. Francaise de Normalisation, Normé homologuee T
.
90
350
.
Barbour, M. T., Gerritsen, J., Snyder, B. D., Stribling, J. B.,
1999
.
Rapid Bioassessment Protocols for Use in Streams and Wadeable Rivers: Periphyton, Benthic Macroinvertebrates and Fish (No. EPA 841-B-99-002)
.
U.S. Environmental Protection Agency, Office of Water
,
Washington, DC
.
Boaventura, R., Pedro, A. M., Coimbra, J., Lencastre, E.,
1997
.
Trout farm effluents: characterization and impact on the receiving streams
.
Environmental Pollution
95
,
379
387
.
Boyd, C.,
2003
.
Guidelines for aquaculture effluent management at the farm-level
.
Aquaculture
226
,
101
112
.
Camargo, J. A.,
1994
.
The importance of biological monitoring for the ecological risk assessment of freshwater pollution: a case study
.
Environment International
20
,
229
238
.
Camargo, J., Gonzalo, C.,
2007
.
Physicochemical and biological changes downstream from a trout farm outlet: Comparing 1986 and 2006 sampling surveys
.
Limnetica
26
,
405
414
.
Camargo, J. A., Gonzalo, C., Alonso, A.,
2011
.
Assessing trout farm pollution by biological metrics and indices based on aquatic macrophytes and benthic macroinvertebrates: A case study
.
Ecologial Indicators
11
,
911
917
.
Chevenet, F., Dolédec, S., Chessel, D.,
1994
.
A fuzzy coding approach for the analysis of long-term ecological data
.
Freshwater Biology
31
,
295
309
.
Day, J. A., Harrison, A. D., de Moor, I. J.,
2002
.
Guides to the freshwater invertebrates of southern Africa. Volume 9: Diptera
.
WRC Report No. TT 201/02. Water Research Commission
,
Pretoria
,
South Africa
.
de Moor, I. J., Day, J. A., de Moor, F. C.,
2003a
.
Guides to the freshwater invertebrates of southern Africa. Volume 7: Insecta I: Ephemeroptera, Odonata and Plecoptera
.
WRC Report No. TT 207/03. Water Research Commission
,
Pretoria, South Africa
.
de Moor, I. J., Day, J. A., de Moor, F. C.,
2003b
.
Guides to the freshwater invertebrates of southern Africa. Volume 8: Insecta II: Hemiptera, Megaloptera, Neuroptera, Trichoptera and Lepidoptera
.
WRC Report No. TT 214/03. Water Research Commission
,
Pretoria, South Africa
.
Dedieu, N., Rhone, M., Vigouroux, R., Céréghino, R.,
2015
.
Assessing the impact of gold mining in headwater streams of Eastern Amazonia using Ephemeroptera assemblages and biological traits
.
Ecological Indicators
52
,
332
340
.
Dempster, T., Uglem, I., Sanchez-Jerez, P., Fernandez-Jover, D., Bayle-Sempere, J., Nilsen, R., Bjorn, P. A.,
2009
.
Coastal salmon farms attract large and persistent aggregations of wild fish: an ecosystem effect
.
Marine Ecology Progress Series
385
,
1
14
.
dos Santos Simões, F., Altair, B., Moreira, A. B., Bisinoti, M. C., Gimenez, S. M. N., Yabe, M. J. S.,
2008
.
Water quality index as a simple indicator of aquaculture effects on aquatic bodies
.
Ecological Indicators
8
,
476
484
.
FAO
,
2012
.
Global aquaculture production fishery statistical collections
.
Food and Agriculture Organization of the United Nations.
Available at: http://www.fao.org/fishery/statistics/global-aquaculture-production/en
Gravningen, K.,
2007
. Driving forces in aquaculture: different scenarios towards 2030. In: Arthur, R., Nierentz, J. (Eds.),
Global Trade Conference on Aquaculture. United Nations
,
Rome
, pp.
19e26
.
Guilpart, A., Roussel, J.-M., Aubin, J., Caquet, T., Marle, M., Le Bris, H.,
2012
.
The use of benthic invertebrate community and water quality analyses to assess ecological consequences of fish farm effluents in rivers
.
Ecological Indicators
23
,
356
365
.
MacMillan, J. R., Huddleston, T., Woolley, M., Forthergill, K.,
2003
.
Best management practice development to minimize environmental impact from large flow-through trout farms
.
Aquaculture
296
,
91
99
.
McVicar, A. H.,
1997
.
Disease and parasite implications of the coexistence of wild and cultured Atlantic salmon populations
.
ICES Journal of Marine Science
54
,
1093
1103
.
Merritt, R. W., Cummins, K. W., Berg, M. B. (Eds.),
2008
.
An introduction to the aquatic insects of North America
. 4th Edition.
Kendall/Hunt Publishing Company
,
Dubuque, Iowa
.
Navarrete-Mier, F., Sanz-Lázaro, C., Marín, A.,
2010
.
Does bivalve mollusc polyculture reduce marine fin fish farming environmental impact?
Aquaculture
306
,
101
107
.
Naylor, R. L., Goldburg, R. J., Primavera, J. H., Kautsky, N., Beveridge, M. C., Clay, J., Folke, C., Lubchenco, J., Mooney, H., Troell, M.,
2000
.
Effects of aquaculture on World fish supplies
.
Nature
405
,
1017
1024
.
Pérez, J. E., Alfonsi, C., Nirchio, M., Muñoz, C., Gómez, J. A.,
2003
.
The introduction of exotic species in aquaculture: a solution or part of the problem
.
Interciencia
28
,
234
238
.
Stals, R., de Moor, I. J.,
2007
.
Guides to the freshwater invertebrates of southern Africa, Volume 10: Coleoptera
.
WRC Report No. TT 320/07. Water Research Commission
,
Pretoria, South Africa
.
Stephen, W., Farris, J. L.,
2004
.
A biomonitoring approach to aquaculture effluent characterization in channel catfish fingerling production
.
Aquaculture
241
,
319
330
.
Tello, A., Corner, R. A., Telfer, T. C.,
2010
.
How do land-based salmonid farms affect stream ecology
.
Environmental Pollution
158
,
1147
1158
.
True, B., Johnson, W., Chen, S.,
2004
.
Reducing phosphorus discharge flow through aquaculture I: facility and effluent characterization
.
Aquaculture Engineering
32
,
129
144
.
Tsuzuki, Y.,
2015
.
Relationships between pollutant discharge and water quality in the rivers from “better” to “worse” water quality
.
Ecological Indicators
52
,
245
255
.
Živić, I., Marković, Z., Filipović-Rojka, Z., Živić, M.,
2009
.
Influence of a trout farm on water quality and Macrozoobenthos Communities of the receiving stream (Trešnjica River, Serbia)
.
International Review of Hydrobiology
94
,
673
687
.