Full Paper
7th Australian Stream Management Conference - Full Paper
Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney? 1
2
3
1
4
Tippler C. , Findlay S. , Wright I.A. , Davies P.J. , Evans C. and Ahmed M.
5
1 Department of Environment and Geography, Macquarie University, NSW 2. Ku-ring-gai Council, Pymble, NSW 3. School of Science and Health, University of Western Sydney, South Penrith , NSW 4. Georges River Combined Councils’ Committee, Hurstville, NSW 5. Blacktown City Council, Blacktown NSW
Key Points • Freshwater macroinvertebrate communities of the Sydney basin do not exhibit significant seasonal variation. • Sampling of macroinvertebrate communities is traditionally undertaken during spring and autumn, however within the Sydney basin this approach is not necessary. • Findings of this study will enable waterway managers to refine monitoring programs.
Abstract This paper challenges the validity of current Australian guidelines that recommend seasonal sampling for freshwater macroinvertebrates. Macroinvertebrates are a commonly used indicator to assess the ecological condition of streams and various sampling methods provide a consistent and repeatable approach. Most methods recommend seasonal surveys, which traditionally occur in spring and autumn. The basis for a seasonal approach is to capture specimens at an ideal life-stage when identification can be easily achieved. This is particularly relevant for many invertebrate taxa with lifecycles that are influenced by and dependent on extreme seasonal climatic change. However in more temperate environments, where daytime temperatures seldom drop below 10 degrees Celsius the biological response of macroinvertebrate communities and in-turn, seasonal differences may not be as pronounced. This study reviewed family level macroinvertebrate data sets across the Sydney basin to investigate if seasonality is a factor that influences community composition between spring and autumn sampling periods. The results suggest the time of year that sampling is undertaken is not an important consideration within the study region and that this should inform future sampling protocols.
Keywords macroinvertebrates, seasonality, Sydney, invertebrate indices, sampling, catchment management
Introduction Throughout the world freshwater macroinvertebrates are commonly used as an indicator of urban stream ecosystem condition (Beavan et al. 2001; Paul and Meyer 2001; Walsh et al. 2001). This has been based on the relative ease of collection, identification of specimens and the sensitivities to pollution and environmental disturbance that many invertebrate taxa display (Chessman 2003). The common approach to sampling macroinvertebrates in streams and rivers is by qualitative surveys using sweep/kick net sampling techniques designed to capture a sample of biota within different environments such as pools, riffles and stream edges. A range of methods have advocated sampling during autumn and spring (Wright 2000; Turak and Waddell 2002) as seasonal variability is well known to influence the distribution and abundance of various invertebrate taxa and the composition of communities (Brooks 2000; Leung et al 2012; Sporka et al 2006). For example in many parts of the northern hemisphere where cold winters are followed by short warm summers, significant seasonal variation in both invertebrate community structure and the abundance of a many invertebrate taxa has been documented (Sporka et al 2006). Similarly, invertebrate communities in tropical monsoonal streams have been shown to exhibit strong seasonal variation due to flow variation resulting from the cyclical wet and dry nature of these ecosystems (Leigh 2012; Mesa 2012). The climate in the Sydney basin is best characterised as temperate with little difference in rainfall between seasons and relatively low differences in temperatures across the seasons. Summers are usually warm with a mean maximum Tippler, C., Findlay, S., Wright, I.A., Davies, P. J., Evans, C. & Ahmed, M. (2014). Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney, in Vietz, G; Rutherfurd, I.D, and Hughes, R. (editors), Proceedings of the 7th Australian Stream Management Conference. Townsville, Queensland, Pages 292-299. 292
7ASM Full Paper Tippler C. et al - Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney? temperature of 25°Celcius and winters mild with mean minimum of 8°Celcius. Outside of drought years, rainfall is regular throughout the year with highest average rainfall falling between February-May and September –October (BOM 2014). Elevation across the Sydney basin ranges from sea level to a maximum of 400 m above sea level. Sydney is fortunate to have a number of stream catchments with high coverage of remnant vegetation and with little to no urban development. This provides a notable contrast to streams within urbanised areas that can are best described as degraded due to catchment modification which has led to higher diurnal temperature differences, flashier hydrology, and poor water quality among other factors (Meyer et al 2005; Tippler et al 2012). When using macroinvertebrates as an indicator of stream health in Australia the most commonly advocated approaches follow that prescribed by the AUSRIVAS (Turak and Waddel 2002) or SIGNAL (Chessman 2003) methods. Both advocate seasonal surveys (autumn and spring) on the basis that this will help capture the greatest diversity of invertebrate taxa across their various life stages and ensure seasonal variation within any specific communities can be accounted for, irrespective of climate. Macroinvertebrate sampling undertaken in Sydney commonly use the AUSRIVAS and SIGNAL methods and apply autumn and spring monitoring protocols. In the published results of the various sampling programs, the influence of seasonality on invertebrate populations is often overlooked in the analysis. Tippler et al (2012) found that seasonality in the Georges River (to the south of Sydney) was not significant at taxonomic level of Order. In a study of streams from the Ku-ring-gai Local Government Area (to Sydney’s north) Wright (2011) reported seasonal differences in invertebrate community structure as measured at the species level was minimal. Outside these two studies no other peer-reviewed commentary on the impact of seasons with the study region has been found. This study specifically examines the influence of seasonality on macroinvertebrate community structure and commonly used invertebrate indices of SIGNAL, EPT, richness and abundance across Sydney. The results of this study are proposed to inform the future design of catchment management monitoring programs within this region and comparable climates.
Method Study Area The Sydney basin, as defined in this study, encompasses an area which extends from Hornsby in the north to Penrith in the west, Appin to the south and Bondi to the east. The study area involved 12 local government areas (Fig. 1) including Hornsby and Ku-ring-gai Councils to the north, Blacktown City Council to the west, Sutherland, Rockdale, Kogarah, Hurstville, and Bankstown City Councils to the south and Fairfield, Liverpool, Campbelltown and Wollondilly Councils to the south west. The Sydney basin is dominated by a temperate climate with warm summers and no dry season (NSW Department of Environment and Heritage 2014). Average annual rainfall varies from approximately 1400 mm in areas closest to the coast and decreases to approximately 700 mm in the western suburbs. Rainfall across the Sydney basin occurs regularly with slightly elevated monthly falls recorded in late autumn and early to mid-winter. The average daytime temperatures range from approximately 16 °C (winter) to 26 °C (summer) in areas closest to the coast and approximately 16 °C (winter) to 30 °C (summer) in areas furthest from the coast (BOM 2014).
Tippler, C., Findlay, S., Wright, I.A., Davies, P. J., Evans, C. & Ahmed, M. (2014). Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney, in Vietz, G; Rutherfurd, I.D, and Hughes, R. (editors), Proceedings of the 7th Australian Stream Management Conference. Townsville, Queensland, Pages 292-299. 293
7ASM Full Paper Tippler C. et al - Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney?
Figure 1. Map of the Sydney basin and participating Councils (coloured).
Aquatic Macroinvertebrate Sampling Macroinvertebrate sampling data was obtained from the participating councils between 2002 to 2013.This included 412 aquatic macroinvertebrate sampling results collected from 56 locations across 50 freshwater streams and rivers. 220 samples were collected during the autumn sample season and 192 collected during the spring sampling season. 60 samples were collected from reference streams (defined as those with intact naturally vegetated catchments) and 352 samples from disturbed streams, the majority with urbanised catchments and a small number impacted by agriculture. Macroinvertebrate samples were collected by technical staff employed by participating Councils or consultants engaged by Council. The methods for collection followed the Australian National River Health Program protocols (DEST et al. 1994; Chessman 2003). This involved collection using a ‘kick’ net, with 250-μm mesh and square 30×30 cm net frame (Chessman 1995) to survey edge and riffle habitat. A total of 10 m of stream habitat was sampled within a 100 m section of each site. Samples were live picked in the field on a sorting tray, preserved in ethanol and either dispatched to a commercial laboratory for identification or identified ‘in-house’ by experienced technical personnel using the recommended Australian taxonomic keys (Hawking and Smith 1997). The majority of taxa were identified to Family level. Due to difficulties in identification of acarina (water mites) and oligochaeta (segmented worms) these were identified at the Order level. Where riffle and edge samples were collected, picked and preserved separately, the data was homogenised to one sample to be representative of the sampling location. Four macroinvertebrate indices were calculated for each macroinvertebrate sample: Tippler, C., Findlay, S., Wright, I.A., Davies, P. J., Evans, C. & Ahmed, M. (2014). Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney, in Vietz, G; Rutherfurd, I.D, and Hughes, R. (editors), Proceedings of the 7th Australian Stream Management Conference. Townsville, Queensland, Pages 292-299. 294
7ASM Full Paper Tippler C. et al - Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney? 1. 2. 3. 4.
Percentage of ephemeroptera (mayfly), plecoptera (stonefly) and trichoptera (caddisfly) (%EPT) (Cairns and Pratt 1993) - a biotic index based on the percentage of pollution sensitive taxa present at a site; Family taxonomic richness(Rosenberg and Resh 1993); Macroinvertebrate abundance; Unweighted family level SIGNAL 2 scores (Chessman 2003) - Biotic index based on the presence/absence of pollution tolerant/intolerant invertebrate taxa.
Statistical Analysis All macroinvertebrate indices were compared between season (spring and autumn) and as a function of stream disturbance (reference and disturbed). Reference streams were defined as having naturally vegetated catchments with minimal urban or agricultural development. F-tests were applied to test for variance between sample groups and t-tests, assuming unequal variance, were applied to establish the significance between sample groups. Multivariate analysis was used to assess macroinvertebrate community response to season and stream disturbance. Non-metric multidimensional scaling (nMDS) was performed on a similarity matrix calculated with fourth root transformed macroinvertebrate data using the Bray-Curtis similarity measure (Clarke 1993; Warwick 1993). Data were grouped according to season of sampling and stream disturbance. Two-dimensional ordination plots were generated to give a representation of the similarity among samples and groups. Two-way crossed analysis of similarity (ANOSIM: Clarke 1993) was used to investigate the significance of difference in invertebrate community structure with data grouped by season of sampling and stream disturbance.
Results A total of 43,374 individual specimens from 141 macroinvertebrate families were collected. Three families contributed to approximately 30% of total abundance: chironomidae (9258 individuals) being the most abundant taxa followed by hydrobiidae (4138 individuals) and physidae (2597 individuals). Comparison between spring and autumn invertebrate indices showed no significant differences are evident between seasonal SIGNAL scores (t (385) = 1.21, p=0.22), seasonal richness (t (410) = 0.81, p=0.41) and seasonal % EPT taxa (t (375) = 0.65, p=0.52). In contrast seasonal abundance was significantly different (t(336) = 2.08, p=0.03) with mean abundance being higher in spring (117) than autumn (95) (Table 1). Comparison of invertebrate indices between reference streams and disturbed streams show SIGNAL scores varied significantly (t (95) = 13.29, p=
Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney? 1
2
3
1
4
Tippler C. , Findlay S. , Wright I.A. , Davies P.J. , Evans C. and Ahmed M.
5
1 Department of Environment and Geography, Macquarie University, NSW 2. Ku-ring-gai Council, Pymble, NSW 3. School of Science and Health, University of Western Sydney, South Penrith , NSW 4. Georges River Combined Councils’ Committee, Hurstville, NSW 5. Blacktown City Council, Blacktown NSW
Key Points • Freshwater macroinvertebrate communities of the Sydney basin do not exhibit significant seasonal variation. • Sampling of macroinvertebrate communities is traditionally undertaken during spring and autumn, however within the Sydney basin this approach is not necessary. • Findings of this study will enable waterway managers to refine monitoring programs.
Abstract This paper challenges the validity of current Australian guidelines that recommend seasonal sampling for freshwater macroinvertebrates. Macroinvertebrates are a commonly used indicator to assess the ecological condition of streams and various sampling methods provide a consistent and repeatable approach. Most methods recommend seasonal surveys, which traditionally occur in spring and autumn. The basis for a seasonal approach is to capture specimens at an ideal life-stage when identification can be easily achieved. This is particularly relevant for many invertebrate taxa with lifecycles that are influenced by and dependent on extreme seasonal climatic change. However in more temperate environments, where daytime temperatures seldom drop below 10 degrees Celsius the biological response of macroinvertebrate communities and in-turn, seasonal differences may not be as pronounced. This study reviewed family level macroinvertebrate data sets across the Sydney basin to investigate if seasonality is a factor that influences community composition between spring and autumn sampling periods. The results suggest the time of year that sampling is undertaken is not an important consideration within the study region and that this should inform future sampling protocols.
Keywords macroinvertebrates, seasonality, Sydney, invertebrate indices, sampling, catchment management
Introduction Throughout the world freshwater macroinvertebrates are commonly used as an indicator of urban stream ecosystem condition (Beavan et al. 2001; Paul and Meyer 2001; Walsh et al. 2001). This has been based on the relative ease of collection, identification of specimens and the sensitivities to pollution and environmental disturbance that many invertebrate taxa display (Chessman 2003). The common approach to sampling macroinvertebrates in streams and rivers is by qualitative surveys using sweep/kick net sampling techniques designed to capture a sample of biota within different environments such as pools, riffles and stream edges. A range of methods have advocated sampling during autumn and spring (Wright 2000; Turak and Waddell 2002) as seasonal variability is well known to influence the distribution and abundance of various invertebrate taxa and the composition of communities (Brooks 2000; Leung et al 2012; Sporka et al 2006). For example in many parts of the northern hemisphere where cold winters are followed by short warm summers, significant seasonal variation in both invertebrate community structure and the abundance of a many invertebrate taxa has been documented (Sporka et al 2006). Similarly, invertebrate communities in tropical monsoonal streams have been shown to exhibit strong seasonal variation due to flow variation resulting from the cyclical wet and dry nature of these ecosystems (Leigh 2012; Mesa 2012). The climate in the Sydney basin is best characterised as temperate with little difference in rainfall between seasons and relatively low differences in temperatures across the seasons. Summers are usually warm with a mean maximum Tippler, C., Findlay, S., Wright, I.A., Davies, P. J., Evans, C. & Ahmed, M. (2014). Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney, in Vietz, G; Rutherfurd, I.D, and Hughes, R. (editors), Proceedings of the 7th Australian Stream Management Conference. Townsville, Queensland, Pages 292-299. 292
7ASM Full Paper Tippler C. et al - Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney? temperature of 25°Celcius and winters mild with mean minimum of 8°Celcius. Outside of drought years, rainfall is regular throughout the year with highest average rainfall falling between February-May and September –October (BOM 2014). Elevation across the Sydney basin ranges from sea level to a maximum of 400 m above sea level. Sydney is fortunate to have a number of stream catchments with high coverage of remnant vegetation and with little to no urban development. This provides a notable contrast to streams within urbanised areas that can are best described as degraded due to catchment modification which has led to higher diurnal temperature differences, flashier hydrology, and poor water quality among other factors (Meyer et al 2005; Tippler et al 2012). When using macroinvertebrates as an indicator of stream health in Australia the most commonly advocated approaches follow that prescribed by the AUSRIVAS (Turak and Waddel 2002) or SIGNAL (Chessman 2003) methods. Both advocate seasonal surveys (autumn and spring) on the basis that this will help capture the greatest diversity of invertebrate taxa across their various life stages and ensure seasonal variation within any specific communities can be accounted for, irrespective of climate. Macroinvertebrate sampling undertaken in Sydney commonly use the AUSRIVAS and SIGNAL methods and apply autumn and spring monitoring protocols. In the published results of the various sampling programs, the influence of seasonality on invertebrate populations is often overlooked in the analysis. Tippler et al (2012) found that seasonality in the Georges River (to the south of Sydney) was not significant at taxonomic level of Order. In a study of streams from the Ku-ring-gai Local Government Area (to Sydney’s north) Wright (2011) reported seasonal differences in invertebrate community structure as measured at the species level was minimal. Outside these two studies no other peer-reviewed commentary on the impact of seasons with the study region has been found. This study specifically examines the influence of seasonality on macroinvertebrate community structure and commonly used invertebrate indices of SIGNAL, EPT, richness and abundance across Sydney. The results of this study are proposed to inform the future design of catchment management monitoring programs within this region and comparable climates.
Method Study Area The Sydney basin, as defined in this study, encompasses an area which extends from Hornsby in the north to Penrith in the west, Appin to the south and Bondi to the east. The study area involved 12 local government areas (Fig. 1) including Hornsby and Ku-ring-gai Councils to the north, Blacktown City Council to the west, Sutherland, Rockdale, Kogarah, Hurstville, and Bankstown City Councils to the south and Fairfield, Liverpool, Campbelltown and Wollondilly Councils to the south west. The Sydney basin is dominated by a temperate climate with warm summers and no dry season (NSW Department of Environment and Heritage 2014). Average annual rainfall varies from approximately 1400 mm in areas closest to the coast and decreases to approximately 700 mm in the western suburbs. Rainfall across the Sydney basin occurs regularly with slightly elevated monthly falls recorded in late autumn and early to mid-winter. The average daytime temperatures range from approximately 16 °C (winter) to 26 °C (summer) in areas closest to the coast and approximately 16 °C (winter) to 30 °C (summer) in areas furthest from the coast (BOM 2014).
Tippler, C., Findlay, S., Wright, I.A., Davies, P. J., Evans, C. & Ahmed, M. (2014). Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney, in Vietz, G; Rutherfurd, I.D, and Hughes, R. (editors), Proceedings of the 7th Australian Stream Management Conference. Townsville, Queensland, Pages 292-299. 293
7ASM Full Paper Tippler C. et al - Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney?
Figure 1. Map of the Sydney basin and participating Councils (coloured).
Aquatic Macroinvertebrate Sampling Macroinvertebrate sampling data was obtained from the participating councils between 2002 to 2013.This included 412 aquatic macroinvertebrate sampling results collected from 56 locations across 50 freshwater streams and rivers. 220 samples were collected during the autumn sample season and 192 collected during the spring sampling season. 60 samples were collected from reference streams (defined as those with intact naturally vegetated catchments) and 352 samples from disturbed streams, the majority with urbanised catchments and a small number impacted by agriculture. Macroinvertebrate samples were collected by technical staff employed by participating Councils or consultants engaged by Council. The methods for collection followed the Australian National River Health Program protocols (DEST et al. 1994; Chessman 2003). This involved collection using a ‘kick’ net, with 250-μm mesh and square 30×30 cm net frame (Chessman 1995) to survey edge and riffle habitat. A total of 10 m of stream habitat was sampled within a 100 m section of each site. Samples were live picked in the field on a sorting tray, preserved in ethanol and either dispatched to a commercial laboratory for identification or identified ‘in-house’ by experienced technical personnel using the recommended Australian taxonomic keys (Hawking and Smith 1997). The majority of taxa were identified to Family level. Due to difficulties in identification of acarina (water mites) and oligochaeta (segmented worms) these were identified at the Order level. Where riffle and edge samples were collected, picked and preserved separately, the data was homogenised to one sample to be representative of the sampling location. Four macroinvertebrate indices were calculated for each macroinvertebrate sample: Tippler, C., Findlay, S., Wright, I.A., Davies, P. J., Evans, C. & Ahmed, M. (2014). Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney, in Vietz, G; Rutherfurd, I.D, and Hughes, R. (editors), Proceedings of the 7th Australian Stream Management Conference. Townsville, Queensland, Pages 292-299. 294
7ASM Full Paper Tippler C. et al - Does seasonality influence freshwater macroinvertebrate communities in the temperate paradise of Sydney? 1. 2. 3. 4.
Percentage of ephemeroptera (mayfly), plecoptera (stonefly) and trichoptera (caddisfly) (%EPT) (Cairns and Pratt 1993) - a biotic index based on the percentage of pollution sensitive taxa present at a site; Family taxonomic richness(Rosenberg and Resh 1993); Macroinvertebrate abundance; Unweighted family level SIGNAL 2 scores (Chessman 2003) - Biotic index based on the presence/absence of pollution tolerant/intolerant invertebrate taxa.
Statistical Analysis All macroinvertebrate indices were compared between season (spring and autumn) and as a function of stream disturbance (reference and disturbed). Reference streams were defined as having naturally vegetated catchments with minimal urban or agricultural development. F-tests were applied to test for variance between sample groups and t-tests, assuming unequal variance, were applied to establish the significance between sample groups. Multivariate analysis was used to assess macroinvertebrate community response to season and stream disturbance. Non-metric multidimensional scaling (nMDS) was performed on a similarity matrix calculated with fourth root transformed macroinvertebrate data using the Bray-Curtis similarity measure (Clarke 1993; Warwick 1993). Data were grouped according to season of sampling and stream disturbance. Two-dimensional ordination plots were generated to give a representation of the similarity among samples and groups. Two-way crossed analysis of similarity (ANOSIM: Clarke 1993) was used to investigate the significance of difference in invertebrate community structure with data grouped by season of sampling and stream disturbance.
Results A total of 43,374 individual specimens from 141 macroinvertebrate families were collected. Three families contributed to approximately 30% of total abundance: chironomidae (9258 individuals) being the most abundant taxa followed by hydrobiidae (4138 individuals) and physidae (2597 individuals). Comparison between spring and autumn invertebrate indices showed no significant differences are evident between seasonal SIGNAL scores (t (385) = 1.21, p=0.22), seasonal richness (t (410) = 0.81, p=0.41) and seasonal % EPT taxa (t (375) = 0.65, p=0.52). In contrast seasonal abundance was significantly different (t(336) = 2.08, p=0.03) with mean abundance being higher in spring (117) than autumn (95) (Table 1). Comparison of invertebrate indices between reference streams and disturbed streams show SIGNAL scores varied significantly (t (95) = 13.29, p=
