Environ Monit Assess (2015) 187:70 DOI 10.1007/s10661-015-4313-0
Characterization of Chironomidae (Diptera) surface-floating pupal exuviae sample sort time from coastal tropical aquatic systems Petra Kranzfelder & Leonard C. Ferrington Jr.
Received: 25 June 2014 / Accepted: 18 January 2015 # Springer International Publishing Switzerland (outside the USA) 2015
Abstract Many studies either ignore chironomids or only identify specimens to subfamily or tribe due to the associated difficulty and high cost with processing and identifying larvae. An efficient form of sampling chironomids involves collections of surface-floating pupal exuviae (SFPE). SFPE sample sorting has been shown to be more time efficient than traditional dipnet methods in temperate urban and peri-urban streams. However, no published studies have tested the time efficiency of SFPE sample sorting from coastal tropical aquatic systems. We calculated sort times for SFPE samples collected from a coastal tropical stream and an estuary and used multiple linear regression analysis to quantify the relationship between sample sort time and number of specimens, average body length of specimens, and dry weight of sample residue. The average amount of time required to sort very small samples was 69.3 min, while moderate samples averaged 85.6 min and large samples averaged 153.5 min. However, on average, small samples were nine times more time consuming per specimen than large samples. Additionally, dry weight of small-sized residue and the number of specimens contributed significantly to sort time. Therefore, we recommend collecting larger samples, which can be achieved by sampling for 20 min over 200-m reaches for stream sites and 500- to 1,000-m reaches for riverine and estuarine sites. Also, we suggest P. Kranzfelder (*) : L. C. Ferrington Jr. Department of Entomology, University of Minnesota, 219 Hodson Hall, 1980 Folwell Ave., St. Paul, MN 55108-6125, USA e-mail: [email protected]
collecting during periods of low wave action and disturbance by boat wake to reduce the amount of sample residue. This research will enhance project planning and budgeting of future studies using the SFPE method to monitor coastal tropical aquatic systems. Keywords Aquatic insects . Time efficiency . Neotropical . Stream . Estuary
Introduction The Chironomidae (Insecta: Diptera) are a species-rich family of aquatic flies (Ferrington 2008) that are typically the most abundant and widely distributed macroinvertebrate taxon in freshwater (Pinder 1986; Ashe et al. 1987). With over 6,000 species currently known worldwide, they exploit a wide range of microhabitats in virtually every aquatic systems, including freshwater and saline environments, tropical and arctic regions, temporary and permanent water, and impacted, as well as, pristine waters (Ferrington et al. 2008; Armitage et al. 1995). Chironomids are ecologically important because they exhibit all functional feeding groups and behavioral categories exhibited collectively by Ephemeroptera, Plecoptera, and Trichoptera (Ferrington et al. 2008). Additionally, as a species-rich family, chironomids show a whole range of responses to anthropogenic stressors, such as eutrophication and acidification, exhibited by macroinvertebrates in general (Rosenberg 1992; Barbour et al. 1999).
70
Environ Monit Assess (2015) 187:70
Page 2 of 8
Many biomonitoring programs either ignore chironomids or only identify them to family, subfamily, or tribe level, due to the associated difficulty and high cost of processing and identifying larvae (Rosenberg 1992; Ferrington et al. 2008; Wilson and McGill 1977). However, taxonomic resolution at supra-generic level can have dramatic effects on bioassessment scores, since sensitivities to anthropogenic contaminants are often unique to individual species (Jones 2008; Barbour et al. 1999). In addition to the difficulties associated with species identification of larvae, collecting chironomids from Neotropical aquatic ecosystems can be particularly challenging, since many taxa burrow during daylight due to predation pressure from invertebrate predators (Rosemond et al. 1998; Ramirez and Pringle 1998; Pringle and Hamazaki 1998) and aquatic systems may be composed of difficult to sample substrates, such as sand or silt (Lenat 1988; Rinella and Feminella 2005). In a rapid bioassessment study, Mora et al. (2003) collected benthic macroinvertebrate dip-net samples from several sample sites in Laguna del Tortuguero, Tortuguero National Park, Costa Rica. They reported a total of six orders representing 11 families (and 12 genera). Chironomids were the dominant macroinvertebrate taxa collected in these samples and the only family present in all samples; however, only 27 individuals were collected. An efficient, effective, low-cost, and easy-to-use form of sampling Chironomidae involves collections of surface-floating pupal exuviae (SFPE), which offers many advantages over the collection of larvae (Coffman 1973; Ferrington et al. 1991). In comparison to collections of larvae from benthic samples, SFPE collections improved taxonomic resolution by enhancing ease of genus and species identification, increased the number of taxa detected, and allowed for collection from a range of microhabitats, including those difficult to sample with other sampling methods (e.g., wood, hyporheic, sand, and deep water) (Ferrington et al. 2008). In addition, this method provides estimates of individuals that survived as immatures and emerged as functional adults; therefore, SFPE collections can be considered a better measure of the water quality characteristics of the aquatic system, since they correspond to an organism that successfully utilized the habitat (Bouchard and Ferrington 2011) (for a more in-depth review of the advantages and disadvantages of the SFPE method, see Kranzfelder et al. (in press)).
Given the limited resources for environmental monitoring and assessment of aquatic ecosystems, it is important to be able to readily predict the costs associated with accomplishing project objectives. Comparing traditional dip-net and SFPE sampling techniques, samples take on average 19 and 5.3 times more time than field collection, respectively, to sort (Ferrington et al. 1991). Therefore, it is important to have accurate estimates of the time required to sort samples for program budgeting purposes. SFPE collections have been shown to be more cost efficient than traditional dip-net sampling techniques in temperate urban and peri-urban streams (Anderson and Ferrington 2012; Ferrington et al. 1991). For example, Ferrington et al. (1991) demonstrated the efficiency of using SFPE methodology compared to dip-net sampling for larvae in an organically enriched urban stream in Kansas. They documented that three to four SFPE samples can be processed in the lab for every one dip-net sample (Ferrington et al. 1991). Similarly, Anderson and Ferrington (2012), working in urban and peri-urban trout streams in northeastern Minnesota, concluded that SFPE samples took approximately 70 % less time to sort than dip-net samples. To date, no studies have tested the time efficiency of SFPE sample processing in coastal tropical streams and estuaries. Additionally, the sample variables that influence individual sample sort time are unknown. Our study was part of an effort to test and standardize rapid biomonitoring methods using Chironomidae as bioindicators of water quality of coastal tropical waters on the Caribbean coast of Costa Rica. Another study was conducted to evaluate the spatial and temporal variability of Chironomidae species emergence in the same estuarine system (Kranzfelder in revision). The aims of this study were to (1) calculate sort times of SFPE samples collected from a coastal tropical stream, Quebrada, and an estuary, Laguna del Tortuguero, and (2) quantify the influence of the number of specimens, average body length of specimens, and dry weight of small-sized and large-sized residues on sort times.
Methods Study area Samples were collected from three contiguous sample sites in a Neotropical estuary, Laguna del Tortuguero, and from one sample site in the freshwater stream,
Environ Monit Assess (2015) 187:70
Quebrada, in Tortuguero National Park, Costa Rica. Tortuguero National Park lies within the Caribbean coastal lowlands of Costa Rica’s Limon providence between 10° 20′ and 10° 35′ latitude north (Fig. 1). The park contains a diverse terrestrial environment ranging from lowland tropical rainforest to palm swamps and halophytic coastal vegetation. The region experiences high annual rainfall, averaging over 5 m per year with two rainy seasons (July to August and November to January) (Winemiller and Leslie 1992). Generally, at least 50 mm of rain is received every month and temperatures show little seasonal variation with an annual mean of 23–26 °C (Hirth 1963; Nuhn et al. 1967). Samples were collected from the northern basin of Laguna Tortuguero, which is a 6.5-km elongate channel with an average depth of 7 m and a width of 300–400 m (Nordlie and Kelso 1975). The freshwater stream, Quebrada, is located on the barrier island that forms the partition between the Caribbean Sea and Laguna del Tortuguero (Winemiller and Ponwith 1998). Nordlie and Kelso (1975) measured saline water (8.7– 10.5 ppt) near the bottom of Laguna del Tortuguero (depths >5 m) during both the dry and wet seasons. Surface salinity measurements were typically less than or equal to 0.1 ppt in both Quebrada and Laguna del Tortuguero (Nordlie and Kelso 1975; Winemiller and Leslie 1992). Laguna del Tortuguero bottom substrate was dominated by sand, while Quebrada substrate was predominately sand with vegetative litter and silt. Field and laboratory methods Chironomidae SFPE samples were collected from each of the four sample sites on seven consecutive days in June 2010 and in January 2011, resulting in a total of 56 samples. June collections occurred within the drier season, with mean daily precipitation of 11 mm and total precipitation of 323 mm for June 2010; comparatively, January collections fell within the wetter season, with mean daily precipitation of 17 mm and total precipitation of 507 mm for January 2011 (Caño Palma Biological Station: +10° 35′ N, −83° 31′ W). Field collections followed the methods outlined by Ferrington et al. (1991). Briefly, SFPE samples were collected by dipping a white enamel pan into areas of known SFPE accumulation, such as areas with foam and/or debris behind fallen trees. Contents from the pan were then poured through a
Page 3 of 8 70
125-μm-aperture US Standard test sieve to retain SFPE and residue. This process was repeated for 10 min, working from downstream to upstream, and then, sieve contents were preserved in 80 % ethanol (Ferrington et al. 1991). In the laboratory, a small portion of the sample was placed in a Petri dish, and under a dissecting microscope, SFPE were picked from the sample into 1-dram vials with 80 % ethanol. After a complete scan of the dish, the dish was swirled and scanned for additional pupal exuviae. This was repeated until there were no more pupal exuviae detected in the dish on two consecutive scans. Samples with fewer than 500 pupal exuviae were completely picked and sorted, while samples with more than 500 pupal exuviae were subsampled. A subsample size of 500 specimens was on average sufficient to identify a large proportion (91 %) of the taxa collected in previous studies that collected SFPE samples (Bouchard and Ferrington 2011). Times (in minutes) needed to pick the specimens from each sample and the total numbers of specimens picked were recorded. All laboratory work reported here was completed by the first author using standardized laboratory protocols to avoid individual variation in sorting times. Average body length of all individuals within a genus were measured from the anterior-most margin of the cephalothorax to the posterior-most margin of the anal lobes of the abdomen (Wiederholm 1986). Body length was measured for all individuals in the genus if there were less than 20 individuals in the genus, or a subsample of 20 individuals if there were more than 20 individuals in the genus. For each selected individual, the cephalothorax and abdomen of slide mounted specimens was measured to the nearest 0.5 mm using × 12 magnification. Several of the SFPE samples collected for this study contained coarse detritus and fine inorganic and organic sediments. Dry weight was quantified to characterize the residue composition in each SFPE sample. Residue less than 2.0 mm in diameter was separated from residue greater than 2.0 mm in diameter by using a 125-μm-aperature sieve that was placed underneath a 2.0-mm-aperture sieve. Then, both residue size classes were transferred from the sieves to separate Petri dishes and dried in an incubator at 40 °C for at least 24 h. Finally, the dry weight of the residue was weighed to the nearest 0.01 g.
70
Page 4 of 8
Environ Monit Assess (2015) 187:70
Environ Monit Assess (2015) 187:70
Page 5 of 8 70
Fig. 1
Sample sites in Tortuguero Conservation Area, Costa Rica. LT Laguna del Tortuguero. Freshwater flows from Rio Penetencia and Rio Tortuguero into Laguna del Tortuguero
Data analysis We used multiple linear regression analysis (Jongman et al. 1995) to quantify the relationship between sample sort time and number of specimens, average body length of specimens, dry weight of the less than 2.0-mm diameter residue, and dry weight of the greater than 2.0-mm diameter residue. The dependent variable, sort time, did not meet linear model assumptions. Therefore, a BoxCox test was performed using R version 3.0.3© to find the optimal inverse square root power transformation on sort time to normalize the data (Box and Cox 1964).
weight of the greater than 2.0-mm diameter residue was 0.65 g per sample. The mean generic body length of individuals was 4.14 mm, and the mean number of specimens per SFPE sample was 62 (Table 2). Sort time was significantly influenced by residue that was less than 2.0 mm in diameter (p≤0.001) and number of specimens (p=0.01), but not by body length (>0.70) or residue greater than 2.0 mm in diameter (>0.10) (multiple regression, R2adj =0.55, p≤0.001) (Table 3). Specifically, there was a significant relationship between inverse square root of sort time and residue that was less than 2.0 mm in diameter (Fig. 2) and a significant relationship between inverse square root of sort time and the number of specimens (Fig. 3). Therefore, the sample sort time increased as functions of both the dry weight of residue less than 2.0 mm in diameter and the number of specimen (Table 3).
Discussion
Results The minimum time required to sort a sample was 15 min and the maximum time required was 320 min, with 83.9 % of the samples being sorted in 100 min or less. For samples that had less than 100 specimens, the average sample had 25.2 specimens, which were sorted in an average of 69.3 or 2.7 min per specimen. For samples that had between 101 and 300 specimens, the average sample had 183.9 specimens, which were sorted in an average of 85.6 or 0.5 min per specimen. For samples that had between 301 and 500 specimens, the average sample had 500 specimens, which were sorted in an average of 153.5 or 0.3 min per specimen (Table 1). On average, the small samples were nine times more time consuming per specimen than large samples. The mean dry weight of the less than 2.0-mm diameter residue was 2.38 g per sample, while the mean dry
Table 1 Mean SFPE sample sort time and mean number of specimens per sample for various subsample groupings Subsample groupings
N
T (S.E.)
Up to 100 specimens
47 69.3 (8.5)
n
T =n 25.2 2.7
Between 101 to 300 specimens
7 85.6 (18.7)
183.9 0.5
Between 301 to 500 specimens
2 153.5 (23.5) 500.0 0.3
N number of samples, T mean SFPE sample sort time (in minutes), n mean number of specimens per sample
Coastal tropical watersheds are among the most dynamic aquatic ecosystems on earth, with enormous spatial and temporal complexity. They are influenced by continental runoff waters, which are often rich in nutrients derived from urban, agricultural, and industrial activities (Pringle et al. 2000). Given the limited resources available for biological monitoring programs of these tropical aquatic systems (Ramírez et al. 2008), it is important to utilize protocols that efficiently determine patterns of chironomid community composition. Our results support other studies showing that Chironomidae SFPE sample sorting has been shown to be more time efficient than traditional dip-net techniques for assessing chironomid community composition (Anderson and Ferrington 2012; Ferrington et al. 1991). Most samples could be sorted in less than 100 min, but on average, it
Table 2 Mean and range of variables measured in SFPE samples Variable
N
Mean (SE)
Range
Residue less than 2.0-mm diameter (g) Residue greater than 2.0-mm diameter (g) Body length (mm)
56
2.38 (0.23)
0.09–5.36
56
0.65 (0.09)
0.06–4.60
44
4.14 (0.18)
1.50–6.67
Number of specimens
56
62 (14)
0–500
N number of samples
70
Environ Monit Assess (2015) 187:70
Page 6 of 8
Table 3 Multiple linear regression, considering sort time as the dependent variable Variable
Partial correlation
Standard error
p value
Residue less than 2.0-mm diameter (g) Residue greater than 2.0-mm diameter (g) Body length (mm)
−0.01325
0.00231
0.10
0.00193
0.00336
>0.70
Number of specimens
−0.00009
0.00004
0.01
took 85.6 min to sort a subsample of 300 specimens. Tropical coastal water SFPE sample sort times were less time efficient than sort times of temperate urban streams. The average time needed to sort a subsample of 300 specimens was approximately twice as much for tropical coastal water samples compared to temperate urban streams (Anderson and Ferrington 2012; Ferrington et al. 1991). The variable contributing most significantly to SFPE sample sort times was the amount of small-sized residue. During field collections, larger organic and inorganic materials were suspended in the water column from the benthic substrates via mechanical disturbances (e.g., wave action) and smaller materials were floating on the surface of the water. Therefore, organic material (e.g., pieces of leaves, flowers, sticks, terrestrial, and aquatic invertebrates) and inorganic material (e.g., plastic, styrofoam, silt, and sand) were incidentally transferred to the sieve and included in the SFPE sample (P. Kranzfelder, personal observations). As a result,
Fig. 2 Relationship between the inverse square root of sort time (in minutes) and dry weight of residue less than 2.0 mm in diameter (in grams)
Fig. 3 Relationship between the inverse square root of sort time (in minutes) and number of specimens
bioassessment programs sampling chironomids in aquatic systems with high total suspended solids should plan for increased SFPE sample sort time. For example, our results suggest that a SFPE sample with 5.0 grams of small-sized residue should require three times as long to sort as a sample with 0.5 g of small-sized residue. While it is impossible to completely eliminate sample residue, caution should be taken during field collection to reduce the amount of mechanical disturbance created from walking on bottom substrates and wave action from boat wakes. While the number of specimens in the sample contributed significantly to the sort time of individual samples, very small samples (i.e., less than 100 specimens) were much more time consuming to sort per specimen than medium samples (i.e., between 101 and 300 specimens) and large samples (i.e., between 301 and 500 specimens). Therefore, our results, similar to Anderson and Ferrington (2012), suggest that samples with larger numbers of exuviae are more efficient, and thus, sampling effort in the field should be adjusted to ensure that a sample contains a large number of pupal exuviae. In addition to increased lab processing efficiency, Ruse (2011) suggests that by increasing the number of pupal exuviae identified from a single sample, there is an increase in probability of including Brare^ species that are not close to their peak emergence period. Increased sample size can be achieved by extending the time spent sampling from 10 to 20 min per site and/or increasing the length of the sampling reach from 100 to 200 m for stream sites and from 500 to 1,000 m for riverine and estuarine sites. While sort times from coastal tropical waters were more time consuming than samples from temperate
Environ Monit Assess (2015) 187:70
urban streams, it appears that alternative sampling methods, such as the traditional dip-net sampling, would be relatively more time consuming. Both Ferrington et al. (1991) and Anderson and Ferrington (2012) found that sample sorting of SFPE samples took about 70 % less time compared to dip-net samples. Stein et al. (2008) found that sort times took an average of 4–7 h for benthic macroinvertebrate dip-net samples collected from the nearby Dos Novillos River on the Caribbean coast of Costa Rica. They suggested that the long sort time was due to separating the invertebrates from organic and inorganic material in the sample (Stein et al. 2008). The SFPE samples processed for this study contained 98 species of chironomids, which are distributed among 44 genera. Dip-net samples were taken from the same sample sites in Laguna del Tortuguero at a later date as part of another study (Kranzfelder, in preparation). The results from these dip-net samples are similar to the findings of Mora et al. (2003), with six orders of aquatic or semiaquatic insects detected. Compared to Mora et al. (2003), additional Hemiptera genera (six versus five), Ephemeroptera genera (two versus one), Coleoptera genera (three versus one), and Chironomidae (eight genera, none reported by Mora et al. (2003)) were detected, resulting in a total of 22 genera now known for Laguna del Tortuguero. However, our dip-net samples only detected four genera of EPT, demonstrating the potential ineffectiveness of metrics based on these three orders for water quality studies in this estuarine system. By contrast, the 98 species of Chironomidae detected using the SFPE method are relatively diverse at genus level (44 genera) and this method should be considered as a preferred replacement to dip-net sampling for biomonitoring in this estuarine environment. The results of this study will enhance planning and budgeting for future field sampling designs by providing guidelines for biomonitoring programs using the SFPE method. Acknowledgments We thank Costa Rica’s Ministerio del Ambiente y Energia for kindly permitting access to the study area and for providing research permit ACTo-GASP-PIN-016-010. We also wish to thank Jessica Miller and Catherine DeGuire for field and laboratory assistance and the Canadian Organization for Tropical Education and Rainforest Conservation for providing precipitation data. Funding for this project was provided by the Department of Entomology in the College of Foods Agricultural and Natural Resource Sciences. This paper is published under the auspices of the Chironomidae Research Group at the University of Minnesota.
Page 7 of 8 70
References Anderson, A. M., & Ferrington, L. C. (2012). Time-efficiency of sorting Chironomidae surface-floating pupal exuviae samples from urban trout streams in Northeast Minnesota, USA. In E. E. Stur & K. Aagaard (Eds.), 18th International Symposium on Chironomidae. Trondheim, Norway: Fauna norvegica. Armitage, P. D., Cranston, P. S., & Pinder, L. C. V. (1995). The Chironomidae: biology and ecology of non-biting midges: Chapman & Hall. Ashe, P., Murray, D. A., & Reiss, F. (1987). The zoogeographical distribution of Chironomidae (Insecta: Diptera). Annales de Limnologie, 23(1), 27–60. 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. (2nd ed., Vol. EPA 841-B-99-002) (2nd ed.). Washington, D.C.: U.S. Environmental Protection Agency, Office of Water. Bouchard, R. W., & Ferrington, L. C. (2011). The effects of subsampling and sampling frequency on the use of surfacefloating pupal exuviae to measure Chironomidae (Diptera) communities in wadeable temperate streams. Environmental Monitoring and Assessment, 181(1–4), 205–223. Box, G. E. P., & Cox, D. R. (1964). An analysis of transformations. Journal of the Royal Statistical Society. Series B (Methodological), 26(2), 211–252. Coffman, W. P. (1973). Energy flow in a woodland stream ecosystem: II. The taxonomic composition of the Chironomidae as determined by the collection of pupal exuviae. Archiv fur Hydrobiologie, 71, 281–322. Ferrington, L. C. (2008). Global diversity of non-biting midges (Chironomidae; Insecta-Diptera) in freshwater. In E. V. Balian, C. Lévêque, H. Segers, & K. Martens (Eds.), Freshwater animal diversity assessment, Developments in Hydrobiology, Vol. 198 (pp. 447–455). Netherlands: Springer Netherlands. Ferrington, L. C., Blackwood, M. A., Wright, C. A., Crisp, N. H., Kavanaugh, J. L., & Schmidt, F. J. (1991). A protocol for using surface-floating pupal exuviae of Chironomidae for rapid bioassessment of changing water-quality. In N. E. Peters & D. E. Walling (Eds.), Sediment and stream water quality in a changing environment: trends and explanation. Oxfordshire, U.K: International Association of Hydrological Sciences Press. Ferrington, L. C., Berg, M. B., & Coffman, W. P. (2008). Chironomidae. In R. W. Merritt, K. W. Cummins, & M. B. Berg (Eds.), An introduction to the aquatic insects of North America (4th ed., pp. 847–989). Dubuque, IA: Kendall/Hunt Publishing Company. Hirth, H. F. (1963). The ecology of two lizards on a tropical beach. Ecological Monographs, 33, 83–112. Jones, F. C. (2008). Taxonomic sufficiency: the influence of taxonomic resolution on freshwater bioassessments using benthic macroinvertebrates. Environmental Reviews, 16, 45–69. Jongman, R. H. G., Ter Braak, C. J. F., & van Tongeren, O. F. R. (1995). Data analysis in community and landscape ecology. Cambridge, UK: Cambridge University Press.
70
Page 8 of 8
Kranzfelder, P. (in revision). Temporal and spatial variability of Chironomidae (Insecta: Diptera) species emergence in a Neotropical estuary. Freshwater Science. Kranzfelder, P. (in preparation). Cumulative effects of coastal watershed land use on Chironomidae (Insecta: Diptera) communities of Neotropical estuaries in Costa Rica. University of Minnesota, Minneapolis, MN. Kranzfelder, P., Anderson, A. M., Egan, A. T., Mazack, J. E., Bouchard, R. W., & Ferrington, L. C. (in press). Use of Chironomidae (Diptera) surface-floating pupal exuviae as a rapid bioasessment protocol for water bodies. Journal of Visualized Experiments (JoVE). Lenat, D. R. (1988). Water quality assessment of streams using a qualitative collection method for benthic macroinvertebrates. Journal of the North American Benthological Society, 7(3), 222–233. Mora, J. M., Rodríguez, M. A., & López, L. I. (2003). Sondeo ecológico rápido y monitoreo de especies indicadoras en el Parque Nacional Tortuguero. In INBIO (Ed.). Guapiles, Costa Rica. Nordlie, F. G., & Kelso, D. P. (1975). Trophic relationships in a tropical estuary. Revista de Biología Tropical, 23, 77–99. Nuhn, H., Pérez, S. R., & Sandner, G. (1967). Estudio geográfico regional de la zona atlántico norte de Costa Rica. In I. d. T. y. Colonización (Ed.). San José, Costa Rica: Instituto de Tierras y Colonización. Pinder, L. C. V. (1986). Biology of freshwater Chironomidae. Annual Review of Entomology, 31, 1–23. Pringle, C. M., & Hamazaki, T. (1998). The role of omnivory in a Neotropical stream: separating diurnal and nocturnal effects. Ecology, 79(1), 269–280. Pringle, C. M., Scatena, F. N., Paaby-Hansen, P., & Nunez-Ferrera, M. (2000). River conservation in Latin America and the Caribbean. In P. J. Boon, B. Davies, & G. E. Petts (Eds.), Global perspectives on river conservation: science, policy and practice (pp. 41–77). Durham, NC: John Wiley and Sons.
Environ Monit Assess (2015) 187:70 Ramirez, A., & Pringle, C. M. (1998). Invertebrate drift and benthic community dynamics in a lowland Neotropical stream, Costa Rica. Hydrobiologia, 386, 19–26. Ramírez, A., Pringle, C. M., & Wantzen, K. M. (2008). Tropical stream conservation. In D. Dudgeon (Ed.), Tropical stream ecology (pp. 285–304). London: Academic Press. Rinella, D. J., & Feminella, J. W. (2005). Comparison of benthic macroinvertebrates colonizing sand, wood, and artificial substrates in a low-gradient stream. Journal of Freshwater Ecology, 20(2), 209–220. Rosemond, A. D., Pringle, C. M., & Ramírez, A. (1998). Macroconsumer effects on insect detritivores and detritus processing in a tropical stream. Freshwater Biology, 39(3), 515–523. Rosenberg, D. (1992). Freshwater biomonitoring and Chironomidae. Netherland Journal of Aquatic Ecology, 26(2–4), 101–122. Ruse, L. (2011). Lake acidification assessed using chironomid pupal exuviae. Fundamental and Applied Limnology, 178(4), 267–286. Stein, H., Springer, M., & Kohlmann, B. (2008). Comparison of two sampling methods for biomonitoring using aquatic macroinvertebrates in the Dos Novillos River, Costa Rica. Ecological Engineering, 34(4), 267–275. Wiederholm, T. (1986). Chironomidae of the Holarctic region. Part 2. Pupae (Vol. 28): Entomologica Scandinavica Supplement. Wilson, R. S., & McGill, J. D. (1977). A new method of monitoring water quality in a stream receiving sewage effluent, using chironomid pupal exuviae. Water Research, 11(11), 959– 962. Winemiller, K. O., & Leslie, M. A. (1992). Fish assemblages across a complex, tropical freshwater/marine ecotone. Environmental Biology of Fishes, 34(1), 29–50. Winemiller, K. O., & Ponwith, B. J. (1998). Comparative ecology of eleotrid fishes in Central American coastal streams. Environmental Biology of Fishes, 53, 373–384.
Characterization of Chironomidae (Diptera) surface-floating pupal exuviae sample sort time from coastal tropical aquatic systems Petra Kranzfelder & Leonard C. Ferrington Jr.
Received: 25 June 2014 / Accepted: 18 January 2015 # Springer International Publishing Switzerland (outside the USA) 2015
Abstract Many studies either ignore chironomids or only identify specimens to subfamily or tribe due to the associated difficulty and high cost with processing and identifying larvae. An efficient form of sampling chironomids involves collections of surface-floating pupal exuviae (SFPE). SFPE sample sorting has been shown to be more time efficient than traditional dipnet methods in temperate urban and peri-urban streams. However, no published studies have tested the time efficiency of SFPE sample sorting from coastal tropical aquatic systems. We calculated sort times for SFPE samples collected from a coastal tropical stream and an estuary and used multiple linear regression analysis to quantify the relationship between sample sort time and number of specimens, average body length of specimens, and dry weight of sample residue. The average amount of time required to sort very small samples was 69.3 min, while moderate samples averaged 85.6 min and large samples averaged 153.5 min. However, on average, small samples were nine times more time consuming per specimen than large samples. Additionally, dry weight of small-sized residue and the number of specimens contributed significantly to sort time. Therefore, we recommend collecting larger samples, which can be achieved by sampling for 20 min over 200-m reaches for stream sites and 500- to 1,000-m reaches for riverine and estuarine sites. Also, we suggest P. Kranzfelder (*) : L. C. Ferrington Jr. Department of Entomology, University of Minnesota, 219 Hodson Hall, 1980 Folwell Ave., St. Paul, MN 55108-6125, USA e-mail: [email protected]
collecting during periods of low wave action and disturbance by boat wake to reduce the amount of sample residue. This research will enhance project planning and budgeting of future studies using the SFPE method to monitor coastal tropical aquatic systems. Keywords Aquatic insects . Time efficiency . Neotropical . Stream . Estuary
Introduction The Chironomidae (Insecta: Diptera) are a species-rich family of aquatic flies (Ferrington 2008) that are typically the most abundant and widely distributed macroinvertebrate taxon in freshwater (Pinder 1986; Ashe et al. 1987). With over 6,000 species currently known worldwide, they exploit a wide range of microhabitats in virtually every aquatic systems, including freshwater and saline environments, tropical and arctic regions, temporary and permanent water, and impacted, as well as, pristine waters (Ferrington et al. 2008; Armitage et al. 1995). Chironomids are ecologically important because they exhibit all functional feeding groups and behavioral categories exhibited collectively by Ephemeroptera, Plecoptera, and Trichoptera (Ferrington et al. 2008). Additionally, as a species-rich family, chironomids show a whole range of responses to anthropogenic stressors, such as eutrophication and acidification, exhibited by macroinvertebrates in general (Rosenberg 1992; Barbour et al. 1999).
70
Environ Monit Assess (2015) 187:70
Page 2 of 8
Many biomonitoring programs either ignore chironomids or only identify them to family, subfamily, or tribe level, due to the associated difficulty and high cost of processing and identifying larvae (Rosenberg 1992; Ferrington et al. 2008; Wilson and McGill 1977). However, taxonomic resolution at supra-generic level can have dramatic effects on bioassessment scores, since sensitivities to anthropogenic contaminants are often unique to individual species (Jones 2008; Barbour et al. 1999). In addition to the difficulties associated with species identification of larvae, collecting chironomids from Neotropical aquatic ecosystems can be particularly challenging, since many taxa burrow during daylight due to predation pressure from invertebrate predators (Rosemond et al. 1998; Ramirez and Pringle 1998; Pringle and Hamazaki 1998) and aquatic systems may be composed of difficult to sample substrates, such as sand or silt (Lenat 1988; Rinella and Feminella 2005). In a rapid bioassessment study, Mora et al. (2003) collected benthic macroinvertebrate dip-net samples from several sample sites in Laguna del Tortuguero, Tortuguero National Park, Costa Rica. They reported a total of six orders representing 11 families (and 12 genera). Chironomids were the dominant macroinvertebrate taxa collected in these samples and the only family present in all samples; however, only 27 individuals were collected. An efficient, effective, low-cost, and easy-to-use form of sampling Chironomidae involves collections of surface-floating pupal exuviae (SFPE), which offers many advantages over the collection of larvae (Coffman 1973; Ferrington et al. 1991). In comparison to collections of larvae from benthic samples, SFPE collections improved taxonomic resolution by enhancing ease of genus and species identification, increased the number of taxa detected, and allowed for collection from a range of microhabitats, including those difficult to sample with other sampling methods (e.g., wood, hyporheic, sand, and deep water) (Ferrington et al. 2008). In addition, this method provides estimates of individuals that survived as immatures and emerged as functional adults; therefore, SFPE collections can be considered a better measure of the water quality characteristics of the aquatic system, since they correspond to an organism that successfully utilized the habitat (Bouchard and Ferrington 2011) (for a more in-depth review of the advantages and disadvantages of the SFPE method, see Kranzfelder et al. (in press)).
Given the limited resources for environmental monitoring and assessment of aquatic ecosystems, it is important to be able to readily predict the costs associated with accomplishing project objectives. Comparing traditional dip-net and SFPE sampling techniques, samples take on average 19 and 5.3 times more time than field collection, respectively, to sort (Ferrington et al. 1991). Therefore, it is important to have accurate estimates of the time required to sort samples for program budgeting purposes. SFPE collections have been shown to be more cost efficient than traditional dip-net sampling techniques in temperate urban and peri-urban streams (Anderson and Ferrington 2012; Ferrington et al. 1991). For example, Ferrington et al. (1991) demonstrated the efficiency of using SFPE methodology compared to dip-net sampling for larvae in an organically enriched urban stream in Kansas. They documented that three to four SFPE samples can be processed in the lab for every one dip-net sample (Ferrington et al. 1991). Similarly, Anderson and Ferrington (2012), working in urban and peri-urban trout streams in northeastern Minnesota, concluded that SFPE samples took approximately 70 % less time to sort than dip-net samples. To date, no studies have tested the time efficiency of SFPE sample processing in coastal tropical streams and estuaries. Additionally, the sample variables that influence individual sample sort time are unknown. Our study was part of an effort to test and standardize rapid biomonitoring methods using Chironomidae as bioindicators of water quality of coastal tropical waters on the Caribbean coast of Costa Rica. Another study was conducted to evaluate the spatial and temporal variability of Chironomidae species emergence in the same estuarine system (Kranzfelder in revision). The aims of this study were to (1) calculate sort times of SFPE samples collected from a coastal tropical stream, Quebrada, and an estuary, Laguna del Tortuguero, and (2) quantify the influence of the number of specimens, average body length of specimens, and dry weight of small-sized and large-sized residues on sort times.
Methods Study area Samples were collected from three contiguous sample sites in a Neotropical estuary, Laguna del Tortuguero, and from one sample site in the freshwater stream,
Environ Monit Assess (2015) 187:70
Quebrada, in Tortuguero National Park, Costa Rica. Tortuguero National Park lies within the Caribbean coastal lowlands of Costa Rica’s Limon providence between 10° 20′ and 10° 35′ latitude north (Fig. 1). The park contains a diverse terrestrial environment ranging from lowland tropical rainforest to palm swamps and halophytic coastal vegetation. The region experiences high annual rainfall, averaging over 5 m per year with two rainy seasons (July to August and November to January) (Winemiller and Leslie 1992). Generally, at least 50 mm of rain is received every month and temperatures show little seasonal variation with an annual mean of 23–26 °C (Hirth 1963; Nuhn et al. 1967). Samples were collected from the northern basin of Laguna Tortuguero, which is a 6.5-km elongate channel with an average depth of 7 m and a width of 300–400 m (Nordlie and Kelso 1975). The freshwater stream, Quebrada, is located on the barrier island that forms the partition between the Caribbean Sea and Laguna del Tortuguero (Winemiller and Ponwith 1998). Nordlie and Kelso (1975) measured saline water (8.7– 10.5 ppt) near the bottom of Laguna del Tortuguero (depths >5 m) during both the dry and wet seasons. Surface salinity measurements were typically less than or equal to 0.1 ppt in both Quebrada and Laguna del Tortuguero (Nordlie and Kelso 1975; Winemiller and Leslie 1992). Laguna del Tortuguero bottom substrate was dominated by sand, while Quebrada substrate was predominately sand with vegetative litter and silt. Field and laboratory methods Chironomidae SFPE samples were collected from each of the four sample sites on seven consecutive days in June 2010 and in January 2011, resulting in a total of 56 samples. June collections occurred within the drier season, with mean daily precipitation of 11 mm and total precipitation of 323 mm for June 2010; comparatively, January collections fell within the wetter season, with mean daily precipitation of 17 mm and total precipitation of 507 mm for January 2011 (Caño Palma Biological Station: +10° 35′ N, −83° 31′ W). Field collections followed the methods outlined by Ferrington et al. (1991). Briefly, SFPE samples were collected by dipping a white enamel pan into areas of known SFPE accumulation, such as areas with foam and/or debris behind fallen trees. Contents from the pan were then poured through a
Page 3 of 8 70
125-μm-aperture US Standard test sieve to retain SFPE and residue. This process was repeated for 10 min, working from downstream to upstream, and then, sieve contents were preserved in 80 % ethanol (Ferrington et al. 1991). In the laboratory, a small portion of the sample was placed in a Petri dish, and under a dissecting microscope, SFPE were picked from the sample into 1-dram vials with 80 % ethanol. After a complete scan of the dish, the dish was swirled and scanned for additional pupal exuviae. This was repeated until there were no more pupal exuviae detected in the dish on two consecutive scans. Samples with fewer than 500 pupal exuviae were completely picked and sorted, while samples with more than 500 pupal exuviae were subsampled. A subsample size of 500 specimens was on average sufficient to identify a large proportion (91 %) of the taxa collected in previous studies that collected SFPE samples (Bouchard and Ferrington 2011). Times (in minutes) needed to pick the specimens from each sample and the total numbers of specimens picked were recorded. All laboratory work reported here was completed by the first author using standardized laboratory protocols to avoid individual variation in sorting times. Average body length of all individuals within a genus were measured from the anterior-most margin of the cephalothorax to the posterior-most margin of the anal lobes of the abdomen (Wiederholm 1986). Body length was measured for all individuals in the genus if there were less than 20 individuals in the genus, or a subsample of 20 individuals if there were more than 20 individuals in the genus. For each selected individual, the cephalothorax and abdomen of slide mounted specimens was measured to the nearest 0.5 mm using × 12 magnification. Several of the SFPE samples collected for this study contained coarse detritus and fine inorganic and organic sediments. Dry weight was quantified to characterize the residue composition in each SFPE sample. Residue less than 2.0 mm in diameter was separated from residue greater than 2.0 mm in diameter by using a 125-μm-aperature sieve that was placed underneath a 2.0-mm-aperture sieve. Then, both residue size classes were transferred from the sieves to separate Petri dishes and dried in an incubator at 40 °C for at least 24 h. Finally, the dry weight of the residue was weighed to the nearest 0.01 g.
70
Page 4 of 8
Environ Monit Assess (2015) 187:70
Environ Monit Assess (2015) 187:70
Page 5 of 8 70
Fig. 1
Sample sites in Tortuguero Conservation Area, Costa Rica. LT Laguna del Tortuguero. Freshwater flows from Rio Penetencia and Rio Tortuguero into Laguna del Tortuguero
Data analysis We used multiple linear regression analysis (Jongman et al. 1995) to quantify the relationship between sample sort time and number of specimens, average body length of specimens, dry weight of the less than 2.0-mm diameter residue, and dry weight of the greater than 2.0-mm diameter residue. The dependent variable, sort time, did not meet linear model assumptions. Therefore, a BoxCox test was performed using R version 3.0.3© to find the optimal inverse square root power transformation on sort time to normalize the data (Box and Cox 1964).
weight of the greater than 2.0-mm diameter residue was 0.65 g per sample. The mean generic body length of individuals was 4.14 mm, and the mean number of specimens per SFPE sample was 62 (Table 2). Sort time was significantly influenced by residue that was less than 2.0 mm in diameter (p≤0.001) and number of specimens (p=0.01), but not by body length (>0.70) or residue greater than 2.0 mm in diameter (>0.10) (multiple regression, R2adj =0.55, p≤0.001) (Table 3). Specifically, there was a significant relationship between inverse square root of sort time and residue that was less than 2.0 mm in diameter (Fig. 2) and a significant relationship between inverse square root of sort time and the number of specimens (Fig. 3). Therefore, the sample sort time increased as functions of both the dry weight of residue less than 2.0 mm in diameter and the number of specimen (Table 3).
Discussion
Results The minimum time required to sort a sample was 15 min and the maximum time required was 320 min, with 83.9 % of the samples being sorted in 100 min or less. For samples that had less than 100 specimens, the average sample had 25.2 specimens, which were sorted in an average of 69.3 or 2.7 min per specimen. For samples that had between 101 and 300 specimens, the average sample had 183.9 specimens, which were sorted in an average of 85.6 or 0.5 min per specimen. For samples that had between 301 and 500 specimens, the average sample had 500 specimens, which were sorted in an average of 153.5 or 0.3 min per specimen (Table 1). On average, the small samples were nine times more time consuming per specimen than large samples. The mean dry weight of the less than 2.0-mm diameter residue was 2.38 g per sample, while the mean dry
Table 1 Mean SFPE sample sort time and mean number of specimens per sample for various subsample groupings Subsample groupings
N
T (S.E.)
Up to 100 specimens
47 69.3 (8.5)
n
T =n 25.2 2.7
Between 101 to 300 specimens
7 85.6 (18.7)
183.9 0.5
Between 301 to 500 specimens
2 153.5 (23.5) 500.0 0.3
N number of samples, T mean SFPE sample sort time (in minutes), n mean number of specimens per sample
Coastal tropical watersheds are among the most dynamic aquatic ecosystems on earth, with enormous spatial and temporal complexity. They are influenced by continental runoff waters, which are often rich in nutrients derived from urban, agricultural, and industrial activities (Pringle et al. 2000). Given the limited resources available for biological monitoring programs of these tropical aquatic systems (Ramírez et al. 2008), it is important to utilize protocols that efficiently determine patterns of chironomid community composition. Our results support other studies showing that Chironomidae SFPE sample sorting has been shown to be more time efficient than traditional dip-net techniques for assessing chironomid community composition (Anderson and Ferrington 2012; Ferrington et al. 1991). Most samples could be sorted in less than 100 min, but on average, it
Table 2 Mean and range of variables measured in SFPE samples Variable
N
Mean (SE)
Range
Residue less than 2.0-mm diameter (g) Residue greater than 2.0-mm diameter (g) Body length (mm)
56
2.38 (0.23)
0.09–5.36
56
0.65 (0.09)
0.06–4.60
44
4.14 (0.18)
1.50–6.67
Number of specimens
56
62 (14)
0–500
N number of samples
70
Environ Monit Assess (2015) 187:70
Page 6 of 8
Table 3 Multiple linear regression, considering sort time as the dependent variable Variable
Partial correlation
Standard error
p value
Residue less than 2.0-mm diameter (g) Residue greater than 2.0-mm diameter (g) Body length (mm)
−0.01325
0.00231
0.10
0.00193
0.00336
>0.70
Number of specimens
−0.00009
0.00004
0.01
took 85.6 min to sort a subsample of 300 specimens. Tropical coastal water SFPE sample sort times were less time efficient than sort times of temperate urban streams. The average time needed to sort a subsample of 300 specimens was approximately twice as much for tropical coastal water samples compared to temperate urban streams (Anderson and Ferrington 2012; Ferrington et al. 1991). The variable contributing most significantly to SFPE sample sort times was the amount of small-sized residue. During field collections, larger organic and inorganic materials were suspended in the water column from the benthic substrates via mechanical disturbances (e.g., wave action) and smaller materials were floating on the surface of the water. Therefore, organic material (e.g., pieces of leaves, flowers, sticks, terrestrial, and aquatic invertebrates) and inorganic material (e.g., plastic, styrofoam, silt, and sand) were incidentally transferred to the sieve and included in the SFPE sample (P. Kranzfelder, personal observations). As a result,
Fig. 2 Relationship between the inverse square root of sort time (in minutes) and dry weight of residue less than 2.0 mm in diameter (in grams)
Fig. 3 Relationship between the inverse square root of sort time (in minutes) and number of specimens
bioassessment programs sampling chironomids in aquatic systems with high total suspended solids should plan for increased SFPE sample sort time. For example, our results suggest that a SFPE sample with 5.0 grams of small-sized residue should require three times as long to sort as a sample with 0.5 g of small-sized residue. While it is impossible to completely eliminate sample residue, caution should be taken during field collection to reduce the amount of mechanical disturbance created from walking on bottom substrates and wave action from boat wakes. While the number of specimens in the sample contributed significantly to the sort time of individual samples, very small samples (i.e., less than 100 specimens) were much more time consuming to sort per specimen than medium samples (i.e., between 101 and 300 specimens) and large samples (i.e., between 301 and 500 specimens). Therefore, our results, similar to Anderson and Ferrington (2012), suggest that samples with larger numbers of exuviae are more efficient, and thus, sampling effort in the field should be adjusted to ensure that a sample contains a large number of pupal exuviae. In addition to increased lab processing efficiency, Ruse (2011) suggests that by increasing the number of pupal exuviae identified from a single sample, there is an increase in probability of including Brare^ species that are not close to their peak emergence period. Increased sample size can be achieved by extending the time spent sampling from 10 to 20 min per site and/or increasing the length of the sampling reach from 100 to 200 m for stream sites and from 500 to 1,000 m for riverine and estuarine sites. While sort times from coastal tropical waters were more time consuming than samples from temperate
Environ Monit Assess (2015) 187:70
urban streams, it appears that alternative sampling methods, such as the traditional dip-net sampling, would be relatively more time consuming. Both Ferrington et al. (1991) and Anderson and Ferrington (2012) found that sample sorting of SFPE samples took about 70 % less time compared to dip-net samples. Stein et al. (2008) found that sort times took an average of 4–7 h for benthic macroinvertebrate dip-net samples collected from the nearby Dos Novillos River on the Caribbean coast of Costa Rica. They suggested that the long sort time was due to separating the invertebrates from organic and inorganic material in the sample (Stein et al. 2008). The SFPE samples processed for this study contained 98 species of chironomids, which are distributed among 44 genera. Dip-net samples were taken from the same sample sites in Laguna del Tortuguero at a later date as part of another study (Kranzfelder, in preparation). The results from these dip-net samples are similar to the findings of Mora et al. (2003), with six orders of aquatic or semiaquatic insects detected. Compared to Mora et al. (2003), additional Hemiptera genera (six versus five), Ephemeroptera genera (two versus one), Coleoptera genera (three versus one), and Chironomidae (eight genera, none reported by Mora et al. (2003)) were detected, resulting in a total of 22 genera now known for Laguna del Tortuguero. However, our dip-net samples only detected four genera of EPT, demonstrating the potential ineffectiveness of metrics based on these three orders for water quality studies in this estuarine system. By contrast, the 98 species of Chironomidae detected using the SFPE method are relatively diverse at genus level (44 genera) and this method should be considered as a preferred replacement to dip-net sampling for biomonitoring in this estuarine environment. The results of this study will enhance planning and budgeting for future field sampling designs by providing guidelines for biomonitoring programs using the SFPE method. Acknowledgments We thank Costa Rica’s Ministerio del Ambiente y Energia for kindly permitting access to the study area and for providing research permit ACTo-GASP-PIN-016-010. We also wish to thank Jessica Miller and Catherine DeGuire for field and laboratory assistance and the Canadian Organization for Tropical Education and Rainforest Conservation for providing precipitation data. Funding for this project was provided by the Department of Entomology in the College of Foods Agricultural and Natural Resource Sciences. This paper is published under the auspices of the Chironomidae Research Group at the University of Minnesota.
Page 7 of 8 70
References Anderson, A. M., & Ferrington, L. C. (2012). Time-efficiency of sorting Chironomidae surface-floating pupal exuviae samples from urban trout streams in Northeast Minnesota, USA. In E. E. Stur & K. Aagaard (Eds.), 18th International Symposium on Chironomidae. Trondheim, Norway: Fauna norvegica. Armitage, P. D., Cranston, P. S., & Pinder, L. C. V. (1995). The Chironomidae: biology and ecology of non-biting midges: Chapman & Hall. Ashe, P., Murray, D. A., & Reiss, F. (1987). The zoogeographical distribution of Chironomidae (Insecta: Diptera). Annales de Limnologie, 23(1), 27–60. 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. (2nd ed., Vol. EPA 841-B-99-002) (2nd ed.). Washington, D.C.: U.S. Environmental Protection Agency, Office of Water. Bouchard, R. W., & Ferrington, L. C. (2011). The effects of subsampling and sampling frequency on the use of surfacefloating pupal exuviae to measure Chironomidae (Diptera) communities in wadeable temperate streams. Environmental Monitoring and Assessment, 181(1–4), 205–223. Box, G. E. P., & Cox, D. R. (1964). An analysis of transformations. Journal of the Royal Statistical Society. Series B (Methodological), 26(2), 211–252. Coffman, W. P. (1973). Energy flow in a woodland stream ecosystem: II. The taxonomic composition of the Chironomidae as determined by the collection of pupal exuviae. Archiv fur Hydrobiologie, 71, 281–322. Ferrington, L. C. (2008). Global diversity of non-biting midges (Chironomidae; Insecta-Diptera) in freshwater. In E. V. Balian, C. Lévêque, H. Segers, & K. Martens (Eds.), Freshwater animal diversity assessment, Developments in Hydrobiology, Vol. 198 (pp. 447–455). Netherlands: Springer Netherlands. Ferrington, L. C., Blackwood, M. A., Wright, C. A., Crisp, N. H., Kavanaugh, J. L., & Schmidt, F. J. (1991). A protocol for using surface-floating pupal exuviae of Chironomidae for rapid bioassessment of changing water-quality. In N. E. Peters & D. E. Walling (Eds.), Sediment and stream water quality in a changing environment: trends and explanation. Oxfordshire, U.K: International Association of Hydrological Sciences Press. Ferrington, L. C., Berg, M. B., & Coffman, W. P. (2008). Chironomidae. In R. W. Merritt, K. W. Cummins, & M. B. Berg (Eds.), An introduction to the aquatic insects of North America (4th ed., pp. 847–989). Dubuque, IA: Kendall/Hunt Publishing Company. Hirth, H. F. (1963). The ecology of two lizards on a tropical beach. Ecological Monographs, 33, 83–112. Jones, F. C. (2008). Taxonomic sufficiency: the influence of taxonomic resolution on freshwater bioassessments using benthic macroinvertebrates. Environmental Reviews, 16, 45–69. Jongman, R. H. G., Ter Braak, C. J. F., & van Tongeren, O. F. R. (1995). Data analysis in community and landscape ecology. Cambridge, UK: Cambridge University Press.
70
Page 8 of 8
Kranzfelder, P. (in revision). Temporal and spatial variability of Chironomidae (Insecta: Diptera) species emergence in a Neotropical estuary. Freshwater Science. Kranzfelder, P. (in preparation). Cumulative effects of coastal watershed land use on Chironomidae (Insecta: Diptera) communities of Neotropical estuaries in Costa Rica. University of Minnesota, Minneapolis, MN. Kranzfelder, P., Anderson, A. M., Egan, A. T., Mazack, J. E., Bouchard, R. W., & Ferrington, L. C. (in press). Use of Chironomidae (Diptera) surface-floating pupal exuviae as a rapid bioasessment protocol for water bodies. Journal of Visualized Experiments (JoVE). Lenat, D. R. (1988). Water quality assessment of streams using a qualitative collection method for benthic macroinvertebrates. Journal of the North American Benthological Society, 7(3), 222–233. Mora, J. M., Rodríguez, M. A., & López, L. I. (2003). Sondeo ecológico rápido y monitoreo de especies indicadoras en el Parque Nacional Tortuguero. In INBIO (Ed.). Guapiles, Costa Rica. Nordlie, F. G., & Kelso, D. P. (1975). Trophic relationships in a tropical estuary. Revista de Biología Tropical, 23, 77–99. Nuhn, H., Pérez, S. R., & Sandner, G. (1967). Estudio geográfico regional de la zona atlántico norte de Costa Rica. In I. d. T. y. Colonización (Ed.). San José, Costa Rica: Instituto de Tierras y Colonización. Pinder, L. C. V. (1986). Biology of freshwater Chironomidae. Annual Review of Entomology, 31, 1–23. Pringle, C. M., & Hamazaki, T. (1998). The role of omnivory in a Neotropical stream: separating diurnal and nocturnal effects. Ecology, 79(1), 269–280. Pringle, C. M., Scatena, F. N., Paaby-Hansen, P., & Nunez-Ferrera, M. (2000). River conservation in Latin America and the Caribbean. In P. J. Boon, B. Davies, & G. E. Petts (Eds.), Global perspectives on river conservation: science, policy and practice (pp. 41–77). Durham, NC: John Wiley and Sons.
Environ Monit Assess (2015) 187:70 Ramirez, A., & Pringle, C. M. (1998). Invertebrate drift and benthic community dynamics in a lowland Neotropical stream, Costa Rica. Hydrobiologia, 386, 19–26. Ramírez, A., Pringle, C. M., & Wantzen, K. M. (2008). Tropical stream conservation. In D. Dudgeon (Ed.), Tropical stream ecology (pp. 285–304). London: Academic Press. Rinella, D. J., & Feminella, J. W. (2005). Comparison of benthic macroinvertebrates colonizing sand, wood, and artificial substrates in a low-gradient stream. Journal of Freshwater Ecology, 20(2), 209–220. Rosemond, A. D., Pringle, C. M., & Ramírez, A. (1998). Macroconsumer effects on insect detritivores and detritus processing in a tropical stream. Freshwater Biology, 39(3), 515–523. Rosenberg, D. (1992). Freshwater biomonitoring and Chironomidae. Netherland Journal of Aquatic Ecology, 26(2–4), 101–122. Ruse, L. (2011). Lake acidification assessed using chironomid pupal exuviae. Fundamental and Applied Limnology, 178(4), 267–286. Stein, H., Springer, M., & Kohlmann, B. (2008). Comparison of two sampling methods for biomonitoring using aquatic macroinvertebrates in the Dos Novillos River, Costa Rica. Ecological Engineering, 34(4), 267–275. Wiederholm, T. (1986). Chironomidae of the Holarctic region. Part 2. Pupae (Vol. 28): Entomologica Scandinavica Supplement. Wilson, R. S., & McGill, J. D. (1977). A new method of monitoring water quality in a stream receiving sewage effluent, using chironomid pupal exuviae. Water Research, 11(11), 959– 962. Winemiller, K. O., & Leslie, M. A. (1992). Fish assemblages across a complex, tropical freshwater/marine ecotone. Environmental Biology of Fishes, 34(1), 29–50. Winemiller, K. O., & Ponwith, B. J. (1998). Comparative ecology of eleotrid fishes in Central American coastal streams. Environmental Biology of Fishes, 53, 373–384.
