Amazon Benthic Report April 2009

Assessment of Global Vision International’s Impact on Water Quality in the Bosque Protector Yachana Preliminary Report 12 April 2009

Authors Jonathan Escolar, Andrew Mercer, Hannah Urpeth

Table of Contents

Summary……………………………………………………………………………………………………………………………….ii 1.0

Introduction……………………………………………………………………………………………………………….1

2.0

Methods…………………………………………………………………………………………………………………….2 2.1

Site selection…………………………………………………………………………………………………….....2

2.2

Collection………………………………………………………………………………………………………………2

2.3

Identification…………………………………………………………………………………………………………2

2.4

Analyses………………………………………………………………………………………………………………..3 2.4.1 EPT Index…………………………………………………………………………………………….....3 2.4.2 Sensitivity Index………………………………………………………………………………………3

3.0

Results……………………………………………………………………………………………………………………….4 3.1

Upper Pump Stream……………………………………………………………………………………………..4

3.2

Lower Pump Stream……………………………………………………………………………………………..4

3.3

EPT Combined Results…………………………………………………………………………………………..4

3.4

Sensitivity Combined Results…………………………………………………………………………………5

4.0

Discussion…………………………………………………………………………………………………………………..6

5.0

Conclusions and Recommendations……………………………………………………………………………7

References…………………………………………………………………………………………………………………………….7 Acknowledgements……………………………………………………………………………………………………………….8 Appendix I……………………………………………………………………………………………………………………………..8

Summary This document is a preliminary report investigating the water quality of a stream directly impacted by GVI’s presence in the Bosque Protector Yachana using benthic macro invertebrates as indicators. The results show that water quality upstream is marginally better than that downstream, suggesting that the effluents released by GVI have negatively affected water quality. However, two different analyses gave different results, causing uncertainty as to which gives a true reflection of water quality. Further investigation and consultation with local experts has revealed that the BMWP/Col index is a more reliable way of determining water quality in this region. Recommendations for the future direction of this investigation include: 1. Resample the same sites using the BMWP/Col index. 2. Use kick nets in combination with Surber nets to sample areas of deeper water in streams as well as riffle sites. 3. Take more samples to obtain a large enough dataset to test statistically. 4. Sample the same areas long term to account for seasonal variation.

ii

1.0

Introduction

GVI has been using a research station based in the Bosque Protector Yachana in Napo Province, Ecuador for approximately three years. The infrastructure of the site unfortunately lacks a proper grey water system, and instead waste water from sinks and showers is drained into a large wooden-sided subterranean tank where it gradually seeps into an adjacent waterway. Chemicals used on the site and therefore released into the ecosystem include bleach, detergent, DEET, laundry soap and other personal hygiene products. Fecal matter from toilets is contained in a similar tank and therefore must also slowly leach into the surrounding soil and waterways. Figure 1.0 shows the layout of GVI base camp in relation to the impacted waterways.

Figure 1.0 Map showing position of grey water and sewage tanks in relation to Pump Stream.

Two study sites were investigated that represented two different treatments: One pre-discharge (sample site 2), that is not impacted by wastewater from GVI Base Camp and a second post-discharge (sample site 1), located downstream from GVI and therefore exposed to any effluents. It is therefore expected that sample site two, the upper Pump Stream, will have higher water quality as it is not impacted by effluents discharged by GVI Base Camp. Sampling benthic invertebrate communities is a reliable and economical way of determining water quality (Feinsinger, 2001). Each invertebrate family has differing sensitivities to contaminants, and their presence or absence as well as abundance can be used in different analytical indices to give an 1

indication of water quality. Over 50 different methods have been developed for the biological assessment of water quality in temperate countries (Cota et. al., 2002) and some have been adapted for use in tropical regions and their associated biota. The analyses selected for this investigation are discussed fully in subsequent sections.

2.0

Methods

2.1

Site selection

Two study sites were selected, each representing a different treatment. One site was located before waste water from the base camp entered the water system (upper Pump stream) and the other site selected in an area downstream to where waste water was discharged (lower Pump stream). Study sites were selected based on the presence of areas of fast flowing shallow water over rocky substrate known as riffles. Valid study sites contained riffles of both suitable size and abundance to allow for the collection of 30 samples to be taken from the selected study area. 2.2

Collection

Samples were collected by employing a modified kick sampling technique (Sutherland, 1996) with the use of a Surber net (300mm x 300mm). The Surber net was placed upon the substrate of identified riffles with the net positioned downstream, allowing for the collection of dislodged individuals. The area of each sample was defined by the frame of the Surber net resting on the substrate, and all loose stones within it were hand scrubbed before being placed outside of the sample area and the remaining substrate disturbed thoroughly by hand to a depth of one inch. After the sample was completed any removed stones were placed back in their original position so as to minimize disturbance. After each sample, the contents of the Surber net were emptied into a large bucket with the net being thoroughly flushed with stream water and then visually checked for remaining specimens. Collected materials from the sample were divided into trays and searched for specimens with any individuals found being removed with tweezers and placed in a killing jar containing 70% alcohol. A separate killing jar was used for case-crafting Trichoptera to aid in the identification process. The above process was repeated until 15 samples from each survey site were collected. A further 15 samples were taken at later dates from both the upper and lower Pump Stream. 2.3

Identification

Collected specimens were taken back to the field base and using a taxonomic key (Reyes & Peralbo, 2001) identified to family level before being tallied. Identification was performed using 10x hand lenses.

2

2.4

Analyses

Tallied results were then used in conjunction with both Ephemeroptera-Plecoptera-Trichoptera (EPT) index and an individual sensitivity index (Reyes & Peralbo, 2001) to determine water quality for the selected study areas. 2.4.1

EPT Index

The EPT index measures the percentage of individuals from the orders Ephemeroptera, Plecoptera and Trichoptera against the total number of individuals in each sample. These three orders are used as they are particularly sensitive to changes in water quality. The higher the percentage on this index, the higher the quality of the water. Table 2.0 shows how the scores relate to water quality.

EPT Total

Water quality

75 - 100% 50 - 74% 25 - 49% 0 - 24%

Very Good Good Moderate Poor

Table 2.0 EPT Index scores.

2.4.2

Sensitivity Index

This index works on the presence or absence of different orders, sub orders or families that each have a different degree of sensitivity to contaminants. Each group is assigned a value of 1 - 10, the groups given a rating of 10 being the most sensitive and the ones rated 1, least sensitive. The score from the groups present is added up and higher scores indicate a higher quality of water. Table 2.1 shows the values given to each group.

Sensitive Species Total

Water quality

101 - 145+ 61 – 100 36 – 60 16 – 35 0 – 15

Very Good Good Moderate Poor Very Poor

Table 2.1 Sensitivity Index scores.

3

3.0

Results

3.1

Upper Pump Stream

The upper Pump Stream was sampled on two occasions, 25.02.2009 and the 04.03.2009. Each time, 15 Surber net samples were collected from different locations. Table 3.1 shows the results for water quality from each sample day and a mean value for both on the EPT and Sensitivity indices. In all three the water quality is rated as moderate by the EPT index and very good on the sensitivity index.

Sample 25.02.2009 04.03.2009 Mean

EPT index score 43.35 34.81 39.08

Sensitivity index score 130 109 119.5

Table 3.1 Water quality results for upper Pump Stream on EPT and Sensitivity indices.

3.2

Lower Pump Stream

The lower Pump Stream was also sampled on two occasions, 23.02.2009 and the 17.03.2009. Again, 15 Surber net samples were taken from individual locations on each sampling date. Table 3.2 shows the results for water quality from each sample day and a mean value for both indices. According to the EPT index water quality is poor in the sample taken on 23.02.2009, and moderate on both the 17.13.2009 and on average. On the Sensitivity index it is rated as very good for all three samples.

Sample 23.02.2009 17.03.2009 Mean

EPT index score 22.95 29.35 26.15

Sensitivity index score 123 143 133

Table 3.2 Water quality results for lower Pump Stream on EPT and Sensitivity indices.

3.3

EPT Combined Results

Results from the EPT index support the null hypothesis that water quality is higher in the upper Pump Stream. Figure 3.1 shows the EPT scores from the upper and lower Pump Stream on both sample dates and a mean score of the two.

4

Figure 3.1 EPT scores for upper and lower Pump Stream on each sample date and mean score.

3.4

Sensitivity Combined Results

The results from the Sensitivity index analysis do not support the null hypothesis. Instead they indicate that on average water quality is marginally higher in the lower Pump Stream. Figure 3.2 shows the Sensitivity index scores from the upper and lower Pump Stream on both sample dates and their mean.

Table 3.2 Sensitivity index scores for upper and lower Pump Stream on each sample date and mean score.

5

4.0

Discussion

The major surprise presented by the results of this study is that the EPT index and Sensitivity index gave contradicting results for all the samples collected. The values obtained from the EPT index fit the null hypothesis that water quality is higher on the upper Pump Stream, although the sample size is too small to test for a significant difference statistically. With the Sensitivity index however, water quality was on average higher in the lower Pump Stream, although again, the sample size was not large enough to test for significant differences. The reason for this disparity is that the analyses used are unsuitable for the Neotropical region, despite the fact that these were taken from a source designed for use in Ecuador by Ecuadorians (Reyes & Peralbo, 2001). The use of the EPT index gives unreliable results because the number of families of stoneflies (Plecoptera) drops off to just one, Perlidae, the closer one is to the equator (Feinsinger, 2001). After further research and consultation a more suitable analysis was found for the Neotropics, the BMWP/Col index (Contreras et. al.,2008; Zarate, pers. comm., 2009) that uses more relevant families as sensitivity indicators. The use of the Sensitivity index is also questionable because unlike the EPT index it does not take abundance into account. This could affect the accuracy of results as some samples yield substantially more individuals than others. For example, the 17.03.2009 sample consisted of 620 individual invertebrates, whereas the other three samples had a mean of 173 individuals. It should be noted however, that the BMWP/Col index also ignores abundance and works solely on presence or absence of families. The identification key provided by Reyes and Peralbo (2001) also limited the accuracy of the study as a large proportion of the invertebrates collected could only be identified as ‘other’. A new key (Contrera, et. al., 2008) that enables identification of a greater number of families will be used for future studies. The use of the Surber net to sample riffles was effective, although other studies have found kick nets to yield better and more cost-effective results (Buss & Borges, 2008). A combination of the two methods would enable sampling of both riffles and pools and may give a more complete sample of benthic invertebrate assemblages. Some other physical factors that may affect the lower Pump Stream and potentially skew results have been noted. First, a landslide has blocked the flow of the stream in one area creating a large, still body of water. The flow then returns to normal but the 17.03.2009 sample was taken downstream from this site as there were not enough suitable riffles higher up the waterway. A road is also in close proximity to a stretch of the lower Pump Stream, and although the level of traffic is low, runoff could potentially have an impact on water quality. Another factor that must be addressed in future studies is non-use of weather data in correlation with these data. Precipitation levels can have a significant effect on benthic macro invertebrate communities, as many may be washed away after periods of heavy rainfall. There was particularly heavy rainfall prior to surveying the upper Pump Stream on the 04.03.2009 that may have washed away a significant number of the invertebrates present. 6

5.0

Conclusions and Recommendations

The main conclusion drawn from this investigation is that the analyses used were unreliable at giving an accurate reflection of water quality as they are unsuited to Neotropical freshwater habitats. The sample sites should be re-sampled and analysed using the BMWP/Col index and a more detailed identification key, and deeper water areas should be sampled using kick nets in addition to the sampling of riffles with Surber nets. Moreover, a greater number of samples should be taken to allow a statistical analysis of the samples to be carried out. Ideally this should be conducted at regular interval throughout the year to look for seasonal variations and these data should be correlated with weather data. As the water quality results are unreliable the authors are unable to make recommendations regarding changes to GVI practice and infrastructure.

References Buss D. F. & Borges E. L., 2008. ‘Application of Rapid Bioassessment Protocols (RBP) for Benthic Macroinvertebrates in Brazil: Comparison between Sampling Techniques and Mesh Sizes.’ Neotropical Entomology 37 (3): 288-295 Contreras J., Roldán G., Arango A. & Álvarez L.F., 2008. ‘Evaluación de la calidad del agua de las microcuencas La Laucha, La Lejía y La Rastrojera, utilizando los macroinvertebrados como bioindicadores, Municipio de Durania, Departamento Norte de Santander, Colombia.’ Rev. Acad. Colomb. Cienc. 32(123): 171-193 Cota, L., Goulart, M., Moreno, P. & Callisto, M. ‘Rapid assessment of river water quality using an adapted BMWP index: a practical tool to evaluate ecosystem health’. Verh. Internat. Verein. Limnol. 28: 1-4 Feinsinger, P. (2001) Designing Field Studies for Biodiversity Conservation. Island Press: Washington Reyes , C.C & Peralbo K.F. (2001) Manual de Monitoreo: Los Macroinvertebrados Acuaticos como Indicadores de la Calidad del Agua. EcoCiencia: Ecuador Sutherland, W.J. (1996) Ecological Census Techniques: A Handbook. Cambridge University Press: UK

7

Acknowledgements The authors would like to thank the following volunteers and staff who participated in the data collection during the expedition 091b at the Bosque Protector Yachana: Katherine Allison, Chris Beirne, Karina Berg, Sophie Cousins, Max Hardman, Amy Hill, Thomas Keating, Victoria Morgan-Hill, James Pitt, Alan Rea, Rachel Reisinger, Piter Silvera, Glen Skelton, Natalie White, Mauro Yumbo.

Appendix I This appendix contains the data sheets from each sample date. Ephemeroptera/Plecoptera/Trichoptera (EPT) Index & Sensitive Species Index Survey Site Date Name of Surveyor/s

Upper pump stream - 1st section 25/02/09 Katherine, Hannah, Jon, Max, James, Natalie, Amy

Classification Phylum / Class

Order

Family

Annelida

Hirudinea Oligochaeta

Aracnida Crustacea

Acari Decapoda

Insecta

Gammaridea Coleoptera

Glossiphonidae Annelidae Tubificidae Hydrachnidae Palaeomonidae Pseudothelpusidae Hyalellidae Chrysomelidae Elmidae (Adult) Elmidae (Larva) Gyrinidae Psephenidae Ptilodactylidae (Larva) Scarabidae Caratopogonidae Chironomidae Ephydridae Simuliidae Stratiomyidae Syrphidae Tipulidae Baetidae Euthyplocidae Leptohyphidae Leptophlebiidae Oligoneuridae Other Belostomidae Naucoridae Veliidae Pyralidae Corydalidae Anisoptera Gomphidae Libellulidae Zygoptera Calopterygidae Coenagrionidae

Diptera

Ephemeroptera

Hemiptera

Lepidoptera Neuroptera Odonata

Abundance

Sensitive Species

0 0

0 0

2

0 0

0 0

4 1

0 0

6 6

0

0

39

0

10

4

0 0

0 2

10

0

8

12

0 0 2 0 55 0 2

3 0 9 0 9 0 0

3 9

0 0 0 0 0

7 0 0 6 8

26

0

8

2 55 2 11

8

EPT Present

Plecoptera Trichoptera

Mollusca

Nematomorpha Platelminta Other (excluding EPT)

Bivalvia Gastropoda

Nematoda Tricladia

18

18 0

10 0

3

3

7

8 4

8 4

9 5

5

5

9

13

13

8

4

4 0

0 0

9 15

0 0 0 0

0 0 0 0

263

114

130

Perlidae Other Calamoceratidae Glossosomatidae Helicopsychidae Hydrobiosidae Hydropsychidae Lampyridae Leptoceridae Odontoceridae Philopotamidae Polycentropodidae Other Sphaeridae Hydrobiidae Physidae Planorbidae Other Gordioidea Planarida

4

Total EPT Total Sensitive Species Total

43.35 130

Ephemeroptera/Plecoptera/Trichoptera (EPT) Index & Sensitive Species Index Survey Site Date Name of Surveyor/s

Upper Pump Stream - Upstream from Access C 4/3/2009 Andy M, Chris, Glen, Victoria, Rachel, Max, Dan, Piter

Classification Phylum / Class

Order

Family

Annelida

Hirudinea Oligochaeta

Aracnida Crustacea

Acari Decapoda

Insecta

Gammaridea Coleoptera

Glossiphonidae Annelidae Tubificidae Hydrachnidae Palaeomonidae Pseudothelpusidae Hyalellidae Chrysomelidae Elmidae (Adult) Elmidae (Larva) Gyrinidae Psephenidae Ptilodactylidae (Larva) Scarabidae Caratopogonidae Chironomidae Ephydridae Simuliidae Stratiomyidae Syrphidae Tipulidae Baetidae Euthyplocidae Leptohyphidae Leptophlebiidae Oligoneuridae Other Belostomidae

Diptera

Ephemeroptera

Hemiptera

Abundance

Sensitive Species

0 0

0 0

2 5

0 0

10 0

3

0 0

6 0

0

0

17

0

10

1 11

0 0

3 2

8

0

8

8

0 0 4 0 17 0 5

3 0 9 0 9 0 0

4 17 5

9

EPT Present

Lepidoptera Neuroptera Odonata

Plecoptera Trichoptera

Mollusca

Bivalvia Gastropoda

Nematomorpha Platelminta Other (excluding EPT)

Nematoda Tricladia

Naucoridae Veliidae Pyralidae Corydalidae Anisoptera Gomphidae Libellulidae Zygoptera Calopterygidae Coenagrionidae Perlidae Other Calamoceratidae Glossosomatidae Helicopsychidae Hydrobiosidae Hydropsychidae Lampyridae Leptoceridae Odontoceridae Philopotamidae Polycentropodidae Other Sphaeridae Hydrobiidae Physidae Planorbidae Other Gordioidea Planarida

2 5

0 0 0 0 0

0 0 0 6 8

24

0

8

10

10 0

10 0

0

0

0 0

0 0

2

2

9

4

4

8

5

5 0

0 0

0 0 0 0

0 0 0 0

47

109

2

Total

135

EPT Total Sensitive Species Total

34.81 109

Ephemeroptera/Plecoptera/Trichoptera (EPT) Index & Sensitive Species Index Survey Site Date Name of Surveyor/s

Main Pump Stream Lower - 1st Sample 23/02/09 Hannah, Andy M, Katherine, Sophie, Alan, Natalie, Mauro

Classification Phylum / Class

Order

Family

Annelida

Hirudinea Oligochaeta

Aracnida Crustacea

Acari Decapoda

Insecta

Gammaridea Coleoptera

Glossiphonidae Annelidae Tubificidae Hydrachnidae Palaeomonidae Pseudothelpusidae Hyalellidae Chrysomelidae Elmidae (Adult) Elmidae (Larva) Gyrinidae Psephenidae Ptilodactylidae (Larva) Scarabidae Caratopogonidae Chironomidae Ephydridae Simuliidae

Diptera

Abundance

5 5

23

5

10

EPT Present

Sensitive Species

0 0

0 0

0 0

0 0

0 0

6 6

0

0

0

10

0 0

0 0

0

8

Ephemeroptera

Hemiptera

Lepidoptera Neuroptera Odonata

Plecoptera Trichoptera

Mollusca

Bivalvia Gastropoda

Nematomorpha Platelminta Other (excluding EPT)

Nematoda Tricladia

Stratiomyidae Syrphidae Tipulidae Baetidae Euthyplocidae Leptohyphidae Leptophlebiidae Oligoneuridae Other Belostomidae Naucoridae Veliidae Pyralidae Corydalidae Anisoptera Gomphidae Libellulidae Zygoptera Calopterygidae Coenagrionidae Perlidae Other Calamoceratidae Glossosomatidae Helicopsychidae Hydrobiosidae Hydropsychidae Lampyridae Leptoceridae Odontoceridae Philopotamidae Polycentropodidae Other Sphaeridae Hydrobiidae Physidae Planorbidae Other Gordioidea Planarida

5

0 0 5 0 3 0 5

3 0 9 0 9 0 0

1 8

0 0 0 0 0

7 0 0 6 8

15

0

8

6

6 0

10 0

3

3

7

1 1

1 1

9 5

2

2

9

0

0

2 0

0 0

1 3 3

0 0 0 0

0 3 0 0

122

28

123

5 3 5 20

2

Total EPT Total Sensitive Species Total

22.95 123

Ephemeroptera/Plecoptera/Trichoptera (EPT) Index & Sensitive Species Index Survey Site Date Name of Surveyor/s

Lower pump Stream (2nd Section - Road) 16/3/2009 Hannah, Jon, Karina, Sophie, James, Tom, Dan, Mauro, Victoria

Classification Phylum / Class

Order

Family

Hirudinea Oligochaeta

Glossiphonidae Annelidae Tubificidae Hydrachnidae Palaeomonidae Pseudothelpusidae Hyalellidae Chrysomelidae Elmidae (Adult)

Annelida

Aracnida Crustacea

Acari Decapoda

Insecta

Gammaridea Coleoptera

Abundance

EPT Present

11

0 0

0 1

2

0 0

0 0

7

0

6

11

Sensitive Species

Diptera

Ephemeroptera

Hemiptera

Lepidoptera Neuroptera Odonata

Plecoptera Trichoptera

Mollusca

Bivalvia Gastropoda

Nematomorpha Platelminta Other (excluding EPT)

Nematoda Tricladia

17

Elmidae (Larva) Gyrinidae Psephenidae Ptilodactylidae (Larva) Scarabidae Caratopogonidae Chironomidae Ephydridae Simuliidae Stratiomyidae Syrphidae Tipulidae Baetidae Euthyplocidae Leptohyphidae Leptophlebiidae Oligoneuridae Other Belostomidae Naucoridae Veliidae Pyralidae Corydalidae Anisoptera Gomphidae Libellulidae Zygoptera Calopterygidae Coenagrionidae Perlidae Other Calamoceratidae Glossosomatidae Helicopsychidae Hydrobiosidae Hydropsychidae Lampyridae Leptoceridae Odontoceridae Philopotamidae Polycentropodidae Other Sphaeridae Hydrobiidae Physidae Planorbidae Other Gordioidea Planarida

0

6

0

0

20

0

10

5 127

0 0

3 2

43

0

8

19

0 0 5 1 77 0 21

3 0 9 7 9 0 0

3 27

0 0 0 0 0

7 8 0 6 8

61

0

8

9

9 0

10 0

7

7

7

30

0 30

0 5

9

9

9

14

14

8

9

9 0

0 0

2 16 49

0 0 0 0

3 0 0 0

620

182

143

5 1 77 21 27 2

Total EPT Total Sensitive Species Total

29.35 143

12