KANDID A T UPPSA TS
Miljö och Hälsoskydd
The environmental and anthropogenic impact on freshwater biodiversity in Lajeado, RS, Brazil.
A study of using dragonflies as indicators for the environmental status in freshwater biomes
Philip Sjöwall och Robert Lundström
Miljövetenskap 15hp
Halmstad 2016-11-30
1
The environmental and anthropogenic impact on freshwater biodiversity in
Lajeado, RS, Brazil.
A study of using dragonflies as indicators for the environmental status in freshwater biomes
Robert Lundström & Philip Sjöwall Miljö- och Hälsoskydd
Högskolan i Halmstad
Handledare: Göran Sahlén Examinator: Stefan Weisner
2
Abstract
Fragmentation of the Atlantic rainforest and alteration of waters due to agricultural expansion has greatly affected the species diversity in Brazil. In this study, we investigate how different
environmental factors affect dragonfly communities and dragonfly species richness in sixteen different locations in Rio Grande do Sul, Brazil. A total of 328 individuals distributed among 46 species and eight families were collected. At each sampling location the water was analyzed and the surrounding environment recorded and plotted. Our goal was to investigate if we could find species for use as bio indicators on water quality and if the fragmentation of the forests in Rio Grande do Sul affects the species richness negatively. Our data suggested that the amount of forest, pH level and water temperature have a strong correlation to the number of species. For the number of specimens we found that amount of forest, pH level, water temperature, conductivity, amount of dissolved solids in the water and amount of surrounding urban area affects the
population. We found that still water holds more species as well as specimens than current water.
The species T binotata, M Ocellata, Oxyagrion Sp., L pictus, M stawiarskii, R planaltica, L auritus and L dichrostigma could possibly be used as indicators for pH level, and that S reticulata, L bipupillatus, B furcata, R bonariensis, D mincki and T cophysa possibly could be used as indicators for conductivity. Further studies has to be done in order to be certain about the use of these species as indicators.
Keywords: Fragmentation, Atlantic rainforest, bio indicators, dragonfly, water quality.
3
Introduction
Since the industrial revolution in the late 18th century where the agricultural society changed into a mechanical mass-producing society (Kubesh, Mcneil & Bellotto 2008), the world population has grown rapidly. In the year 1850 there were approximately 1.2 billion people in the world, which is estimated to grow from 7.4 billion people today (Worldometers, 2016) into 9.2 billion people by 2050 (Mayhew 2015). The industrial revolution made people move to urban areas for work at the factories which led to a decrease in farmers, but the entry of agricultural inventions such as the seed drill made it possible to farm with less people on larger farms in order to feed the growing population (Kubesh, Mcneil & Bellotto 2008). Intense utilization of the land led to nutrient losses in the soil up until the age of mineral fertilizers in the 19th century, where
minerals were used to replace the removed nutrients (Isermann 1990). This new found technique resulted in higher yields and nutrient removal, but also led to an increased nutrient level in the soil due to the input of mineral fertilizers (Isermann 1990).
After the industrial revolution there was an increase of substances to the air, water and soil due to human activities (Cowling 1990). Increased sulfur dioxide and nitrogen oxide emissions to the air along with acid deposition led to acidification of lakes, streams and ground waters. This
acidification led to death and reproductive failure in some freshwater fishes and also decreased the fertility of soils (Cowling 1990). The availability of a number of nutrients in soil are reduced along with pH reduction, and phosphorus is less available to the soil when pH is below 5.5 (Carly et. al. 2009). Carly et al. (2009) also presented evidence that support the hypothesis that soil acidification is the dominant process responsible for a decline in species richness with increasing N deposition. Increased nutrient levels to the water leads to an increase of organic matter. This leads to a growth in primary production and oxygen-consuming drift-algae and therefore less oxygen available in the water for other organisms. This has led to a transformation in the biodiversity in some fresh-water systems in the Baltic and Brazil (Bonsdorff 1997; Eskinazi- Sant’Anna et al. 2013). The change in biodiversity due to the eutrophication of water counts as one of the major conservation problems the world faces today, and that is one of the reasons why the United Nations signed the Convention on Biological Diversity: The Aichi Biodiversity Targets
4
in Japan 2010 as the basis for halting and eventually reversing the loss of biodiversity of the planet (Convention on Biological Diversity, 2010).
Size of agriculture
Agricultural land usage has grown fast since the Industrial revolution due to new technology. The total land that were used for agriculture in 1700 was 265 million ha, which has grown with high pace to 1.2 billion ha in 1950. Today, over 1.5 billion ha in the world consist of agricultural land (Groom & Vynne 2006. 184-186). Brazil has since the modernization and industrialization of agriculture become one of the world's major producers of food, and agriculture make up 28% of the country's national exports (Ferreira et. al. 2012). While the economic impact of this has been greatly beneficial for the country and its human population, the biodiversity in the country has not been given the same benefits. The expansion in agriculture has led to that one third, 264 million ha, of the land in Brazil is currently being used for agricultural purpose with the area growing every day (Martinelli et. al. 2010). Most of the new farmland is made available by cutting down rainforest, which now makes up 330 million ha of Brazil (The Nature
Conservancy). A majority of this, 200 million ha, consist of pasture. The soybean is the crop that make up the biggest amount of the land used for crops, followed by corn and sugarcane. Other crops include beans and rice (Martinelli et. al. 2010). To make up new farmland, forests are cut down at one of the fastest paces in human era (Martinelli et. al. 2010) leading to deforestation, fragmentation and loss of habitats. The fragmentation of forest landscapes has greatly reduced the population of many forest species (Tabarelli, Mantovani & Peres 1990).
Atlantic rainforest
Brazil is the home of a lot of different species, many of which have vulnerable status. It holds about one seventh of the total number of species in the world (Agostinho 2005) and you can find threatened aquatic invertebrates in every state. The Atlantic rainforest stretches down across the Atlantic coast from Rio Grande do Norte (latitude 3°S) down to Rio Grande do Sul (latitude 30°S) (Tabarelli et al. 2010) and should not be confused with the two other Brazilian forest ecosystems, the Amazonas and Andean, since it is isolated from them. The Atlantic rainforest is separated from the Amazonas by the Cerrado and Catinga regions, and separated from the Andean by the Gran Chaco region (Silva & Casteleti 2003). The Atlantic rainforest reaches
5
further in from the coast with increasing latitude and stretches all the way into Paraguay and Argentina (Tabarelli et al. 2010). It consists of several different forest types depending on soil type, distance to the coast etc.The Atlantic rainforest used to cover 1,315,460 km² and were present in over seventeen different Brazilian states (SOS Mata Atlântica 2014). The Atlantic rainforest was cut down by the European settlers, who came to Brazil in the 16th century (Bethell 1984), to make space for cropland (Silva & Casteleti 2003). Because of human activities, the remaining native forest is divided into 245,173 different forest fragments where 83,4% of them have a surface area of less than 0.5 km² (Ribeiro et al. 2009). The biodiversity in the Atlantic rainforest is very high as it holds 1-8% of the world's species and thousands of endemic plants and animals (Ribeiro et al. 2009; Tabarelli et al. 2010). The survival of these species are in jeopardy due to the loss of and fragmentation of their habitats.
Fragmentation
Fragmentation of the landscape has affected the species composition that once were in the
original environments and have led to many species becoming threatened or extinct due to loss of habitat (Renner et al. 2015; Murcia 1995). Studies on mammals and birds in fragmented forests show that the more isolated fragment, the poorer it was in species (Vieira 2009) but that it contains rare and endangered species and therefore has a high conservation value (Chiarello 2000).
Parallels could be drawn from fragmentation to MacArthur and Wilson’s (1967) theory of island biogeography and the species-area curve to further show the effects of fragmentation. According to the theory of island biogeography the number of species inhabiting an island is determined by the ratio of extinction and colonization. The rate of colonization is determined by the islands distance from a source of colonist, usually the mainland or other islands and the size of the island (Brown 1978). The term “island” does not only apply to land surrounded by water, but also includes other isolated landscapes such as lakes, mountain peaks and natural areas surrounded by anthropogenically altered landscapes (Noss et al. 2006). The rate of extinction is determined by the species-area curve. This theory tells us that smaller islands and islands further away from the mainland population generally have smaller populations than bigger island near mainland
population and therefore the per species likelihood of extinction is greater on small islands than
6
on bigger islands (Losos & Ricklefs 2009). This is also applied to the fragmentation of the Atlantic rainforest. The initial problem with fragmentation is that the species endemic to the disrupted or destroyed environment are eliminated unless they can rapidly move to a similar environment or adapt to a new one (Noss, Csuti & Groom 2006). Remaining habitats are initially over flooded with individuals fleeing from affected areas. This increase in individuals is known as crowding and the number of animals later drops to equilibrium as the island can't produce enough energy to ensure every inhabitants survival. This is often later followed by a loss of species (Debisnksi & Holt 2000).
Living in isolated area comes with several disadvantages. Certain species may breed in certain environments but need feed in others (Noss, Csuti & Groom 2006). The risk of inbreeding is also a lot higher in fragmented areas, both for animals and plants who may also lose its pollinators (Keller & Waller 2002). Invasive species or changes to the top-down dynamics due to
fragmentation is a risk and can lead to cascade effects of the ecosystem (Noss, Csuti & Groom 2006). The classic example of these cascade effects is how extensive hunting of the sea otters on the North American Pacific coast led to their main food source, sea urchins, to reproduce
unchecked by their main predator. The uncontrolled sea urchin population then overexploited the kelp forest of which they feed, completely laying waste to the ecosystem (Estes, Duggins,
Rathbun. 1989).
Agriculture and anthropogenic impact on freshwater
Deforestation also leads to an increase in sedimentation of nearby aquatic ecosystems (Barbosa de Oliveira-Junior et al. 2015). The freshwater systems in these areas are also greatly affected when channels are dug and water flows are regulated to supply humans, cattle and crops (Ward 1998). Much like the fragmentation of forests has reduced the population of forest species, the anthropogenic regulation of the freshwater systems has greatly reduced the freshwater species (Sala et al. 2000). Mineral fertilizers and pesticides together with cattle producing highly nitrogen containing manure, has greatly affected water quality in freshwater systems. Channelizing also heavily alter the water quality in terms of nutrients, turbidity, salinity, oxygen etc. (Silk & Ciruna 2005). Generalist species can usually adapt quite well to these changes while specialists who are not able to adapt will most likely perish (Silk & Ciruna 2005) leading to a loss of biodiversity.
7
Problems with intense agriculture in Brazil
Half of the wood- and grassland, mainly Atlantic rain forest, in Cerrado region, located in central Brazil, has been transformed into cash crop fields (such as soybeans) due to the agricultural land expansion (Groom & Vynne 2006). Animal-based agriculture has also grown rapidly in Brazil, primarily in the southern states like Rio Grande do Sul and Santa Catarina where the country’s major proportion of swine and poultry production is located (Shigaki 2006). Animal agriculture has not only grown in the last twenty years, it has also changed. Conventional types of farms with a combination of cattle and crops are being replaced with intensive livestock farms, which leaks out more nutrients than conventional farms (Theobald, Daedlow & Kern 2015). One big problem with the manure from animal agriculture is that the ratio of Nitrogen (N) and Phosphorus (P) is not balanced to what the crop needs. The crops usually needs nutrients with a N to P ratio of 8:1 and the typical manure consist of a N to P ratio of 4:1 which results in an abundance of P in the soil (Shigaki 2006). This abundance of P will accumulate and spread to the water with the rain, resulting in eutrophication and an algal-bloom (Bouwman, Beusen & Billen 2009).
Eutrophication is a common threat when it comes to inland waters and its species in Brazil, but other factors like pollution, flood control and introduction of new species play a big part in the loss of biodiversity (Agostinho 2005).
Indicator species
Usage of species as an indicator for environmental changes is a commonly used method. One important feature with an indicator is that it shows clear responses to the environmental changes, which can be measured (Sahlén & Ekestubbe 2001). Insects are commonly used as indicators since they are present in almost any area and respond quickly to changes in their environment (Koch, Wagner & Sahlén 2013). Dragonflies (Odonata) works as good indicators since they are sensitive to human disturbance and have a high niche specificity in aquatic environments.
Theoretically, this means that disturbed areas would have a smaller odonate species variation than non-disturbed areas and a generally unhealthier ecosystem (Sahlén & Ekestubbe 2001;
Renner, Sahlén & Périco 2015). The majority of all dragonflies life cycle is spent in aquatic environments as their eggs are laid in water and the larvae live in water until metamorphosis
8
(Leipelt & Suhling 2005). Hence dragonflies are greatly affected by water quality and serve as good indicators for the quality of water (Smith, Samways & Taylor 2007).
As predators dragonflies are dependent on the prey population. Therefore by looking higher up in the food chain we can get an insight in changes of the entire food web (Koch, Wagner &
Sahlén 2013). Dragonflies are also easily sampled and identified. There are several studies published which has used dragonflies as indicators (Suhling et al. 2006; Sahlén & Ekestubbe 2001; Smith, Samways, Taylor 2007; Renner, Sahlén & Périco 2015) where the last one took place in southern Brazil. These studies shows that using dragonflies as indicators is a well-
established method, considering that they are one of the most visible indicators of the diversity in waters (Renner, Sahlén & Périco 2015).
Ecosystems of Rio Grande do Sul
Activities, such as agriculture, leading to deforestation in Rio Grande do Sul plays a big part in the creation of highly fragmented environments that now exist in the area (Tabarelli et al. 1990).
Since a lot of the landscape in Rio Grande do Sul has changed from a riverine landscape with grasslands and forests into a cultured land with ponds and canals due to the increased agriculture usage of the land (Ward 1998), the freshwater biodiversity in Rio Grande do Sul is threatened and therefore could have a high protection value. In these areas dragonflies serves as one of the groups of species used to indicate a healthy ecosystem and therefore have a big value when determining the environmental condition of the landscapes in Rio Grande do Sul (Renner et al.
2015). Even though only 0.01% of all of earth's water is easily available for us as freshwater in lakes and rivers (McAllister et al. 1997) it is still crucial for the survival of every living species.
Lakes and rivers also hold one of the proportionally largest amounts of species on the planet (Abell et al. 2002). The wellbeing of these ecosystems should therefore be in everyone's interest.
Rio Grande do Sul has, together with Sao Paulo, the highest number of threatened aquatic invertebrate in the whole country. This could be explained by the fact that southern Brazil is the wealthiest part of Brazil and therefore has the highest anthropogenic impact, holds a lot of endemic species and that it’s the area in Brazil where most studies on aquatic species has been made (Agostinho 2005).
9
Aim
The aim of this study is to see if we can observe a correlation between habitats with varying anthropogenic and environmental impact and dragonfly species diversity. By measuring the water quality in freshwater systems and dragonfly species caught at the location we hope to find
different species that indicates different water variables. With this information one could make a system for using dragonflies to locate different kinds of polluted or otherwise affected bodies of water. This could prove very useful since using species as bio indicators is a well-known method for environmental work. We will also try to observe if areas with a lot of forest holds a larger amount of individuals and more species than fragmented areas and thus further prove that conservation efforts should be made to preserve the Atlantic rainforest.
Materials and methods
Collection
This project took place in the area of a town called Lajeado which is located in Rio Grande do Sul, the most southern state of Brazil. We sampled dragonflies close to waters in two different areas (Fig. 1). Sampling was done on 16 different locations on 13/5/2016 and 20/5/2016. The first field trip took place around Santa Clara do Sul and nine different locations, named A2-A10, were sampled. Three people were sampling on these locations. The second field trip took place around the town Anta Gorda where we sampled four different location, and around the town Arvorezinha where we sampled three different locations. Six people were sampling on these locations, named B1-B7. A2-A10 were areas with high anthropogenic impact and B1-B7 with less. The water at the sampling sites were either standing or running. At each of the sixteen different locations we sampled adult dragonflies using nets in areas close to water. The specimens were immediately killed by being placed in a container filled with alcohol which also served as a preservative. If there was an observation of a known species that we were unable to catch, we noted the presence of the species at the site and added it to the database. We spent 10-20 minutes at each location and stopped when we had not caught any new species in some time. This
however was no guarantee that every single species had been caught.
10
At each site we further measured the variables; pH level, ORP, conductivity, turbidity, dissolved oxygen and total dissolved solids using a water measurement probe either by putting the probe directly in the water or by putting it in a bucket containing a collected sample of water. On each location several different environmental parameters were noted and rated from one to three.
These were; wind were one meant there was no wind at the time of sampling, two breeze and three intense winds, overcast were one meant there were no clouds, for two the sky was partially covered and three the sky was covered, water vegetation were one meant there were no
vegetation, two some and three the location was overgrown and condition of light and shade were one meant no shade, two partially shaded and three completely shaded. Some parameters were rated differently, these were: water condition which was either clear, murky, polluted or eutrophic and vegetation at water’s edge which was rated as forest, bushes, grass or none. More than one alternative could be filled in for the vegetation at water’s edge. We used a Garmin GPSMAP 78s GPS to get the coordinates for each sampling location. For all the determination work, analyses and writing we used the facilities and equipment provided to us by Univates, the university in Lajeado. For the map in Fig.1 we used ZeeMaps.
Fig 1.Map of Brazil with the city Lajeado marked in green and sampling locations from first field trip marked in red, sampling locations from the second field trip marked in yellow.
11 Data analysis and identification
The species collected was identified in the lab using microscopes and literature. For identification of the families and genera we used Garrison, von Ellenrieder & Louton (2006, 2010). To identify the odonata to species level we used Lencioni (2005, 2006) and Heckman (2006). We used Kompier (2015) and the website All Odonata (2016) as a final picture comparison for
identification. Since dragonflies are aerial insects and some species are incredibly quick, we were unable to collect every single species on every single location. To compensate for this we drew a collector's curve in which the sample locations were entered randomly using (random.org 2016).
From each sample site we drew a one square kilometer circle with the sample site in the center, using Google Earth Pro v7.1.5.1557. We then plotted each area into 5 different categories: water, human structures, grazing field and farmland and calculated the areas in percent of total area in the circle (Fig. 2). Water turned out to make up less than one percent of the total area and were therefore later excluded. We also made a Jackknife estimation by using the formula presented in Journal of Statistical Software by Smith & Pontius 2006 to see if our collection represents the estimated total dragonfly community.
Fig. 2. The different land use in a 1 square kilometer circle around location B3 where agriculture is highlighted in red, urban areas such as roads and buildings marked in black, fields marked in blue and forest marked in purple.
12
All of the data comparison were made with SPSS v20. We conducted a Kolmogorov-Smirnov test to see if the data were normally distributed or not. Then we used Pearson correlation to test if the number of species and the number of specimens of dragonflies were correlated with the different environmental variables; pH level, ORP, conductivity, turbidity, dissolved oxygen, total dissolved Solids and the percent of different land types surrounding the locations. We further tested the dragonfly species variation and total number with dragonflies caught with a t-test to see if there was a difference between different water types. We used Anova to see differences in species numbers related to overcast, conditions of light, shade and different vegetation.
A Canonical Correspondence Analysis CCA (ter Baak 1986) was performed to analyze the relationships between our water variables (temperature, pH, ORP, conductivity, turbidity, dissolved oxygen and total Dissolved Solids) and the occurrence of individual species of dragonflies at the site. Here we chose to use only the variables which we found to have a
significant correlation to either number of species or number of individuals. We used the software PAST v2.17 and v3.0 for the CCA analysis (Hammer, Harper & Ryan 2001).
Results
A total of 328 individuals distributed among 46 species and eight families were collected (Appendix 1). Nineteen species were Zygoptera (damselflies) and 27 were Anisoptera
(dragonflies). An average of 7, 3 different odonata species was caught on each location where A2 held the least number of species with only two. B6 and A9 held the most with a total of twelve different species.
Our Jackknife estimation suggests that there is 62 species in the region, implying that our
collection covers 74% of the total dragonfly species variation. The collector's curve further shows that there are more species in this area whom are not collected since the curve is still rising (Fig.
3). If the curve instead would have even out in the end it would tell us that we probably caught most of the species.
13
Fig. 3. Collector's curve showing the total number of individual dragonflies collected on the x- axis against the number of different species on the y-axis.
We found that all of our data used in SPSS was normally distributed. The number of species were correlated to the amount of surrounding forest (r = 0.588, p = 0.017) and water temperature (r = 0.564, p = 0.023) (Fig. 4). We also found a negative correlation between the number of species in each location and the pH level (r = -0.547, p = 0.028). We found no correlation between number of species with ORP, amount of surrounding urban area, amount of surrounding fields, amount of surrounding agriculture, conductivity, turbidity, dissolved oxygen and total dissolved Solids (p >
0.05).
14
(A) (B)
(C)
Fig. 4. Results from Pearson correlation tests that shows the two positive correlations, A and B, and the negative correlation C.
The first three axes generated by CCA explained 76% of the total dragonfly composition (Fig. 5).
The significance of the axes generated was tested 1000 times using the Monte Carlo test. The most important variables for dragonfly composition were pH, amount of forest, conductivity, temperature and total dissolved solid. The eigenvalues of the first three axes were 0,572, 0,517 and 0,323 respectively. As seen in Fig.5 species 1 and 40-46 are dependent on the amount of forest while species 9, 18, 19 and 20 are not. Species 14, 38, 39 prefers water with high pH, conductivity and Total Solids while the species preferring forest don’t like these locations,
15
Fig. 5. CCA results showing how the occurrence of species are related to different environmental factors. For list of species see appendix 1.
Several environmental factors were strongly correlated to the number of specimens found at each water body (Fig.6). We found a positive significant correlation between the number of
dragonflies and the amount of surrounding forest (r = 0.735, p = 0.001) and with water temperature (r = 0.759, p = 0.001). We also found a strong negative significant correlation between number of dragonflies and pH level (r = -0.686 p = 0.003), the water conductivity (r = - 0.711, p = 0.002), the waters total amount of dissolved solids (r = -0.709, p = 0.002) and with the amount of surrounding urban area (r = -0.541, p = 0.030). We found no significant correlation between number of dragonflies with ORP, amount of surrounding fields, amount of surrounding agriculture, conductivity and dissolved oxygen (p > 0.05).
16
(A) (B)
(C) (D)
(E) (F)
Fig. 6. Results from the Pearson correlation that shows the two positive correlations, A and B, as well as the negative correlations, C, D, E and F
17
The t-test confirmed that there was a significant difference in both the number of species (t = 2.408, df = 14, p = 0.030) and dragonfly specimens (t = 3.503, df = 14, p = 0.004) depending on if there is standing or running water present. We found that there is significantly higher number of species as well as number of specimens in standing (8.67 ± 3.28; 28.78 ± 13.95) than in running (5.43 ± 1.51; 9.86 ± 2.91) water, as shown in Fig. 7. We found no significant difference in dragonfly species richness and number of individual dragonflies with different vegetation types, overcast and the conditions of light and shade (p > 0.05).
(A) (B)
Fig. 7. Mean number of species (±standard deviation), A, and number of specimens mean, B, caught depending on if there was standing or running water present.
18
Discussion
The Pearson correlation result shows that the amount of surrounding forest, water temperature and pH level affects the dragonfly species richness in different water bodies (Fig. 4). Still water bodies with low pH and high temperature that are heavily surrounded by forest would have a significantly higher diversity of dragonfly species than running water bodies with the opposite values. It would also consist of bigger dragonfly communities (Fig. 6).
The CCA analysis shows that T binotata, Oxyagrion sp., L pictus, M stawiarskii, R planaltica, L auritus and L dichrostigma prefer areas with a lot of forest.
Fragmented forests in anthropogenic disturbed areas holds less specialist species and more generalists than forests in well preserved areas (Renner et. al. 2015). Our results show that areas with less amount of surrounding forest holds a smaller species diversity. The difference in species variation could possibly be explained by the absence of specialist species present in the disturbed areas with small forest fragments. Another reason for the connection could be that adult
dragonflies prefer to lay their eggs in waters surrounded by areas that are beneficial for them in later stages of their life cycle, where forests provide adult dragonflies with shelter and a place to rendezvous for mating (Watanabe et al. 1987). Oxyagrion sp. belongs to the Coenagrionidae family which has previously been shown to prefer habitats with big and untouched forests (Renner et al. 2015). This would explain its presence in local B6, since it was the area with the highest amount of forest of all the locals (see Appendix 2) and it also held the most species together with local A9. Local A2 held the least amount of species, which could be explained by the fact that it was the location with least amount of surrounding forest with only 11% (see Appendix 2).
S reticulata, L. bipupillatus, B furcata, T cophysa, R bonariensis and D mincki prefers waters with higher pH. M ocellata and the species connected to areas with a lot of forest, such as T binotata, Oxyagrion Sp, L pictus, M stawiarskii, R planaltica, L auritus and L dichrostigma prefers lower pH levels. Since both adult dragonflies and dragonfly eggs tolerate low pH values (Hudson & Berrill 1986) and the eggs only show negative effects from pH levels below 3.0 (Rychla, Benndorf & Buczyński 2011), there is a possibility that some dragonflies take advantage
19
of waters with low pH where the predation pressure is lower than in more pH neutral water (Rychla, Benndorf & Buczyński 2011). A study made in Sweden shows that some dragonfly species prefers acidified waters over limed waters which could be because liming of water raises the pH level (Al Jawaheri & Sahlén). This would affect the species who prefers more acidified lakes. These studies help explain our result where the number of species increased with lower pH.
Specialist species would benefit from the acidified water and since the lowest measured pH level was 4.62 it would not have a negative effect on the dragonflies nor its eggs. The t-test showed that standing water holds a bigger number of species and dragonflies than running water. This could be explained by the fact that running water is usually not as acidic as standing water and those waters could therefore be more appealing to dragonflies since they prefer water with lower pH levels.
S reticulata, L. bipupillatus, B furcata, R bonariensis, D mincki and T cophysa
prefers waters with high conductivity and with a lot of dissolved solids (Fig. 5). Dragonfly abundance is supposed to be negatively correlated to the water’s conductivity (Matchik et. al.
2010) which is confirmed by our results (Fig. 6). Conductivity is a well-known measurement for changes in the water (Fondriest environmental inc. 2015) and low conductivity can increase the effect of pollution in fish (Irrgang, Saynisch & Thomas 2015).
The amount of solids in the water also affects the number of dragonflies where locations with low level of dissolved solids holds bigger communities than water bodies with the opposite values.
Our results shows that conductivity and total dissolved solids are co-dependent since it follows the same pattern in both the CCA analysis (Fig. 5) and the Pearson correlation (Fig. 6). And since high conductivity are dependent on the amount of ions in the water (Irrgang, Saynisch & Thomas 2015), it would explain why amount of solids in the water also affects the number of dragonflies as well as conductivity.
High water temperature are known to be preferable when it comes to dragonfly species richness (Corbet 1999). Temperature and sunlight are important variables for dragonflies since the female chooses certain waters as home for her eggs. Since dragonflies prefers areas with a lot of
sunlight, it is possible that females chooses the waters with low shade and high amount of sunlight (Samways & Steytler 1996). This would increase the water temperature because of the exposure to sunlight.
20
The amount of urban area, in form of buildings and roads, affects the amount of dragonflies present where locations surrounded by a low percentage of urban areas holds bigger communities than the opposite (Fig. 6). The amount of human structures in an area is also directly linked to that area's amount of forest (Appendix 2), where areas with a large amount of human structures usually lack high amounts of forest and vice versa. Fields and agriculture, which are
anthropogenic disturbance, affects the amount of forest present but were classified separately.
Samways & Steytler (1996) found that the number of dragonfly species decreased in industrial areas but was compensated with the increase of individuals. This could possibly be explained by the method of classify the area, where we probably looked at a bigger total area. If we would have classified the landscape as only forest and anthropogenic disturbance, in form of fields, agriculture and urban structures, we might have gotten a different result.
Other species not mentioned were considered generalist since they were close to origo and therefore did not have a preference for a specific variable that we tested. They made up 76 % of the total species caught.
Possible error sources
Since there was a difference in people catching dragonflies in the field trips, where three people were catching on the first trip and six people catching in the second, it could affect the results in both the amount of species and dragonflies caught. The collector's curve implies that there still is still a good number of species not caught and the jackknife estimate tells us that we caught about 74% of the total species in the area. This implies that we caught a good number but that there is more. If we caught a bigger number the results might vary.
Dragonflies are active when there is sunlight, and if there were partially shade at some locals it could affect the dragonfly activity and therefore lower the catching rate.
The identification of the species was hard and there is a possibility that some of the species were incorrectly identified which would lead to a wrong number of the total species.
We found an unknown female that could not be identified which we labeled “unidentified female” and counted as a separate species. The female could be a species that we already caught and should therefore not be counted as a separate species. But since the unknown female did not
21
show any special preference in the CCA it is considered a generalist and therefore would not alter the result.
Conclusion
As mentioned in the introduction, acidification of lakes is shown to affect the biodiversity.
Dragonflies could benefit from acidification at a certain level because of the decrease in
competition from other species who are negatively affected by low pH value. But if the pH level drops below 3 if would also affect the dragonfly community in a negative way. Some species are certainly benefitted by some acidification, such as T binotata, Oxyagrion Sp, L pictus, M
stawiarskii, R planaltica, L auritus and L dichrostigma who are present in more acidified waters.
S reticulata, L. bipupillatus, B furcata and T cophysa are species that prefer waters with high conductivity. These species could possibly be used as indicators since pH and conductivity are known measures for the water quality, where pH can indicate the waters current status and conductivity can be used to measure changes. But more studies of this has to be done since our study repetition and the sampling only took place in a time span of two weeks.
Forest is shown to play a big part in the number of species present. Therefore forest
fragmentation pose a threat to the dragonfly species variation where the specialist species such as T binotata, Oxyagrion sp., L pictus, M stawiarskii, R planaltica, L auritus and L dichrostigma would suffer the most. So by halting the fragmentation of forests one would not only benefit the mammals and birds who is well known to suffer by the forest fragmentation, but also benefit the freshwater biodiversity in form of dragonflies. Dragonflies are in their role as predators also an indicator of the other trophic levels where an abundance of dragonflies indicates that there is a lot of food present. So by benefiting the dragonflies you could avoid a cascade effect, as with the sea otters on the North American pacific coast, where the removal of a predator would affect the whole food chain.
The increase in still waters in Rio Grande do Sul due to agriculture have had a positive effect for the dragonfly species richness and dragonfly community. But since the agriculture has been cutting down forests for land use, the benefit from an increase in still waters would probably be canceled out and even surpassed by the decrease in forest available. Species dependent on
22
running water are also vulnerable to the agricultural expansion since they will lose their habitats.
Since the Atlantic rainforest holds a big number of threatened species (Agostinho 2005) and our results shows that fragmentation could affect the dragonfly species and possibly cause a cascade effect, conservation measures has to be done to ensure that the Atlantic rainforest can recover from the huge amount of fragmentation in the area.
Acknowledgments
Our gratitude to Göran Sahlén for all his time, help and guidance during this project and for giving us the opportunity to come to Brazil. Samuel Renner for being immensely helpful with the identification. Our supervisor Eduardo Périco for giving us the opportunity to work in the lab.
Marina Schmidt Dalzochio for helping us with the water probe and for being immensely patience with all our questions. Thank to everyone else at the Lab de Evolução e Ecologia for helpful input, patience and help with the sampling. Jasmine Lindeberg for all the help with the
identification. A special thanks to Victor Brandt for the identification, valuable input and for the time spent together while sharing apartment which led to many helpful discussions that
eventually formed this thesis.
23
References
Abell, R., Thieme, M., Dinerstein, E. and Olson, D. 2002. A Sourcebook for Conducting Biological Assessments and Developing Biodiversity Visions for Ecoregion
Conservation. Volume II: Freshwater Ecoregions. World Wildlife Fund, Washington, DC, USA.
Agostinho, A. A., Thomaz, S. M. & Gomes, l. c. 2005. Conservation of the Biodiversity of Brazil's Inland Waters
Al Jawaheri, Raad. Sahlén, Göran. Lake liming proframs affect species richness of
dragonflies (Insecta: Odonata) negatively - a study from southern Sweden. Manuscript, Halmstad University, Sweden.
All Odonata. http://www.allodonata.com/ (Hämtad 2016-04-14)
Barbosa de Oliveira-Junior, J., Shimano, Y., Gardner, T., Hughes, R., de Marco Júnior P. and Juen, L. 2015. Neotropical dragonflies (Insecta: Odonata) as indicators of ecological condition of small streams in the eastern Amazon. Austral Ecology Volume 40, Issue 6:
733–744.
Bethell, L. The Cambridge History of Latin America. Cambridge University Press (July 25, 1986) 2: 47
Bonsdorff, E., Blomqvist, E. M., Mattila, J. and Norkko, A. 1997. Coastal eutrophication:
Causes, consequences and perspectives in the Archipelago areas of the northern Baltic Sea. Estuarine, Coastal and Shelf Science, 44, Supplement 1: 63-72.
Bouwman, A.F., Beusen, A.H.W. and Billen, G. 2009. Human alteration of the global nitrogen and Phosphorus soil balances for the period 1970–2050. Global Biogeochemical Cycles, Volume 23, Issue 4: 1-16 (Hämtad: 2016-03-03).
Brown, J. H. 1978. The Theory of Insular Biogeography and the Distribution of Boreal Birds and Mammals. Great Basin Naturalist Memoirs 2: 209–227 (Hämtad: 2016-03-07)
Chiarello, A. G. 2000. Conservation value of a native forest fragment in a region of extensive agriculture. Revista Brasileira de Biologia, 60:237-247.
Conservacion de la Biodiversidad de las Aguas Interiores de Brasil. Conservation Biology, 19, 646-652.
Convention on Biological Diversity, Strategic Plan for Biodiversity 2011-2020.
https://www.cbd.int/decision/cop/?id=12268 (Hämtad 2015-02-19).
Corbet, P. S. 1999. Dragonflies: Behavior and ecology of Odonata. Third printing. Cornell University Press:188-189
Cowling, E. 1990. Acid Rain and Other Airborne Pollutants: Their Human Causes and Consequences. Population and Development Review. Vol 16: 205-220
Debinski, D.,M. and Holt, R.D. 2000. A Survey and Overview of Habitat Fragmentation
Experiments.Conservation Biology Volume 14, Issue 2: 342–355 (Hämtad: 2016-03-07).
Dormann, C. F., Schweiger, O., Augenstein, I., Bailey, D., Billeter, R., De Blust, G., Defilippi, R., Frenzel, M., Hendrickx, F., Herzog, F., Klotz, S., Liira, J., Maelfait, J.-P., Schmidt, T., Speelmans, M., Van Wingerden, W. K. R. E. and Zobel, M. 2007. - Effects of landscape structure and land-use intensity on similarity of plant and animal communities. 16: 787.
24
Eskinazi-Sant’Anna, E.M., Menezes, R., Costa, I.S., Araújo, M., Panosso, R. and Attayde, J.L.
2013. Zooplankton assemblages in eutrophic reservoirs of the Brazilian semi-arid.
Brazilian Journal of Biology. Vol 73 no 1.
Estes, J.A., Duggins, D.O. and Rathbun G.B. 1989. The Ecology of Extinctions in Kelp Forest Communities. Conservation Biology. Volume 3, 1989. Issue 3: 252-264.
Fondriest environmental inc. 2015. Fundamentals of environmental measurements.
http://www.fondriest.com/environmental-measurements/parameters/water- quality/conductivity-salinity-tds/ (Hämtad 2016-05-19)
Garrison, R. W., von Ellenrieder, N. and Louton, J. A. 2006. Dragonfly Genera of the New World. Baltimore: The Johns Hopkins University Press.
Garrison, R. W., von Ellenrieder, N. and Louton, J. A. 2010. Damselfly Genera of the New World. Baltimore: The Johns Hopkins University Press.
Groom, M.J. and Vynne, C.H. 2006. Habitat degradation and loss. In Martha J, Groom, Gary K.
Meffe and C. Ronald Carrol. and contributors. Principles of conservation biology, 3rd edition. Massachusetts, USA: Sinauer Associates, Inc. Publishers Sunderland: 184-186.
Hammer, Ø., Harper, D.A.T. and Ryan, P.D. 2001. PAST: Paleontological Statistics Software Package for Education and Data Analysis. Palaeontological Association, 22 June 2001.
Heckman, C., W. 2006. Encyclopedia of South American Aquatic Insects: Odonata - Anisoptera:
Springer Netherlands
Hudson, J. and Berrill, M. 1986, "Tolerance of low pH exposure by the eggs of Odonata (dragonflies and damselflies)", Hydrobiologia, vol. 140, no. 1: 21-25.
Irrgang, C., Saynisch, J. & Thomas, M. 2015, "Impact of variable sea-water conductivity on motional induction simulated with an OGCM", Ocean Science Discussions,vol. 12, no. 4, pp. 1869-1891.
Isermann, K. 1990. Share of agriculture in nitrogen and phosphorus emissions into the surface waters of Western Europe against the background of their eutrophication. Fertilizer research, 26: 253-269.
Keller, L.F. and Waller, D.M. Inbreeding effects in wild populations. 2002. Trends in Ecology &
Evolution. Volume 17, Issue 5: 230–241.
Koch,K., Wagner, C and Sahlén, G. 2014. Farmland versus forest: comparing changes in Odonata species composition in western and eastern Sweden. Insect Conservation and Diversity. Volume 7, Issue 1: 22-31 (Hämtad: 2016-03-28).
Kompier, T. 2015. A guide to the Dragonflies and Damselfies of the Serra dos Organos: REGUA Publications.
Kubesh, K., Mcneil, N. and Bellotto, K. 2008. The Industrial Revolution, HOCPP1271.
Leipelt, K.G. and Suhling, F. 2005, Larval biology, life cycle and habitat requirements of Macromia splendens, revisited (Odonata: Macromiidae)", International Journal of Odonatology, vol. 8, no. 1: 33
Lencioni, F. A. A. 2005. Damselfies of Brazil - An illustrated identification guide 1 – Non Coenagrionidae families. Sao Paulo: All Print Editora.
Lencioni, F. A. A. 2006. Damselfies of Brazil - An illustrated identification guide 2 – Coenagrionidae. Sao Paulo: All Print Editora.
Losos, J B., Ricklefs, R E. 2009. Theory of Island Biogeography Revisited (Hämtad: 2016-03- 07).
MacArthur, R. H. and Wilson E.O. 1967. The Theory of Island Biogeography. Princeton University Press
25 Matchik, L., Stenert, C., Kotzian C. B. and Pires, M. M. 2010. Responses of Odonate
Communities to Environmental Factors in Southern Brazil Wetlands. Journal of the Kansas Entomological Society. Vol. 83: 208-220.
Mayhew, S. 2015. A Dictionary of Geography, Oxford University Press.
McAllister, D. E., Hamilton, A.L. and Harvey, B. 1997. Global freshwater biodiversity: Striving for the integrity of freshwater ecosystems. Sea Wind 11(3): 1-140 (Hämtad: 2016-02-23).
Murcia, C. 1995. Edge effect in fragmented forests: implications for conservation. Trends in Ecology & Evolution Volume 10, Issue 2: 58–62.
Noss, R., Csuti B., Groom, M,J and Fainsinger, P. 2006. Habitat Fragmentation. In Groom, M.J., Meffe, G.K., & Carroll, C.R and contributors. (2006). Principles of conservation biology, 3rd edition. Massachussetts, USA: Sinauer Associates, Inc. Publishers Sunderland.
Random.org. https://www.random.org (Hämtad 2016-05-16)
Renner, S., Sahlén, G. and Périco, E. 2015. Testing Dragonflies as Species Richness Indicators in a Fragmented Subtropical Atlantic Forest Environment. Neotropical Entomology: 1-9.
Renner, S., Périco, E., Sahlen, G., Martins dos Santos, D. and Consatti, G. 2015. Dragonflies (Odonata) from the Taquari River valley region Rio Grande do Sul, Brazil. Check List the Journal of Biodiversity Data. Vol 11, No 5
Ribeiro, M.C., Metzger, J.P., Martensen, A.C., Ponzoni, F.J. and Hirota, M.M. The Brazilian Atlantic Forest: How much is left, and how is the remainingforest distributed?
Implications for conservation. 2009. Biological Conservation 142: 1141-1153 Rychla, A., Benndorf, J. and Buczyński, P. 2011. Impact of pH and conductivity on species
richness and community structure of dragonflies (Odonata) in small mining lakes.
Fundamental and Applied Limnology. Vol. 179: 41-50
Sahlén, G. and Ekestubbe, K. Identification of dragonflies (Odonata) as indicators of general species richness in boreal forest lakes. Biodiversity & Conservation, 10: 673-690.
Sala, O. E., Stuart Chapin, F., Iii, Armesto, J. J., Berlow, E., Bloomfield, J., Dirzo, R., Huber Sanwald, E., Huenneke, L. F., Jackson, R. B., Kinzig, A., Leemans, R., Lodge, D. M., Mooney, H. A., Oesterheld, M. N., Poff, N. L., Sykes, M. T., Walker, B. H., Walker, M.
& Wall, D. H. 2000. Global Biodiversity Scenarios for the Year 2100. Science, 287:
1770-1774.
Samways, M.J. & Steytler, N.S. 1996, "Dragonfly (Odonata) distribution patterns in urban and forest landscapes, and recommendations for riparian management", Biological Conservation, vol. 78, no. 3, pp. 279-288.
Shigaki, F., Sharpley, A. and Prochnow, L. I. 2006. Animal-based agriculture, phosphorus management and water quality in Brazil: options for the future. Scientia Agricola, 63:
194-209.
Silk, N. and Ciruna, K. 2005. Practitioner's Guide to Freshwater Biodiversity Conservation Island Press.
Silva, J. M. C., Casteleti, C.H., 2003. Status of the biodiversity of the Atlantic Forest of
Brazil. In: Galindo-Leal, C., Câmara, I.G. (Eds.), The Atlantic Forest of South America:
Biodiversity Status, Threats, and Outlook. Island Press, Washington, DC: 43–59.
Smith, J., Samways, M. J. and Taylor, J. 2007. Assessing riparian quality using two complementary sets of bioindicators. Biodivers Conserv 16:2695–2713.
Smith, C. D. and Pontius, J. S. 2006. Jackknife estimator of Species Richness. Journal of Statistical Software. Volume 15, Issue 3.
26 SOS Mata Atlântica, Instituto Nacional de Pesquisas Espaciais, 2014. Atlas dos
remanescentes florestais da Mata Atlântica. https://www.sosma.org.br/projeto/atlas-da- mata-atlantica/ (Hämtad 2016-05-28)
Stevens, C A., Thompson, K., Grime, J P,. Long C J, and Gowing D J G. 2009. Contribution of acidification and eutrophication to declines in species richness of calcifuge grasslands along a gradient of atmospheric nitrogen deposition. Functional Ecology. 24: 478-484.
Suhling, F., Sahlén, G., Martens, A., Marais, E. and SchéTte, C. 2006. Dragonfly assemblages in arid tropical environments: a case study from western Namibia. In: HAWKSWORTH, D.
L. & BULL, A. T. (eds.) Arthropod Diversity and Conservation. Dordrecht: Springer Netherlands.
Tabarelli, M., Mantovani, W. and Peres, C. A. 1999. Effects of habitat fragmentation on plant guild structure in the montane Atlantic forest of southeastern Brazil. Biological Conservation, 91: 119-127.
Tabarelli, M., Aguiar, A.V. ,Ribeiro, M.C., Metzger, J.P. and Peres, C.A. 2010. Prospects for biodiversity conservation in the Atlantic Forest: Lessons from aging human-modified landscapes. Biological Conservation 143: 2328-2340.
ter Baak, C.J.F. 1986. Canonical Correspondence Analysis: A New Eigenvector Technique for Multivariate Direct Gradient Analysis. Ecology. 67(5). 1986: 1167-1179
Theobald, T.F.H., Daedlow, K. and Kern, J. Phosphorus availability and farm structural factors: examining scarcity and oversupply in north-east Germany. Soil Use and
Management Volume 31, Issue 3: 350–357 (Hämtad: 2016-03-04).
The Nature Conservancy
http://www.nature.org/ourinitiatives/regions/southamerica/brazil/placesweprotect/amazon -rainforest-infographic.xml (Hämtad: 2016-05-10)
Ward, J.V. 1998. Riverine landscapes: biodiversity patterns, disturbance regimes, and aquatic conservation. Biol. Conserv. 83: 269–278.
Watanbe, M., Ohsawa, N. and Taguchi, M. 1987. Territorial behaviour in Platycnemis echigoana Ashina at sunflecks in climax deciduous forests (Zygopera:
Platycnemididae). Odonatologica 16:273-280.
Worldometers http://www.worldometers.info/world-population/ (Hämtad: 2016-03-03).¨
ZeeMaps https://www.zeemaps.com.
27
Appendix 1 All caught species and their respective number as seen in Fig 5 Number Species
1 Libellulidae Micrathyria ocellata 2 Coenagrionidae Acanthagrion gracile 3 Coenagrionidae Oxyagrion chapadense 4 Coenagrionidae Oxyagrion terminale 5 Protoneuridae Peristicta aeneoviridis 6 Libellulidae Erythrodiplax atroterminata 7 Libellulidae Macrothemis imitans
8 Libellulidae Orthemis discolor 9 Aeshnidae Staurophlebia reticulata 10 Coenagrionidae Oxyagrion hempeli 11 Calopterygidae Hetaerina rosea 12 Coenagrionidae Argia albistigma 13 Libellulidae Pantala flavescens 14 Libellulidae Perithemis iceteroptera 15 Coenagrionidae Ischnura fluviatilis 16 Libellulidae Erythrodiplax hyalina
17 Libellulidae Erythrodiplax paraguayensis 18 Lestidae Lestes bipupillatus
19 Libellulidae Brachymesia furcata 20 Libellulidae Tramea cophysa 21 Libellulidae Erythrodiplax media 22 Libellulidae Erythemis plebeja 23 Libellulidae Erythemis peruviana
28 24 Libellulidae Miathyria marcella 25 Coenagrionidae Telebasis corallina 26 Coenagrionidae Ischnura capreolus 27 Unidentified female
28 Gomphidae Progomphus sp.
29 Libellulidae Perithemis mooma 30 Protoneuridae Neoneura leonardoi 31 Coenagrionidae Acanthagrion lancea 32 Libellulidae Micrathyria hesperis 33 Libellulidae Nephepeltia berlai 34 Coenagrionidae Argia Sp.
35 Calopterygidae Mnesarete pruinosa 36 Libellulidae Erythrodiplax melanorubra 37 Libellulidae Dasythemis mincki
38 Aeshnidae Rhionaeschna bonariensis 39 Libellulidae Erythrodiplay Sp.
40 Libellulidae Tramea binotata 41 Coenagrionidae Oxyagrion Sp.
42 Lestidae Lestes pictus
43 Libellulidae Micrathyria stawiarskii 44 Aeshnidae Rhionaeschna planaltica 45 Lestidae Lestes auritus
46 Lestidae Lestes dichrostigma
Table showing the amount of forest, fields, agriculture and human structures on each location
29 Appendix 2
Local Forest % Fields Agriculture Human
structures
A2 11 % 9 % 17 % 47 %
A3 15 % 0% 12 % 57 %
A4 15 % 7 % 18 % 42 %
A5 15 % 17 % 17 % 30 %
A6 13 % 35 % 2 % 22 %
A7 18 % 15 29 % 11 %
A8 14 % 9 % 56 % 5 %
A9 21 % 7 % 45 % 6 %
A10 24 % 8 % 45 % 6 %
B1 14 % 14 % 35 % 16 %
B2 28 % 23 % 27 % 9 %
B3 27 % 15 % 32 % 8 %
B4 27 % 47 % 0 % 10 %
B5 34 % 5 % 42 % 3 %
B6 84 % 0 % 0 % 1 %
B7 80 % 0 % 0 % 1 %
Besöksadress: Kristian IV:s väg 3 Postadress: Box 823, 301 18 Halmstad Telefon: 035-16 71 00
E-mail: registrator@hh.se www.hh.se
Philip Sjöwall Robert Lundström
