**1. Introduction**

[51] Riznichenko GY. Mathematical models in biophysics [Internet]. 2012. Available from:

[52] Tchuraev RN. On a stochastic model of a molecular genetic system capable of differ‐ entiation and reproduction of the initial state. Biometrical J. [Internet]. 1980 [cited 2013 Nov 5];22(2):189–94. Available from: http://doi.wiley.com/10.1002/bimj.

[53] Tchuraev R., Stupak I., Tropynina T., Stupak E. Epigenes: design and construction of new hereditary units. FEBS Lett. [Internet]. 2000 [cited 2013 Nov 5];486(3):200–2. Available from: http://www.sciencedirect.com/science/article/pii/S0014579300023000

[54] Lashin S a., Suslov V V., Matushkin YG. Comparative Modeling of Coevolution in Communities of Unicellular Organisms: Adaptability and Biodiversity. J. Bioinform. Comput. Biol. [Internet]. 2010 Jun [cited 2013 Jan 31];08(03):627–43. Available from:

[55] Lashin S a., Matushkin YG, Suslov V V., Kolchanov N a. Evolutionary trends in the prokaryotic community and prokaryotic community-phage systems. Russ. J. Genet. [Internet]. 2011 Dec 8 [cited 2013 Jan 31];47(12):1487–95. Available from: http://

http://www.worldscientific.com/doi/abs/10.1142/S0219720010004653

www.springerlink.com/index/10.1134/S1022795411110123

http://www.biophysics.org/Portals/1/PDFs/Education/galina.pdf

4710220212

136 Biodiversity - The Dynamic Balance of the Planet

The Amazon River and its main tributaries are known as downstream migratory way of eggs and larvae of many migratory fishes [1, 2, 3]. The young fish drift may occur during all year but the great abundance is associated with the flood pulse pattern, generally with the rising of the river level [4, 5]. The eggs, larvae and juveniles may drift for hundred or thousand kilometers and repopulate the stocks in the down river portion [6; 7; 8; 9].

The maintaining of the larval drift process is fundamental for the conservation of a great number of migratory fishes as well as for the maintaining of local economy and food security. The migratory species are the most important fisheries resource of the Amazon Basin and they are responsible for about 87% of the total landing in 66 ports of the main Amazon cities [10]. The fishery production, in special the fishery of migratory species, maintains an important sector of the Amazon economy, generating about US\$130 million yr-1 and more than 160 thousand jobs [11]. The fish consumption rate by Brazilian Amazonian population is 24 kg annual per capita or 343 thousand ton/year, almost four times of the fish consumption rate of the Brazilian population [12].

One of the biggest threats to the drift movement of young fish drift movement are large dams, which interrupt the river connection and reduce or change the natural flood pulse pattern [13, 14]. There were few large hydroelectric dams in the Amazon basin, but this situation is changing by the construction of several hydroelectric dams in large tributaries of the Amazon River [15]. The most recent dams are two hydroelectric power plants nearby Porto Velho city, in the State of Rondônia, Brazil, which are damming the largest tributary of the Amazon River, the Madeira River.

© 2014 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

The Madeira River Basin occupies 1.4 million km2 or 20% of the Amazon Basin, which 50% belongs to Bolivia, 40% to Brazil and 10% to Peru. The Andes headwater is in Bolivia and Peru and the lowland portion and the Brazilian Shields headwater are in Brazil. The Madeira River at the frontier border of Bolivia and Brazil is characterized by a sequence of waterfalls, which separate the Llanos de Mojo savanna in Bolivia, the largest floodplains area of the Madeira River, and its lowland portion [16]. The Jirau and Santo Antonio hydroelectric power plants are being built above of the low portion in order to generate energy from the waterfalls of the Madeira River. The reservoirs of both dams should occupy an area around 530 km2 , which will flood only the Brazilian territory. The highest difference between the upper and lower limits of both reservoirs is about 50 m (Environmental Impact Assessment (EIA) of Madeira River, http://www.ibama.gov.br/licenciamento/index.php).

In order to assess the impact of the dams over the ichthyoplankton drifting along the Madeira River, it was established by the hydroelectric company a sampling protocol to collect period‐ ically along the river before closuring the dam and, then in the reservoir, just below the dam. One methodology adopted was used to compare the larval density estimated from samples taken at different depths along the river transect [2, 3, 4, 5, 23]. In addition, it was developed a methodology in order to estimate the larval flux in different cross section of the river. This methodology was adapted from the direct methods for measuring suspended-sediment discharge in rivers [25]. The utilization of both methods will help to decrease the gear selec‐ tivity of the nets and methodologies and to compare the results of each one.

The aim of this study is to compare the selectivity of three sampling methods over the species composition and developmental stage of fish, as well as characterize the drift movement of ichthyoplankton in the waterfalls stretch of the Madeira River, considering the flood seasonal and spatial variation, before the closure of the Jirau Dam. The present study is a reference to the future impact assessment of the ichthyoplankton drift pattern related to the establishment of power plants.

**Figure 1.** Perennial and seasonal rapids between the two sampling sites in the Madeira River.

/s, in March, and 5,959 m3

/s (River Discharge Database-Sage: http://www.sage.wisc.edu).

ranges from 34,512 m3

**3. Sampling and data analysis**

19,222 m3

The river's length between two sampling sites is about 140 km, river's width ranges from 300 to 1200 m, and altitude varies from 72 to 92 m. The flooding area in this stretch is very narrow if compared to the flooding area of the Beni or Mamoré River in upriver of the Jirau´s Reservoir. The annual precipitation is about 2,650 mm, with intensive rainfall from December to April, while the driest period goes from May to November. The difference between high and low water level, in general, goes from 10 to 12 m [16]. The monthly mean of freshwater discharge

Diversity and Abundance of Fish Larvae Drifting in the Madeira River, Amazon Basin: Sampling Methods Comparison

Ichthyoplankton samples were obtained in two perpendicular transects in the Madeira River located at two sampling sites. In the Abunã site, at upriver, was established one transect above and other below of the confluence with the Abunã River and Madeira River. At downriver

/s, in September, showing an annual mean of

http://dx.doi.org/10.5772/57404

139

### **2. Study area**

The fish eggs, larvae and juveniles were collected monthly from October 2009 to September 2012 in two sampling sites in the Madeira River. The locations of those sites were defined by the limits of the planned reservoir of the Jirau Hydroelectric Power Plant (HPP). The sample sites were established in the upper and downriver limits of the planned reservoir in order to compare future modifications of the drift fish larvae patterns. At upriver site is the confluence of Abunã and Madeira River (9o 40' S-65o 26'W), and at downriver site is the Jirau Hydroelectric Power Plant (9o 15' S-64o 38'W). Between these two sampling sites, there are four perennial rapids (Pederneiras or Tamborete, Paredão, Jirau and Caldeirão do Inferno) and five seasonal rapids (Machado, Prainha, Vai Quem Quer, Dois Irmãos and Embaúba). Along this stretch there is the biggest rapid, known as Jirau waterfall, which is located about 13 km upside of Jirau HPP (Figure 1).

Diversity and Abundance of Fish Larvae Drifting in the Madeira River, Amazon Basin: Sampling Methods Comparison http://dx.doi.org/10.5772/57404 139

**Figure 1.** Perennial and seasonal rapids between the two sampling sites in the Madeira River.

The river's length between two sampling sites is about 140 km, river's width ranges from 300 to 1200 m, and altitude varies from 72 to 92 m. The flooding area in this stretch is very narrow if compared to the flooding area of the Beni or Mamoré River in upriver of the Jirau´s Reservoir. The annual precipitation is about 2,650 mm, with intensive rainfall from December to April, while the driest period goes from May to November. The difference between high and low water level, in general, goes from 10 to 12 m [16]. The monthly mean of freshwater discharge ranges from 34,512 m3 /s, in March, and 5,959 m3 /s, in September, showing an annual mean of 19,222 m3 /s (River Discharge Database-Sage: http://www.sage.wisc.edu).

### **3. Sampling and data analysis**

The Madeira River Basin occupies 1.4 million km2

138 Biodiversity - The Dynamic Balance of the Planet

http://www.ibama.gov.br/licenciamento/index.php).

of power plants.

**2. Study area**

Power Plant (9o

Jirau HPP (Figure 1).

of Abunã and Madeira River (9o

15' S-64o

belongs to Bolivia, 40% to Brazil and 10% to Peru. The Andes headwater is in Bolivia and Peru and the lowland portion and the Brazilian Shields headwater are in Brazil. The Madeira River at the frontier border of Bolivia and Brazil is characterized by a sequence of waterfalls, which separate the Llanos de Mojo savanna in Bolivia, the largest floodplains area of the Madeira River, and its lowland portion [16]. The Jirau and Santo Antonio hydroelectric power plants are being built above of the low portion in order to generate energy from the waterfalls of the

flood only the Brazilian territory. The highest difference between the upper and lower limits of both reservoirs is about 50 m (Environmental Impact Assessment (EIA) of Madeira River,

In order to assess the impact of the dams over the ichthyoplankton drifting along the Madeira River, it was established by the hydroelectric company a sampling protocol to collect period‐ ically along the river before closuring the dam and, then in the reservoir, just below the dam. One methodology adopted was used to compare the larval density estimated from samples taken at different depths along the river transect [2, 3, 4, 5, 23]. In addition, it was developed a methodology in order to estimate the larval flux in different cross section of the river. This methodology was adapted from the direct methods for measuring suspended-sediment discharge in rivers [25]. The utilization of both methods will help to decrease the gear selec‐

The aim of this study is to compare the selectivity of three sampling methods over the species composition and developmental stage of fish, as well as characterize the drift movement of ichthyoplankton in the waterfalls stretch of the Madeira River, considering the flood seasonal and spatial variation, before the closure of the Jirau Dam. The present study is a reference to the future impact assessment of the ichthyoplankton drift pattern related to the establishment

The fish eggs, larvae and juveniles were collected monthly from October 2009 to September 2012 in two sampling sites in the Madeira River. The locations of those sites were defined by the limits of the planned reservoir of the Jirau Hydroelectric Power Plant (HPP). The sample sites were established in the upper and downriver limits of the planned reservoir in order to compare future modifications of the drift fish larvae patterns. At upriver site is the confluence

rapids (Pederneiras or Tamborete, Paredão, Jirau and Caldeirão do Inferno) and five seasonal rapids (Machado, Prainha, Vai Quem Quer, Dois Irmãos and Embaúba). Along this stretch there is the biggest rapid, known as Jirau waterfall, which is located about 13 km upside of

26'W), and at downriver site is the Jirau Hydroelectric

38'W). Between these two sampling sites, there are four perennial

Madeira River. The reservoirs of both dams should occupy an area around 530 km2

tivity of the nets and methodologies and to compare the results of each one.

40' S-65o

or 20% of the Amazon Basin, which 50%

, which will

Ichthyoplankton samples were obtained in two perpendicular transects in the Madeira River located at two sampling sites. In the Abunã site, at upriver, was established one transect above and other below of the confluence with the Abunã River and Madeira River. At downriver site, the positions of transects were established upside and downside of the Caldeirão do Inferno rapid, where the Jirau Hydroeletric Power Plant (Jirau HPP) is being built (Figure 2).

with 1 m2

of area, 2 m of length, and 5 mm of mesh size. The volume of water filtered was

It was tested two different sampling methods to estimate the abundance of the ichthyoplankton drift in the Madeira River. The first used the point sampling method [18] and fixed five sites sampling in a linear transect to take samples in margins, main channel and zones in-between the main channel and the margins [3, 5]. The second used the integrating sampling method [18], which applied the systematic sampling method to define the sites sampling along of transect. The Point sampling with Larval net - PL - collects two samples in different depth at the water column. The remote open/close system set up in the mouth of the larval net ensured the net to collect samples at one meter from surface and at 70% of the total depth. The time spent in each sample was five minutes, and the distance from the margin ranged from 5 to 20 m. The Integrating sampling using Larval net – IL - and Juvenile net – IJ - takes one integrating samples of the water column, from the surface to the bottom and back. The velocity of net moving down and up was constant and determined by an electric winch. The distance of sampling sites was about 100 m each other and, at least, 20 m of the margin. The number of

Diversity and Abundance of Fish Larvae Drifting in the Madeira River, Amazon Basin: Sampling Methods Comparison

The physical and chemical characteristics of the Madeira River's water were monitored by

(ntu) in the surface and at 70% of the bottom depth of each sample spot. The estimative of the freshwater discharge of the Madeira River was based on the river level of the Abunã site. The river level and the river discharge of the Abunã station was supplied by fluviometric station located in the district of Abunã, in the city of Porto Velho / RO (UTM 20L 240534 and 8926519),

The ichthyoplankton samples were fixed in a 4% formalin solution and then they were counted and identified considering the taxonomy and the developmental stages. The developmental stages considered in this study were egg, larva yolk, pre-flexion, flexion, post-flexion and

The larval density was determined for each sample considering the number of ichthyoplankton

the river was estimated multiplying the average of the larval density of the transect by the

study area due to this minimal contribution to the overall freshwater discharge. The larval flux variation was analyzed by Factorial ANOVA, which tested the effect of the sampling method, river site and ichthyoplankton development stage. Pearson coefficient (r) was used to test the

The freshwater discharge of the Madeira River is characterized by an annual unimodal cycle defined by four phases: low, rising, high and falling water level. The low water level phase is

/s). It was not considered the discharge of the small tributaries along the

). The larval flux (larvae/s) of the cross section of

C) and turbidity

http://dx.doi.org/10.5772/57404

141

measuring dissolved oxygen (mg/l), pH, conductivity (µS/cm), temperature (o

juvenile [19, 20]. The fish larvae were identified according to [21, 17].

measured by a flowmeter set up in the mouth of each net.

spots in each transect was related with width of the river.

operated by the Geological Survey of Brazil (CPRM).

and the volume of water filtered (larvae/m3

**4. Hydrology and environment variables**

relationship between the variables.

diary discharge (m3

**Figure 2.** Four transects where ichthyoplankton samples are collected in reservoir area of the Jirau HPP, in Madeira River Basin.

The quantitative samples of ichthyoplankton were obtained during the daylight by towing an ichthyoplankton net from an outboard motor boat equipped with an electric winch. During sampling, the boat's position was maintained by the motor, which was moving upstream with the same speed of the river. Nocturnal samples were not collected because was assumed that the ichthyoplankton drift does not vary along day periods, as reported for rivers that show similar physicochemical characteristics of Madeira River [2].

Two different nets were used to collect ichthyoplankton in the Madeira River. The first one, nominated as larval net, has a conical-cylindrical format, with an aperture diameter of 0.5 m (0.1935 m2 ), 1.5 m length, and a mesh size of 0.5 mm [17]. This type of net has a collector cup used to open and close the net at different depth. The second one, nominated as juvenile net, was developed to collect large larvae and juvenile fish. It has a square frame in the open mouth, with 1 m2 of area, 2 m of length, and 5 mm of mesh size. The volume of water filtered was measured by a flowmeter set up in the mouth of each net.

site, the positions of transects were established upside and downside of the Caldeirão do Inferno rapid, where the Jirau Hydroeletric Power Plant (Jirau HPP) is being built (Figure 2).

**Figure 2.** Four transects where ichthyoplankton samples are collected in reservoir area of the Jirau HPP, in Madeira

The quantitative samples of ichthyoplankton were obtained during the daylight by towing an ichthyoplankton net from an outboard motor boat equipped with an electric winch. During sampling, the boat's position was maintained by the motor, which was moving upstream with the same speed of the river. Nocturnal samples were not collected because was assumed that the ichthyoplankton drift does not vary along day periods, as reported for rivers that show

Two different nets were used to collect ichthyoplankton in the Madeira River. The first one, nominated as larval net, has a conical-cylindrical format, with an aperture diameter of 0.5 m

used to open and close the net at different depth. The second one, nominated as juvenile net, was developed to collect large larvae and juvenile fish. It has a square frame in the open mouth,

), 1.5 m length, and a mesh size of 0.5 mm [17]. This type of net has a collector cup

similar physicochemical characteristics of Madeira River [2].

River Basin.

140 Biodiversity - The Dynamic Balance of the Planet

(0.1935 m2

It was tested two different sampling methods to estimate the abundance of the ichthyoplankton drift in the Madeira River. The first used the point sampling method [18] and fixed five sites sampling in a linear transect to take samples in margins, main channel and zones in-between the main channel and the margins [3, 5]. The second used the integrating sampling method [18], which applied the systematic sampling method to define the sites sampling along of transect. The Point sampling with Larval net - PL - collects two samples in different depth at the water column. The remote open/close system set up in the mouth of the larval net ensured the net to collect samples at one meter from surface and at 70% of the total depth. The time spent in each sample was five minutes, and the distance from the margin ranged from 5 to 20 m. The Integrating sampling using Larval net – IL - and Juvenile net – IJ - takes one integrating samples of the water column, from the surface to the bottom and back. The velocity of net moving down and up was constant and determined by an electric winch. The distance of sampling sites was about 100 m each other and, at least, 20 m of the margin. The number of spots in each transect was related with width of the river.

The physical and chemical characteristics of the Madeira River's water were monitored by measuring dissolved oxygen (mg/l), pH, conductivity (µS/cm), temperature (o C) and turbidity (ntu) in the surface and at 70% of the bottom depth of each sample spot. The estimative of the freshwater discharge of the Madeira River was based on the river level of the Abunã site. The river level and the river discharge of the Abunã station was supplied by fluviometric station located in the district of Abunã, in the city of Porto Velho / RO (UTM 20L 240534 and 8926519), operated by the Geological Survey of Brazil (CPRM).

The ichthyoplankton samples were fixed in a 4% formalin solution and then they were counted and identified considering the taxonomy and the developmental stages. The developmental stages considered in this study were egg, larva yolk, pre-flexion, flexion, post-flexion and juvenile [19, 20]. The fish larvae were identified according to [21, 17].

The larval density was determined for each sample considering the number of ichthyoplankton and the volume of water filtered (larvae/m3 ). The larval flux (larvae/s) of the cross section of the river was estimated multiplying the average of the larval density of the transect by the diary discharge (m3 /s). It was not considered the discharge of the small tributaries along the study area due to this minimal contribution to the overall freshwater discharge. The larval flux variation was analyzed by Factorial ANOVA, which tested the effect of the sampling method, river site and ichthyoplankton development stage. Pearson coefficient (r) was used to test the relationship between the variables.
