Mecanismos de emissão de CO em reservatórios do semiárido ... · Agradeço à Carlos Eduardo Alencar (Cadú), pelo companheirismo e ajuda em momentos decisivos deste trabalho e

1

UNIVERSIDADE FEDERAL DO RIO GRANDE DO NORTE – UFRN

CENTRO DE BIOCIÊNCIAS-CB

PROGRAMA DE PÓS-GRADUAÇÃO EM ECOLOGIA

Mecanismos de emissão de CO2 em reservatórios do

semiárido brasileiro

Caroline Gabriela Bezerra de Moura

Orientador: André Megali Amado

Natal

2015

2

“Mecanismos de emissão de CO2 em reservatórios do semiárido brasileiro”

Caroline Gabriela Bezerra de Moura

Tese, apresentada ao Programa de Pós-Graduação em

Ecologia – PPGE – da Universidade Federal do Rio Grande do

Norte – UFRN – para o exame de defesa para obtenção de

título em nível de Doutorado

BANCA EXAMINADORA

______________________________________

Prof. Dr. André Megali Amado – Orientador

Universidade Federal do Rio Grande do Norte

_____________________________________________

Prof. Dr. Vinícius Fortes Farjalla – Membro Externo

Universidade Federal do Rio de Janeiro

____________________________________________________________

Prof. Dr. Hugo Miguel Preto de Moraes Sarmento – Membro Externo

Universidade Federal de São Carlos

______________________________________

Prof. Dr. Vanessa Becker – Membro Interno


__________________________________________

Prof. Dr. José Luiz de Attayde – Membro Interno


Natal

2015

3

Universidade Federal do Rio Grande do Norte - UFRN

Sistema de Bibliotecas - SISBI

Catalogação de Publicação na Fonte. UFRN - Biblioteca Setorial do Centro de Biociências - CB

Moura, Caroline Gabriela Bezerra de.

Mecanismos de emissão de CO2 em reservatórios do semi-árido

brasileiro / Caroline Gabriela Bezerra de Moura. - Natal, 2015.

116 f.: il.

Orientador: prof. Dr. André Megali Amado.

Tese (Doutorado) - Universidade Federal do Rio Grande do

Norte. Centro de Biociências. Programa de Pós-Graduação em

Ecologia.

1. Balanço de carbono - Tese. 2. Metabolismo aquático - Tese.

3. Translocação de nutrientes - Tese. 4. Peixes onívoros - Tese.

5. Peixes bentívoros - Tese. I. Amado, André Megali. II.

Universidade Federal do Rio Grande do Norte. III. Título.

RN/UF/BSE-CB CDU 621.564.2

4

AGRADECIMENTOS

À Deus, por me conceder saúde e equilíbrio emocional para concluir esta etapa tão

importante da minha vida.

Aos meus pais – Gilson José de Moura e Maria da Conceição Bezerra de Moura - ,

infinitamente serei grata por tudo o que me ensinaram e tenham certeza (assim na

terra como no céu), que eternamente levarei vocês dois como modelo para a

educação dos meus filhos, amo vocês!

Aos meus irmãos (Gilsinho, Lisa e Ananda), sem vocês acho que não conseguiria

atravessar e vencer momentos tão difícies pelos quais tive oportunidade de trilhar

e aprender muito. Sou grata à vocês, minha família amada!

Agradeço à Carlos Eduardo Alencar (Cadú), pelo companheirismo e ajuda em

momentos decisivos deste trabalho e da vida pessoal.

Agradeço a toda a minha família, por todo o amor e carinho que sinto de cada

componente.

Aos meus companheiros amados de jornada, principlamente os que fazem parte da

velha guarda do DOL (Fafá, Anizão, Iagê, Andrievisk, Renatinha), levo vocês no

meu coração, e a amizade linda que levarei eternamente, fofinhos, obrigada por

toda a infinita ajuda que sempre estiveram prontos à oferecer. Agradeço também a

nova guarda do DOL (Veró, Lenice, Haig They, Dedé), vocês foram cruciais em

vários momentos ao longo do desenvolvimento deste trabalho.

5

Aos meus companheiros de muitas aventuras, aprendizado, e acima de tudo a

amizade que levo para sempre, a equipe da ESEC (Pablito, Maricotinha, Bibinha,

Léo, Dan, Marcô, Jura), vocês são especiais na minha vida!

Aos meus co-autores na vida pessoal e profissional (Fabiana Araújo – Bibinha,

Maria Marcolina Cardoso – Marcô, Fabíola Dantas – Fafá, Mariana Amaral –

Maricota, Pablo Rubim – Pablito, Danhyelton Dantas – Dan), vocês foram cruciais

em vários momentos da minha vida acadêmica como um todo (graduação,

mestrado e doutorado). Agraço infinitamente a todos vocês pelo apoio, força,

amizade e parceria que sempre estiveram prontamente dispostos a me oferecer.

Amo todos vocês, meu infinito OBRIGADA!!!!!!!

Agradeço ao meu amado estratosférico, Bruno Wanderley, sem você certamente

seria muito difícil chegar ao final desta etapa. Sou grata, meu amigo, por tudo!

Agraço à Mister Edson Santana, por todo o suporte técnico em nossas coletas de

campo.

Agradeço a todos os componentes do Laboratório de Limnologia – DOL, os

componentes do Laboratório de Ecologia Aquática – LEA e aos componentes do

LARHISA. Meu infinito obrigada, por todo o suporte e ajuda que me ofereceram

ao longo desses 4 anos de trabalho.

Ao meu orientador, não tenho palavras para expressar o que sinto, tamanha a

gratidão e o respeito que aprendi a ter por você ao longo desses anos que tivemos a

oportunidade de trabalharmos juntos, André. Ao longo desse processo

amadurecemos juntos e quero dizer que sou muito orgulhosa em tê-lo como meu

orientador. Em sentir que você acredita em mim, mesmo em momentos em que eu

6

mesmo não acreditava. Se hoje eu cheguei a etapa final desta jornada, devo o

infinito à você. Saiba que você mora em meu coração. Obrigada, André!

Agradeço ao professor Coca (José Luis de Attayde), você sabe que foi através da

sua empolgação que descobri e me encantei pela limnologia. Sou eternamente

grata por você um dia ter confiado em mim e me dado a oportunidade de começar

a trilhar este caminho.

Agradeço à professora Vanessa Becker, talvez você não saiba, mas foi muito

importante no processo de construção deste trabalho. Você é uma das inspirações

que levarei para a minha vida profissional. Muito obrigada Vanessa, por também

acreditar em mim, me inspirar. Grata por tudo, minha querida.

Gostaria de Agradecer à banca de qualificação, pelas valiosas sugestões: Prof. Dr.

José Luiz de Attayde e o Prof. Dr. Paulo Abreu.

À banca de defesa do doutorado, por terem aceitado o convite: Prof. Dr. Hugo

Sarmento (UFScar), Prof. Dr. Vinicius Farjalla (UFRJ), Prof. Dr. José Luiz de

Attayde (UFRN), Prof. Dr. Vanessa Bécker (UFRN).

À Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

(CAPES/REUNI) pela bolsa de estudos concedida durante todo o período do

doutorado.

À Pós Graduação de Ecologia – UFRN e ao Departamento de Oceanografia e

Limnologia – DOL.

7

Agradeço à todos que passaram na minha vida, e alegraram o meu dia. Pelas

experiências trocadas, pelos momentos ímpares (alegres e tristes) vividos ao longo

desses anos.

Meu muito obrigada à todos que contribuíram direta ou indiretamente para a

concretização deste trabalho!

Meu muito obrigada!

8

“Cada um tem o seu ritmo, e no seu ritmo você é perfeito”

Gabi de Moura

Chuva no Sertão

“Terra árida, pingos de fogo a se alastrar... pelo céu, a nuvem cinza a impregnar

uma nova fase que do céu cairá... densas nuvens que de poucas se acumulam e

explodirão num espetáculo único de ressurgimento, de esperança, de renovação...

finalmente... é a chuva no sertão!!”

Gabi de Moura

9

DEDICO ESTA TESE AOS MEUS PAIS!

Assim na terra como no céu!

10

SUMÁRIO:

Resumo .......................................................................................................................... 17

Abstract ......................................................................................................................... 18

Introdução Geral .......................................................................................................... 19

CAPÍTULO I ................................................................................................................ 25

Contrasting patterns of CO2 emission in two small eutrophic reservoirs in the

tropical semiarid ........................................................................................................... 26

Introduction ................................................................................................................ 28

Material and Methods ................................................................................................. 30

Sampling ..................................................................................................................... 30

Results ......................................................................................................................... 33

Discussion ................................................................................................................... 35

Reference .................................................................................................................... 38

CAPÍTULO II ............................................................................................................... 59

Benthivorous fish increase CO2 emission in a shallow semiarid eutrophic reservoir.

........................................................................................................................................ 60

Introduction ................................................................................................................ 62


Results ......................................................................................................................... 67

Discussion ................................................................................................................... 68

Reference .................................................................................................................... 72

CAPÍTULO III ............................................................................................................. 83

Effects of the omnivorous fish Nile tilapia on the CO2 emission in eutrophic lakes84

Introduction ................................................................................................................ 86


Results ......................................................................................................................... 94

Discussion ................................................................................................................... 95

Reference .................................................................................................................... 99

Considerações Finais: ................................................................................................. 110

11

Reference ..................................................................................................................... 111

12

Lista de Figuras:

Capítulo I:

Figure 1: CO2 concentration during two years of monitoring in the ESEC reservoir and

FARM reservoir (September – 2012 to September – 2014). The black bars correspond

to the CO2 concentration in the ESEC reservoir. The white bars correspond to the CO2

concentration in the FARM reservoir. The red line correspond to the equilibrium

atmosphere concentration (CO2 = 390 µATM ). When the CO2 concentration > 390

µATM, the environment is emmiting CO2 to the atmosphere. When the CO2 < 390

µATM, the environment is synk by atmosphere.

Figure 2: Principal Component Analyse (PCA) scores of 18 environmental variables in

the ESEC reservoir (E – dark circles) and FARM reservoir (F – open circles).

Environmental variables monitored during two years (from September - 2012 to

September – 2014), particulate organic carbon (POC), dissolved nitrogen total (DN),

total nitrogen (TN), total phosphorus (TP), soluble reactive phosphorus (SRP), gross

primary production (GPP), ecosystems respiration ( R ), carbon dioxide concentration

(CO2), bacterial respiration (BR), planktonic respiration (PR), dissolved oxygen (DO),

temperature (TEMP), suspended fixed solids - inorganic (SFS), suspended volatile

solids - organic (SVS), maximum depth (zMAX), transparence of Secchi (secchi), PH.

Figure 3: Monitoring of limnologicalvariables during two years (from September of

2012 to September of 2014) of two reservoirs inserted in the semiarid region of

northeastern Brazil. The dark bar correspond the values of ESEC reservoir. The white

bar correspond the values of FARM reservoir. Insert in this figure are: (a) gross primary

production – GPP (µmol.h-1

.d-1

), (b) net ecosystem production – NEP (µmol.h-1

.d-1

), (c)

ecosystem respiration – R (µmol.h-1

.d-1

) and (d) Chlorophyll – a – Chla (µg.L-1

).

Figure 4: Suspended solids (mg.L-1

) concentration during two years of monitoring in the

ESEC reservoir and FARM reservoir (September – 2012 to September – 2014). The

black bars correspond to the SFS (suspended inorganic solids) and white bars

correspond to the SVS (suspended organic solids). (a) SFS and SVS concentration in

the ESEC reservoir. (b) SFS and SVS concentration in the FARM reservoir.

13

Figure 5: Monitoring of variables during two years (from September of 2012 to

September of 2014) of two reservoirs inserted in the semiarid region of northeastern

Brazil. The dark line correspond the values of ESEC reservoir. The gray line correspond

the values of FARM reservoir. The variables insert in this figure are: (a) bacterial

respiration – BR (µmol.O2.h-1

), (b) planktonic respiration – PR (µmol.O2.h-1

).

Figure 6: BR:PR ratio in the ESEC reservoir and FARM reservoir during two years of

monitoring (September – 2012 to September – 2014). The mid black of the bars

correspond to the ESEC values. The mid white of the bars correspond to the FARM

values.

Figure 7: (a) zMax (m-1

) and (b) transparency of secchi – secchi (m-1

), during two years

of monitoring in the ESEC reservoir and FARM reservoir (September – 2012 to

September – 2014). The dark line correspond the values of ESEC reservoir. The gray

line correspond the values of FARM reservoir.

Capítulo II

Figure 1: Schematic representation of experimental design and the mechanism accessed

in this study.

Figure 2: Effect size (mean and confidence interval) of benthivorous fish with and

without access to the sediment over response variables. A) CO2; B) Chlorophyll-a; C)

Bacterial respiration; D) Planktonic respiration; E) Total organ carbon; F) Total

nitrogen; G) Total phosphorous; H) C:N ratio ; I) C:P ratio; J) N:P ratio; K) Water

transparency; L) Methane (CH4); M) Dissolved oxygen; N) Bacterial abundance; O);

Flagellate abundance.

Figure 3: Mean values (±standard deviation) of response variables of the treatments and

reservoir during the experiment. A) CO2; B) Chlorophyll-a; C) Bacterial respiration; D)

Planktonic respiration; E) Total organ carbon; F) Total nitrogen; G) Total phosphorous;

H) C:N ratio ; I) C:P ratio; J) N:P ratio; K) Secchi; L) Methane (CH4) – (Day 30); M)

Dissolved oxygen (Day 0; Day 30); N) Bacterial abundance; O); Flagellate abundance

(Day 0; Day 30). Gray line on CO2 graph means 390 µATM (boundary between

supersaturation and undersaturation).

14

Figure 4: A) Effect size (mean and confidence interval) of benthivorous fish with and

without access to the sediment over response CO2 variable; B) Mean values (±standard

deviation) of response variable of the treatments and reservoir during the experiment of

Total Zoo – (Day 30);

Capítulo III


in this study.

Figure 2: Effect size (mean and confidence interval) of fish with and without access to

the sediment over response variables. Treatments legends: effect of fish without

sediment access ( ); effect of fish with sediment access ( ). A) pCO2; B) Chl-a; C)

Bacterial respiration; D) Planktonic respiration; E) Total Organ Carbon; F) Total

nitrogen; G) Total phosphorous; H) Secchi depth; I) C:N ratio; J) C:P ratio; K) N:P

ratio; L) Dissolved oxygen; M) Bacterial abundance; N) Flagellate abundance; O)

SFS;P)SVS.


reservoir during the experiment. Treatments legends: without sediment access and

without fish ( ), without sediment access but with fish ( ), sediment access without

fish ( ), sediment access with fish ( ), and reservoir ( ). A) pCO2; B) Chl-a; C)



ratio; L) Dissolved oxygen; M) Bacterial abundance; N); Flagellate abundance O) SFS.

Figure 4: A) Effect size (mean and confidence interval) of fish with and without access

to the sediment over response of Total Zoo variable; B) Mean values (±standard

15

deviation) of response Total Zoo variable of the treatments and reservoir during the

experiment.

Lista de Tabelas:

Capítulo I:

Table 1: Mean, standard deviation, minimum, median and maximum of dissolved

organic carbon (DOC), particulate organic carbon (POC), dissolved nytrogen (DN),

total nytrogen (TN), total phosphorus (TP), soluble reactive phosphorus (SRP), gross

primary production (GPP), net ecosystem production (NEP), ecosystem respiration (R),

partial pressure of carbon dioxide (pCO2), bacterial respiration (BR), total plankton

respiration (PR), dissolved oxygen (DO), temperature (TEMP), Chlorophyll - a

concentration (Chla), suspended fixed solids (SFS), suspended volatile solids (SVS),

maximum depth (zMAX), secchi depth (secchi), pH in ESEC reservoir.








maximum depth (zMAX), secchi depth (secchi), pH in FARM reservoir.

Table 3: Correlation coefficients between environmental variables and the first

components PCA axes.

Table 4: Test t for independent samples (groups), to test the differences between

variables of ESEC reservoir and FARM reservoir during two years of monitoring. The

variables were: dissolved organic carbon (DOC), particulate organic carbon (POC),

dissolved nytrogen (DN), total nytrogen (TN), total phosphorus (TP), soluble reactive

phosphorus (SRP), gross primary production (GPP), net ecosystem production (NEP),

ecosystem respiration (R), partial pressure of carbon dioxide (pCO2), bacterial

respiration (BR), total plankton respiration (PR), dissolved oxygen (DO), temperature

(TEMP), Chlorophyll - a concentration (Chla), suspended fixed solids (SFS), suspended

16

volatile solids (SVS), maximum depth (zMAX), secchi depth (secchi), pH. Inside the

table are t-value, degrees of freedom (df) and p-value (p).

Table 5: Simple regression between dependent variable CO2 and independent variables

( SFS, GPP, BR TP) separeted that showed significant effect (p<0.1) to explained the

variance of CO2 at ESEC reservoir.


(SVS, R, PR, SRP, zMAX) separeted that showed significant effect (p<0.1) to explained

the variance of CO2 at FARM reservoir.

Capítulo II:

Table 1: Results of two – way ANOVA testing the effect of fish (F), access to sediment

(A) and its interaction (A x F) over the mean of studied variables (Days 15 and 30).

Capítulo III:

Table 1: Results of two – way ANOVA to test the effect of tilapia fish (F), access to

sediment (A) and the interaction (A x F) over the mean of variables (Days 15 and 30).

17

Resumo

O objetivo desta tese é compreender os fatores que influenciam o balanço de

carbono em reservatórios do semiárido do nordeste brasileiro e avaliar o efeito de peixes

com diferentes hábitos alimentares no balanço de carbono destes ambientes. Os

resultados desta tese nos mostraram que peixes com diferentes hábitos alimentares

podem influenciar o balanço de carbono em reservatórios (Capítulo I; Capítulo II;

Capitulo III). Demonstramos através de experimentos de mesocosmos que peixes

bentívoros (detritívoros) aumentam a heterotrofia e emissão de CO2 para atmosfera,

através da ressuspensão de matéria orgânica e nutrientes presos ao sedimento, que

estimulam as taxas de respiração planctônica e microbiana, assim como os processos de

metanotrofia (Capítulo II). Por outro lado, peixes onívoros como a Tilápia do Nilo,

favorecem a diminução da emissão de CO2 para a atmosfera, através do estímulo da

biomassa fitoplanctônica ocasionado principalmente via cascata trófica pela diminuição

da biomassa de zooplâncton (Capítulo III). Além disso, reservatórios que apresentam

uma dominância de sólidos inorgânicos em suspensão pode indicar que o ambiente está

emitindo CO2 para a atmosfera. Em contrapartida, reservatórios que apresentam uma

dominância de sólidos orgânicos em suspensão pode indicar que o ambiente esteja

apreendendo CO2 da atmosfera (Capítulo I). Podemos concluir, que alguns fatores como

a dominância de sólidos em suspensão pode ser um indicativo da função do ecossistema

aquático frente ao balanço de carbono. Além disso, peixes com diferentes hábitos

alimentares podem influenciar o balanço de carbono em reservatórios.

18

Abstract

The objective of this thesis is to understand the factors that influence the carbon

balance in semi-arid reservoirs in northeastern Brazil and to evaluate the effect of the

feeding characteristics in the carbon balance of these environments. The results of this

thesis have shown that fish with different feeding characteristics can influence carbon

balance in reservoirs (Chapter I, Chapter II, Chapter III). We have demonstrated

through experiments mesocosms that benthivorous fish enhance heterotrophic and

emission of CO2 to the atmosphere, through the suspension of organic matter and

nutrients attached to sediment, which stimulate plankton and microbial respiration rates

as well as methanotrophy processes (Chapter II). On the other hand, omnivorous fish

like Nile Tilapia favor decrease CO2 emissions to the atmosphere by stimulating

phytoplankton biomass mainly caused via trophic cascade by decrease in zooplankton

biomass (Chapter III). In addition, reservoirs which have a predominance of inorganic

suspended solids may indicate that the environment is emitting CO2 to the atmosphere.

However, reservoirs have a dominance of organic suspended solids may indicate that

the environment is uptake CO2 by the atmosphere (Chapter I). We can conclude, that

some factors such as the dominance of suspended solids may be indicative of the of

carbon balance function by aquatic ecosystem. In addition, fish with different feeding

characteristics can influence the carbon balance in reservoirs.

19

Introdução Geral

Nos ecossistemas aquáticos continentais de maneira geral, o dióxido de carbono

(CO2) é captado pelos produtores primários através da fotossíntese, sendo então

reduzido a moléculas orgânicas (e.g. glicose) e incorporado à biomassa vegetal. Ao

utilizar a energia estocada na biomassa, produtores primários e, os consumidores,

oxidam moléculas orgânicas resultando na formação de CO2 pelo processo conhecido

como respiração (remineralização). Desta forma, o carbono pode retornar a atmosfera

ou ser reutilizado pelos produtores primários (Wetzel, 2001). Os ambientes aquáticos

continentais funcionam desta maneira, como importantes atores no ciclo global do

carbono e podem funcionar como emissores ou apreensores de dióxido de carbono

(CO2) da atmosfera (Cole et al., 2007). Quando as taxas de respiração são maiores que

as taxas de produção primária, o ambiente pode funcionar como emissor de CO2

(heterotrófico). Entretanto, quando as taxas de produção primária forem maiores do que

as taxas de respiração do sistema, os ambientes aquáticos podem funcionar como

apreensores de CO2 (autotrófico) da atmosfera (Cole et al., 2000).

Ao longo das últimas duas décadas diversos estudos vem classificando a maioria

dos ecossistemas aquáticos continentais como heterotróficos, i.e. emissores de dióxido

de carbono (CO2) para a atmosfera (e.g. Cole et al., 1994, Duarte and Prairie 2005,

Marotta et al. 2009). A esse padrão vem sendo atribuído, principalmente à entrada de

matéria orgânica alóctone na forma de carbono orgânico dissolvido (Cole et al., 1994;

Duarte and Prairie, 2005).

Alguns fatores podem afetar o balanço de carbono nos ambientes aquáticos

continentais. Por exemplo, a entrada de nutrientes (e.g. esgoto, agricultura, bacia de

drenagem) pode estimular os produtores primários (Smith and Schindler,2009), e com

20

isso aumentar a absorção de carbono da atmosfera (Pacheco et al., 2013). Ou ainda

estimular os organismos heterotróficos e aumentar as taxas de decomposição do sistema

(Cotner et al., 2000; Smith and Kemp, 2003). A Temperatura é um importante fator

apontado por afetar o balanço de carbono nos ambientes aquáticos. Através do aumento

da temperatura as taxas metabólicas dos organismos aumentam e as taxas de respiração

pelos organismos consumidores aumentam proporcionalmente mais que as taxas de

produção de biomassa (Amado et al., 2013). Com isso as concentrações de CO2 no

sistema também tendem a aumentar (Kosten et al., 2010). Outro importante fator

apontado por afetar o balanço de carbono é o carbono orgânico dissolvido (COD). A

entrada de carbono dissolvido estimula as taxas de degradação da matéria orgânica e

com isso estimula as taxas de produção/emissão do CO2 (Cole et al., 1994; Duarte and

Prairie, 2005). A estrutura da cadeia alimentar também pode influenciar no balanço final

de carbono. Quando num ecossistema aquático existe a predominância de um peixe

planctívoro, por exemplo, a biomassa de zooplâncton é reduzida e a biomassa dos

produtores primários aumenta, consequentemente. Com isso, podemos verificar uma

redução na emissão de CO2 neste ambiente (Schindler et al.,1997).

Nas regiões tropicais, a maioria dos ambientes aquáticos também funcionam

como emissores de dióxido de carbono (CO2) para a atmosfera (Richey et al., 2002;

Marotta et al., 2009; Barros et al., 2011). Além disso, apresentam uma distinta

variabilidade (valores extremamente baixos e altos), maior que a observada em

ambientes temperados (Marotta et al., 2009). No entanto, os padrões que regem o

balanço de carbono em ambientes tropicais ainda não estão claros. Alguns fatores são

indicados como os prováveis responsáveis pelo padrão heterotrófico, mas o principal é a

elevada temperatura encontrada nos trópicos (Marotta et al., 2009; Kosten et al., 2010).

21

Recentemente trabalhos realizados no semiárido do nordeste do Brasil destacaram que

muitos dos ambientes aquáticos continentais apesar de eutrofizados, comportam-se

como ecossistemas emissores de CO2 para atmosfera, sendo que apenas poucos

ambientes analisados tenham se comportado como apreensores de CO2 (Junger et

al.,2015; Dantas et al.,2015).

A região semiárida do nordeste brasileiro apresenta algumas características

marcantes, assim como os reservatórios e lagos inseridos nesta região, os quais sofrem

influência das características do clima e apresentam alto tempo de residência, acúmulo e

alta concentração de nutrientes, fatores que favorecem a eutrofização da maioria dos

reservatórios e lagos (Bouvy et al., 2000; Barbosa et al., 2012). O acúmulo de nutrientes

na maioria das vezes ocorre pela grande erosão dos solos, entrada de esgotos de áreas

urbanas e inadequado uso e ocupação do solo. A precipitação anual ocorre entre 400 e

800 mm, com a estação chuvosa compreendida entre os meses de janeiro a julho, e a

estação seca entre os meses de agosto a dezembro. Além disso, como consequência da

escassez de chuva, das altas taxas de evapotranspiração e do contínuo consumo da água

nestes ambientes aquáticos, o nível d`água na maioria dos meses é baixo, caracterizando

a maioria destes ambientes como lagos rasos (profundidade máxima de 5m) (Barbosa et

al., 2012). Devido à baixa profundidade existe uma grande conexão entre coluna d’água

e sedimento o que pode tornar esses ambientes bastante túrbidos pela resuspensão dos

sólidos através da ação dos ventos ou por organismos de hábito bentônico, eg.

macroinverterbrados e peixes (Scheffer, 2004; Freitas et al., 2011; Braga et al., 2015).

Os sólidos suspensos são compostos de material orgânico e inorgânico. A composição

dos sólidos orgânicos é basicamente alga, bactéria e outros componentes planctônicos.A

fração inorgânica é geralmente composta por areia, silte e argila (Billota and Brazier,

22

2008, Soeken et al., 2009). Os sólidos em suspensão modificam diretamente a

penetração de luz na água, afetando a estrutura da comunidade fitoplanctônica e

zooplanctônica, como também a dinâmica de produção primária e decomposição nos

ecossistemas aquáticos afetando o balanço de carbono em lagos (Cotner et al., 2000;

Souza et al., 2008; Liu et al., 2011; Mendonça et al., 2014; Braga et al., 2015; Medeiros

et al., 2015).

A comunidade de peixes pode afetar o funcionamento dos ecossistemas através

da ressuspensão de sedimentos e também do seu hábito alimentar e, por conseguinte,

também pode afetar diretamente o balanço de carbono em ecossistemas aquáticos por

efeitos de cascata trófica (Schindler et al., 1997). A dominância de peixes piscívoros em

lagos temperados, mostrou-se capaz de aumentar a biomassa zooplanctônica e

consequentemente inibir a biomassa fitoplanctônica estimulando, por conseguinte a

emissão de CO2 para atmosfera. Por outro lado, a predominância de peixes

zooplanctívoros, mostrou-se capaz de aumentar a biomassa fitoplanctônica e reduzir a

emissão de CO2 para atmosfera (Schindler et al., 1997). A onivoria é um

comportamento dominante entre os peixes distribuídos em região tropical (Gonçález-

Bergonzoni et al., 2012). O onívoro se alimenta em mais de um nível trófico (Polis and

Strong, 1996) e pode enfraquecer as relações tróficas (Lazzaro et al., 1997). Por isso o

efeito de peixes onívoros no balanço final de CO2 pode não ser observado, pela ausência

de cascata trófica (Marotta et al., 2012). Apesar de existir uma boa compreensão do

efeito de peixes piscívoros e planctívoros no balanço de carbono em ambientes

aquáticos, não é bem compreendido o efeito de peixes bentívoros/detritívoros (e.g.

Prochilodus brevis) ou mesmo planctívoros/onívoros (e.g. Tilápia do Nilo) no balanço

de carbono em lagos e reservatórios. Através do processo de bioturbação peixes

23

bentívoros de hábito detritívoro podem liberar nutrientes no sistema, bem como matéria

orgânica estocada no sedimento e estimular a atividade heterotrófica no ambiente

aquático (Cotner et al., 2000; Jeppesen et al., 2010). Além disso, através da

ressuspensão de sedimento inorgânico pode inibir o crescimento fitoplâncton em

ecossistemas aquáticos (Wahl et al., 2011), como também liberar gases diretamente do

sedimento tais como CO2 e metano (CH4) (Figueireido-Barros et al., 2009). Logo, em

um lago/reservatório dominado por peixes bentívoros-detritívoros, o CO2 pode ser

afetado positivamente e a emissão de CO2 nesse ecossistema pode ser estimulada.

Por outro lado, a dominância por peixes onívoros como a tilápia do Nilo

(Oreochromis niloticus) pode aumentar o crescimento fitoplanctônico através do

consumo de zooplâncton (top-down), assim como através da excreção de nutrientes

(bottm-up) (Vanni et al., 1997; Starling et al., 2002; Lazarro et al., 2003; Domine et al.,

2009; Menezes et al., 2010). Contudo, a tilápia do Nilo também pode causar a

resuspensão do sedimento através do hábito alimentar detritívoro, ou mesmo pelo

cuidado parental no período de desova durante a construção de locas no sedimento

(Beveridge et al., 2000; Starling et al., 2002). Esse processo de bioturbação pode tanto

estimular o crescimento fitoplanctônico através da liberação de nutrientes estocados no

sedimento, ou inibir através da resuspensão de sólidos inorgânicos no sistema que

podem causar a redução da transparência da água (Vanni et al.1997; Gu et al., 2011;

Wahl et al., 2011). Por fim, a Tilápia é um peixe que altera as condições ambientais do

ecossistema onde é predominante e a sua presença pode estar relacionada à má

qualidade ambiental de reservatórios, contribuindo para a eutrofização destes ambientes

(Starling et al., 2002; Attayde et al., 2011). Desta maneira, a Tilápia pode afetar

24

positivamente o crescimento fitoplanctônico e consequentemente inibir a emissão de

CO2 para atmosfera.

Apesar de compreendermos as possibilidades do efeito da resuspensão de

sedimento em ecossistemas aquáticos tropicais (Xu et al., 2009), o efeito de peixes com

diferentes hábitos alimentares no balanço de carbono nesses ambientes ainda não foi

bem demonstrado e compreendido. Diante do exposto, o objetivo desta tese é

compreender os fatores que influenciam o balanço de carbono em reservatórios do

semiárido do nordeste brasileiro e avaliar o efeito de peixes com diferentes hábitos

alimentares no balanço de carbono destes ambientes.

25

CAPÍTULO I

26

Contrasting patterns of CO2 emission in two small eutrophic

reservoirs in the tropical semiarid

Caroline Gabriela Bezerra de Moura1, Maria Marcolina Cardoso, Mariana Rodrigues

Amaral da Costa2, Pablo Rubim

2, Fabíola Dantas, José Luiz de Attayde

2, Vanessa

Bécker3 and André Megali Amado

1.

1 – Departamento de Oceanografia e Limnologia

Pós Graduação em Ecologia

Universidade Federal do Rio Grande do Norte – UFRN - Brasil

2 - Departamento de Ecologia



3 – Departamento de Engenharia Civil

Pós Graduação em Engenharia Sanitária e Ambiental

Universidade Federal do Rio Grande do Norte – UFRN – Brasil

Corresponding author: [email protected]

Key – words: carbon balance, respiration rates, suspended solids.

27

Abstract

The aim of this study was investigate the dynamics of carbon dioxide (CO2) and

the relationship with suspended solids in two eutrophic reservoirs inserted in tropical

semi-arid. To meet this goal we conducted a monitoring of two years (September 2012-

to-September 2014) in two eutrophic reservoirs inserted in the semiarid region of

northeastern Brazil. We observed that the origins of sediment resuspension by

reservoirs in semiarid region can influence the carbon balance in these environments. In

the ESEC reservoir the prevalence of suspended inorganic solids explain the high

microbial metabolism and consequently the high CO2 emissions to the atmosphere. On

the other hand, in the FARM reservoir the prevalence of suspended organic solids

(phytoplankton biomass) explains the high primary production and consequently CO2

uptake from the atmosphere. Thus, we concluded that the origin of the suspended solids

is the main driver of these lakes metabolism. Finally, recalling to the high pCO2

variability in tropical (shallow) lakes, we suggest that local factors, such as physical

lakes characteristics or different fish communities composition, may drive the

ecosystem metabolisms confounding general or global trends as temperature gradient

effects on metabolism.

28

Introduction

The freshwater ecosystems have an important role in the global carbon cycle

acting as synk or source of carbon dioxide (CO2) to the atmosphere (Cole et al, 2007;

Tranvik et al., 2009). Through photosynthesis, the inorganic carbon is incorporated into

the biomass of primary producers (eg. phytoplankton, macrophytes), turning into

organic carbon. The organic carbon incorporated into the biomass of primary producers

can be transferred along the aquatic food web to the higher trophic levels (e.g.

heterotrophic organisms) (O`Sullivan and Reynolds, 2003). The heterotrophic

metabolism of the organisms return the inorganic carbon in the form of CO2 by

respiration to the aquatic ecosystem and then to the atmosphere. Finally, the refractory

part of the organic carbon can accumulate in the sediment, e.g. the detritus may

precipitate and be stored in the bottom of the ecosystems (O`Sullivan and Reynolds,

2003).

The aquatic ecosystem behave either as sink or source of CO2 to the atmosphere

depending on its dominant metabolic processes; ie. whether net ecosystem production

(NEP) is positive or negative. When gross primary production (GPP; which is the sum

of all primary production of the organisms in a given ecosystem), exceeds the net

ecosystem respiration (R; which is the results of all respiration rates from organisms

heterotrophic and autotrophic), the system turns to autotrophic (NEP > 0) state and it

acts as a sink of CO2. However, when the R exceeds GPP (NEP < 0) the system moves

to a heterotrophic state acting as a source of CO2 to the atmosphere (Cole et al., 2000).

Over the past two decades, most aquatic ecosystems have been classified as

heterotrophic (Cole et al., 1994; Del Giorgio et al., 1997; Rickey et al, 2002;. Duarte

and Prairie., 2005; Marotta et al, 2009; Kosten et al, 2010) because of the relevant entry

of allochthonous organic matter in the form of dissolved or particulate organic carbon

(DOC and POC) in these ecosystems (Cole et al., 1994;Duarte and Prairie, 2005). In the

tropics, most aquatic environments also act as CO2 source to the atmosphere on average

at a higher intensity than temperate environments (Marotta et al., 2009). Temperature is

known as an important factor regulating ecosystem metabolism. As in the tropics the

average temperature is high over the year, the high variability of CO2 concentrations in

tropical ecosystems (Marotta et al., 2009) suggest that local factors may have great

29

relevance to this process. For instace, the availability of nutrients and organic matter,

the trophic chain structure (e.g. top-down control), or landscape and lakes

characteristics, etc, have great impact in the aquatic metabolism (Kosten et al. 2010,

Amado et al., 2013, Hall et al., 2016 - In press). Recent work carried out in semi-arid

northeast of Brazil noted that many of the continental aquatic environments despite

eutrophic, act as CO2 source to the atmosphere, though few environments analyzed have

behaved as CO2 synk (Junger et al, 2015.; Dantas et al., 2015).

The semi-arid region of northeastern Brazil has important features such as high

spatial and temporal variability of precipitation, average temperatures above 25 °C,

occurance of temporary rivers and streams and distinct coverage of deciduous

vegetation, called Caatinga (Barbosa et al., 2012), that can directly affect the ecosytems

functioning and thus, metabolic rates. The low rainfall results in high residence times,

and high concentration of nutrients, that favor the shalowness and eutrophication of

most reservoirs and lakes (Bouvy et al., 2000, Barbosa et al. , 2012). Due to the shallow

nature of these Ecosystems, specially in dry periods, there is a great connection between

the water column and sediment through the action of winds or bioturbation, which can

make these environments quite turbid by resuspension of solids quite often (Scheffer.,

2004; Freitas et al., 2011; Braga et al., 2015; Costa et al., 2016 – in press).

In one hand, the predominance of organic suspended solids, which denote high

phytoplankton biomass in the semi-arid eutropphic ecosystems, would cause the

decrease of CO2 through consumption by primary producers (Carigan et al, 2000; Gu et

al, 2011). On the other hand, the suspension of inorganic solids from the sediment to the

water column can affect the light incidence in the water and, as a consequece, reducing

the phytoplankton primary production in lakes and reservoirs (Souza et al, 2008; Wahl

et al. 2011; Braga et al, 2015; Medeiros et al, 2015; Costa et al., 2016- – in press). As a

consequence, the predominance of inorganic solids cause shading, which may affect the

carbon balance favouring the CO2 emissions in these ecosystems (Cotner et al., 2000;

Mendonca et al, 2014). In addition, the prevalence of native bentivorous fish (e.g.

Prochilodus brevis) or planktivrous fish, e.g. Nile Tilapia, can influence phytoplankton

dynamics of the system Lazzarro et al. (2003), and the interaction sediment water

column and can affect the carbon balance of aquatic ecosystems causing the increase or

decrease in CO2 emissions into the atmosphere (Jeppesen et al, 2010; Gu et al, 2011).

30

However, the relationship between sediment resuspension and the carbon balance in

tropical aquatic ecosystems are not yet well understood (Xu et al., 2009).

In this study we investigate the dynamics of lake metabolism and the CO2

dynamics in the water column and the relationships with environmental characteristics

in two eutrophic reservoirs inserted in the semi-arid Brazilian. To achieve our goal, we

conducted a two years monthly monitoring study (September 2012 to September 2014)

in two eutrophic reservoirs inserted in the semiarid region of northeastern Brazil. Our

main results demonstrate that the origin of the suspended solids are key components to

the carbon balance in these shalow ecosystems.

Material and Methods

Study area

The current study was conducted in two small reservoirs (ESEC and FARM

reservoirs) built in the rivers Espinharas and Sabugi, in the Piranhas-Assu River basin,

located in the semi-arid region in the Municipality of Serra Negra do Norte, Rio Grande

do Norte State, northeastern of Brazil. The ESEC reservoir is a shallow (max. depth 4m)

and small (11ha) reservoir situated in a conservation unit named Seridó Ecological

Station (ESEC) (06°34'49,3″S; 37°15'20″W). In the beginning of monitoring the ESEC

reservoir was considered eutrophic and slightly heterotrophic and the chlorophyll-α

(Chlɑ), total phosphorus (TP), total nitrogen (TN) and concentration of carbon dioxide

(CO2) were respectively: 50 (µg.L-1

), 115 (µg.L-1

), 1810 (µg.L-1

) and 600 (µATM). The

FARM reservoir is a shallow (max. depth 6m) and small (11ha) reservoir located in the

Solidão farm (6º34’42,49’’S; 37º19’47,5” W), 20 kilometers far from the ESEC

reservoir. In the beginning of monitoring the FARM reservoir was considered eutrophic

and autotrophic the chlorophyll-α (Chlɑ), total phosphorus (TP), total nitrogen (TN) and

concentration of carbon dioxide (CO2) were respectively: 40 (µg.L-1

), 66,14 (µg.L-1

),

1530 (µg.L-1

) and 300 (µATM).

Sampling

Water samplings were performed from September of 2012 to September of

2014. At first, in each reservoir the secchi disk depth and the water temperature and

dissolved oxygen profile were measured in water column using a portable oxygen meter

(Instrutherm MO-900). Water samples were taken in the sub-surface at eight different

31

sampling stations along the limnetic and littoral zones of the reservoirs and integrated in

one sample in a plastic bucket (50L). In the field laboratory part of the samples were

filtered through glassfiber filter (1.2µm; VWR INTERNATIONAL). The unfiltered

water was used to estimate gross primary production (GPP), net ecosystem production

(NEP), ecosystem respiration (R), total plankton respiration (PR), bacterial abundance

(BA), heterotrophic nanoflagellates abundance (HNF), particulate organic carbon

(POC), total nitrogen (TN), total phosphorus (TP) and soluble reactive phosphorus

(SRP). The filtered water was used to estimate bacterial respiration (BR), dissolved

organic carbon (DOC) and dissolved nitrogen (DN). The filters were used to estimate

Chlorophyll - a concentration (Chla), suspended total solids (STS), suspended fixed

solids (SFS) and suspended volatile solids (SVS). For the partial pressure of CO2

(pCO2) estimation, water samples from sub-surface were taken with polycarbonate

bottle (100mL), which were overflown a few times until complete removal of internal

atmosphere or bubbles. Samples were then taken immediately to the field laboratory for

pH and alkalinity measurements (c.a. 30 min).

Analytical methods

The pCO2 was estimated from the pH and alkalinity through the acid titration

method using 0.02N H2SO4, adjusting for temperature, ionic strength and air pressure

(Cole et al., 1994). Subsequently, the results were compared to the atmosphere CO2

concentration (considered here as being 390 uATM) and classified as undersaturated

when lower than 390uATM or supersaturated with CO2 relative to the atmosphere when

above 390 uATM. We assumed as the pCO2 in equilibrium with the atmosphere as

being between 380 and 400 µATM according to the last five years from estimatives of

Mauna Loa Observatory.

PR rates were estimated as oxygen consumption in unfiltered water samples,

while BR rates were estimated as oxygen consumption in filtered (glass fiber; 1.2µm

average pore size; VWR INTERNATIONAL) water using a golden tip oxygen

microsensor connected to a picoammeter controlled by the MicOx software (Unisence

©; Briand et al., 2004). The samples for PR and BR measurements were incubated in

exetainers (5.9mL; Labco®) with no internal atmosphere in 5 replicates for each

mesocosm in the dark and room temperature (25⁰C ± 1) for 24 hours. The respiration

32

rates were transformed to the carbon basis using the conservative respiration quotient

(RQ) of 1. We are aware that the RQ can be very variable among different ecosystems

(Bergreen et al., 2011), but as our work was developed in ecosystems were the carbon

source is basically autochthonous planktonic organic matter, we believe that any artifact

would not affect the overall results of our work.

The GPP, NEP and R measurements were performed with unfiltered water

samples using the clear and dark bottles (300 mL winkler bottles) method (Wetzel and

Likens, 2000). Both clear and dark bottles were incubated (in five replicates each) for

24 hours in the sub-surface of the ESEC reservoir (10 cm deep) and the dissolved

oxygen were measured in the beginning and at the end of the incubation using an

oxygen microsensor connected to a picoamperimeter controlled by the MicOx software

(Unisence ©; Briand et al., 2004). NEP was calculated from the oxygen concentration

changes in the clear bottles while R was calculated from the oxygen concentration

changes in the dark bottles. GPP was estimated as the sum of NEP and the module of R

(Wetzel and Likens, 2000).

Chla concentrations were estimated from the 1,2µm glassfiber filters through the

95% ethanol extraction and spectrophotometer method (Jespersen and Cristoffersen,

1987).

The bacterioplankton abundance was estimated by flow cytometry in

glutaraldeheyde (final concentration 1%) preserved samples. The abundance was

determined after nucleic-acid staining with Syto13 (Molecular Probe) at 2.5 µM final

concentration (del Giorgio et al., 1996). Fluorescent latex beads (Polysciences, 1.5µm

diameter) were added to each sample for calibration of side scatter and green

fluorescence signals, and as an internal standard for the cytograms.

Nanoflagellates abundance was estimated on glutaraldehyde (final concentration

1%) fixed samples. 1 ml was stained with DAPI and then filtered through 0.6µm

polycarbonate black membrane (Nuclepore, diameter 25mm) and counted in an

epifluorescence microscope (Porter and Feig, 1980). On average 400 individuals were

counted in each sample, at a magnification of 1000x.

DOC and TN concentration were measured from the filtered water by the catalytic

combustion method in a Total Organic Analyzer (TOC – V, Shimadzu – 2.0) with a TN

33

analyzer attached (VNP module) and the POC in the SSM module. Total phosphorus

(TP) and soluble reactive phosphorus (SRP) measured from the unfiltered water with a

spectrophotometer by the acid ascorbic method after persulphate digestion (Murphy and

Riley,1962).

STS, SFS and SVS were determined by gravimetry after drying the filters

overnight at 100 ºC and ignition of filters at 500ºC for three hours (APHA, 2005). The

organic suspended solids were measured by the difference between total suspended

solids and inorganic suspended solids (APHA, 1998).

Statistical Analyses

Previously to the statistical analysis, the data matrix was inspected for the

presence of collinearity by calculating the Variance Inflection Factor (VIF). Variables

that showed VIF higer than 3 were removed from the data matrix through a manual

stepwise procedure according to the proposed by Zuur et al., (2010). All data satisfied

the homogeneity of variances and normality premises.

A principal components analysis (PCA) was performed with 17 environmental

variables dataset (CO2, GPP, R, PR, BR, DO, SFS, SVS, zMAX, Secchi, POC, DN, TN,

TP, SRP, PH, TEMP) to inspect data variation during the study period in the ESEC

reservoir and FARM reservoir. Previously, data were transformed to Z-Score

standardization to remove dimensionality and scale of the different variables.

A t test with independent samples was performed to test the differences between

each variables of ESEC reservoir and FARM reservoir during two years of monitoring.

Simple regressions were performed to examine the best set of predictor variables

affecting the CO2 in ESEC reservoir and FARM reservoir, separately.

Results

Overall we found that in most samplings (96% of months) along the two years of

monitoring the ESEC reservoir was supersaturated in CO2 (Figure 1). On the other hand,

we found that the in the FARM reservoir, in most samplings (76% of months) over the

two years, was undersaturated in CO2 undersaturated or close to the CO2 concentration

of atmosphere equilibrium (390 µATM) (Figure 1). The FARM reservoir and ESEC

reservoir showed high concentrations of nutrients (N and P) and Chla (Table 1; Table

34

2). The ESEC reservoir showed during the monitoring the rates of productivity lower

than FARM reservoir (Table 1; Table 2; Figure 3), high concentration of nutrients

(Table 1; Table 2), a predominance of inorganic suspended solids (SFS) (Table 1; Table

2; Figure 4), high rates of bacterial respiration (Figure 5; Figure 6) and showed a zMAX

lower than FARM reservoir (Table 1; Table 2; Figure 7). The FARM showed during the

study high rates of productivity (Table 2; Figure 3), high concentration of nutrients

(Table 2), showed a predominance of organic suspended solids (SVS) (Table 2; Figure

4), high rates of planktonic respiration (Table 2; Figure 5; Figure 6) and showed a

zMAX higher than ESEC reservoir (Table 1; Table 2; Figure 7).

Principles Components Analysis (PCA)

In PCA, the first two components together explained 58% of the total variance

of the environmental variables data. The first component explained 38.3% of the

variance of the data and the second component explained 20.4% of the variance.

The first component presented negative association with CO2 and, positive with

SVS, GPP, R, PR, POC, DN, TN. The second component presented positive association

with CO2, SFS, SRP, BR, TEMP and the zMAX showed negative association with this

axis (Table 3). Besides, TP showed a positive association, Secchi and ZEU showed

negative association with the second and the first components (Table 3). Most of

FARM data were widely distributed along of the first component (Figure 2), while the

most of ESEC data were widely distributed along of the second component (Figure 2).

T test

The t test between ESEC reservoir variables and FARM variables showed that

the DOC, POC, DN, TN, SRP, GPP, NEP, R, pCO2, BR, PR, Chla, SFS, SVS, zMAX,

pH was statistical different between each reservoir (p<0.05) (Table 4). The TP, DO,

TEMP and zMAX not showed statistical difference between reservoirs (Table 4).

Linear Regression

The results of simple regression with the ESEC reservoir data showed a positive

relationship between CO2 and fixed suspended solids – SFS (inorganic solids), BR, TP

and negative relationship with GPP (Table 5). The simple regression with the FARM

reservoir data showed a negative relationship between CO2 concentration and volatile

35

suspended solids – SVS (organic solids), R, PR and SRP. Furthemore, the zMAX

showed a positive relationship with CO2 (Table 6).

Discussion

Although the reservoirs investigated in the current study are very similar in

terms of size, depth, location and trophic state (both considered eutrophic according to

Thorton and Rast (1993) classification), they presented very distinct limnological

characteristics and metabolic behavior as well. The ESEC reservoir was mostly CO2

super-saturated being classified as net heterotrophic, while the FARM reservoir was

mostly CO2 under-saturated being classified as net autotrophic (Figure 1). Our group

previously registered in a seasonal study and in a lakes survey (N=100) that even

though most aquatic ecosystems in this semi-arid region are eutrophic, the heterotrophic

metabolism is predominant (Junger et al, 2015.; Dantas et al., 2015). Here we discuss

that this contrasting metabolic pattern between the two studied lakes is probably related

to the origin of the suspended solids, since the pCO2 in the ESEC lake was clearly

related to the SFS and to the SVS in the FARM lake (Table 4; Table 5; Table 6; Figure

4).

The FARM reservoir presented a clear pattern of autotrophic environment in

which CO2 concentrations were lower or equal, in the most months, of the 390µATM

(Figure 1) in agreement to the high GPP recorded (Table 2; Table 6; Figure 3). In

addition, we observed high concentration suspended organic solids (SVS), which is

related with the high autochthonous organic matter (e.g. planktonic primary producers,

high Chlorophill a) (Table 6; Figure 2) (Xu et al., 2009). The FARM reservoir showed

similar limnological characteristics with the most of reservoirs distributed in Brazilian

semiarid region, such as eutrophic conditions, mainly by high concentrations of

nutrients, and consequently high primary production (Bouvy et al., 2000; Barbosa et al.,

2012; Costa et al., 2016 – in press). However, the metabolism was autotrophic

contradicting our previous studies (Junger et al., 2015; Dantas et al., 2014), but in

agreement to some lakes from the literature, which were eutrophic and autotrophic

(Pacheco et al., 2013). It is important to notice though, that both reservoirs studied here

are smaller and shallower than the ones in our previous studies.

36

The ESEC reservoir was mostly heterotrophic during the study (Figure 1).

Despite the fact that this reservoir had high nutrients concentrations (Table 1), the Chla

concentrations were not as high as in the FARM reservoir (Table 1; Table 2; Figure 3).

The high concentrations of suspended inorganic solids (SFS) may have caused shading

in the water column reducing phytoplankton biomass (Table 1; Figure 3; Figure 4) and

consequently the primary production (Table 1; Figure 3) (Sousa et al., 2008; Freitas et

al., 2011; Braga et al., 2015; Costa et al., 2016 – in press). The resuspention of

sediments is the most probable explanation to the high inorganic turbudity in the ESEC

reservoir accordingly to theory of the shallow lakes (Scheffer et al., 2004). In fact, some

reservoirs located in the Brazilian semiarid region usually present high rates of turbidity

caused by sediment resuspension (Freitas et al., 2011; Braga et al., 2015; Costa et al.,

2016 – in press). Besides the lower primary production, the sediments ressuspension

releases nutrients (e.g. N and P) in the water column that under low light conditions

stimulate the microbial metabolism, especially bacterial respiration (Cotner et al., 2000;

Biddanda and Cotner, 2002; Liu et al., 2011). Accordingly we registered higher average

BR respiration in the ESEC reservoir when compared to the FARM reservoir (Table 1;

Table 2; Figure 5; Figure 6), which probably contributed to the higher pCO2 at the

ESEC (Table 5; Figure 1; Figure 6) in agreement with the positive correlation of BR

and pCO2 in the ESEC reservoir (p = 0.07; Table 5). Considering the importance of the

sediment to the total CO2 production in lakes, especially the shallow ones (Jonsson et

al., 2001; Sobek et al., 2005), we assumed the p-value of 0.07 as a significant

relationship between BR and CO2.

The wind action could be an important factor causing sediments ressuspension in

shallow lakes (Scheffer et al., 2004) and thus, one could argue that the high inorganic

turbidity in the ESEC reservoir was caused by the sharp depth decrease (Table 1; Figure

4; Figure 7), and strong wind action. However, we believe that the wind action is not

likely the only explanation for the case here since the FARM reservoir, which is only a

few kilometers apart from the ESEC reservoir, did not present this high inorganic

turbidity even thoug both had similar depths mainly at the and of our samplings (Table

1; Figure 4). One possible explanation for the high turbidity (resuspended sediments)

and the consequent high CO2 emission in the ESEC reservoir may rely on the fish

community composition (Wahl et al., 2011). The predominance of benthivorous fish in

37

shallow lakes may both facilitate the wind ressuspension of the sediment, because they

reduce the erosion resistence of the sediment (Scheffer et al., 2003), and cause high

direct resupension when searching for over-exploited prey in the sediments (Zambrano

et al., 2001; Jeppesen et al., 2010; Wahl et al., 2011; Jeppesen et al., 2014).

In March 2013, a massive fish removal (c.a. 7 tons of fishes) was performed in

the ESEC reservoir and showed that the fish community was dominated (almost 6 tons

out of 7 tons) by bentivorous fish, especially the Prochilodus brevis (localy known as

Curimatã). Thus, in agreement to the literature, this community domination by the P.

brevis could explain, at least in part, the predominance of inorganic solids in this

reservoir. Unfortunatelly, we do not have consistent data on the fish community

composition in the FARM reservoir. However, as the wind action should be similar

between two studied reervoirs, we could rule out this factor. Thus, it is plausible that the

FARM reservoir has a fish community with different composition than ESEC (e.g.

planktivorous fish – Nile Tilapia), which would help explain the low suspended

sediments and the prevalence of autotrophy along of the studied months (Gu et al.,

2011). We present strong evidences in the later chapters of this thesis that the

dominance of fishes with these two different habits could lead to contrasting carbon

balances (Chapters 2 and 3).

This work becomes important because it shows that the origin of suspended solids

(either sediment resuspension or phytoplankton particulate matter) in the reservoirs in

semiarid region can influence the carbon balance in these environments. In the ESEC

reservoir the prevalence of suspended inorganic solids explain the high microbial

metabolism and consequently the high CO2 emissions to the atmosphere. On the other

hand, in the FARM reservoir the prevalence of suspended organic solids (phytoplankton

biomass) explains the high primary production and consequently CO2 uptake from the

atmosphere. Thus, we concluded that the origin of the suspended solids is the main

driver of these lakes metabolism. Finally, recalling to the high pCO2 variability in

tropical (shallow) lakes (Marotta et al. 2009), we suggest that local factors, such as

physical lakes characteristics or different fish communities composition, may drive the

ecosystem metabolisms puzzling general or global trends as temperature gradient effects

on metabolism.

38

Reference

Amado, A.M., Meirelles-Pereira, F., Vidal, L.O., Sarmento, H., Suhett, A.L., Farjalla,

V.F., Cotner, J.B. and Roland, F., 2013, “Tropical freshwater ecosystems have lower

bacterial growth efficiency than temperate ones”, Frontiers in Microbiology, v.4, pp.1-

8.

APHA, 1998. Standard Methods for the Examination of Water and Wastewater, 20th

edn. American 348 Public Health Association, Washington DC.

Barbosa, J.E.L., Medeiros, E.S.F., Brasil, J., Cordeiro, R.S., Crispim, M.C.B. and da

Silva, G.H.G., 2012, “Aquatic systems in semiarid Brazil: limnology and management”.

Acta Limnologica Brasiliensia.

Bergreen, M., Lapierre, J.F. and Del Giorgio, P., 2011, “Magnitude and regulation of

bacterioplankton respiratory quotient across freshwater environmental gradients”. The

ISME Journal, pp. 1-10.

Biddanda, B. and Cotner, J., 2002, “Small players, large role: microbial influence on

biogeochemical processes in pelagic aquatic ecosystems”. Ecosystems, v.5, pp. 105 –

121.

Biddanda, B., Ogdahl, M. and Cotner, J., 2001, “Dominance of bacterial metabolism in

oligotrophic relative to eutrophic waters”. Limnology and oceanography, v.46, pp. 730

– 739.

Braga, G.G., Becker, V., de Oliveira, J.N.P., Mendonça Júnior, J.R., Bezerra, A.F.M.,

Torres, L.M., Galvão, A.M.F. and Mattos, A., 2015, “Influence of extended drought on

water quality in tropical reservoirs in a semiarid region”. Acta Limnologica Brasiliensia,

v.27, pp. 15 – 23.

Briand, E., Pringault, O., Jacquet, S. and Torréton, J.P., 2004, “The use of oxygen

microprobes to measure bacterial respiration for determining bacterioplankton growth

efficiency”, Limnology and Oceanography: Methods, v.2, pp.406-416.

Bouvy, M., Falcão, D., Marinho, M., Pagano, M. and Moura, A., 2000, “Ocurrence of

Cylindrospermopsis (Cyanobacteria) in 39 Brazilian tropical reservoirs during the 1998

drought”. Aquatic Microbial Ecology, v. 23, pp. 13-27.

39

Carignan, R., Planas, D. and Vis, Chantal, 2000, “Planktonic production and respiration

in oligotrophic shield lakes”. Limnology and Oceanography, v.45, pp. 189-199.

Cole, J.J., Pace, M.L., Carpenter, S.R. and Kitchell, J.F., 2000, “Persistence of net

heterotrophy in lakes during nutrient addition and food web manipulations”. Limnology

and oceanography, v.45, pp. 1718-1730.

Cole, J.J., Caraco, N.F., Kling, G.W. and Kratz, T.K., 1994, “Carbon dioxide

supersaturation in the surface waters of lakes”, Science, v. 265, pp. 1568-1570.

Cole, J. J., Prairie, Y. T., Caraco, N. F., Mcdowell, W. H., Tranvik, L. J., Striegl, R. G.

et al., 2007, ”Plumbing the global carbon cycle:Integrating inland waters into the

terrestrial carbon budget”. Ecosystems, v. 10, pp. 171-184.

Costa., M.R.A., Attayde, J.L. and Becker, 2016, “Effects of water level reduction on the

dynamics of phytoplankyon functional groups in tropical semi-arid shallow lakes”.

Hydrobiologia. – In press.

Cotner, J.B., 2000, “Intense winter heterotrophic production stimulated by benthic

resuspension”. Limnology and Oceanography, v. 45, pp. 1672-1676.

Dantas, F.C. and Amado, A.M., 2015, “Saturação em CO2 e regulação metabólica do

bacterioplâncton em ecossistemas aquáticos de baixa latitude”. In Portuguese.

Universidade Federal do Rio Grande do Norte.

Moura, C.G.B., 2015, “Mecanismos de emissão de CO2 em reservatórios do semiárido

brasileiro”. Capítulo 2, Thesis.

del Giorgio, P.A., Bird, D.F., Prairie, Y.T. and Planas, D., 1996, “Flow cytometric

determination of bacterial abundance in lake plankton with the green nucleic acid stain

SYTO 13”. Limnology and Oceanography, v. 41, pp. 783-789.

del Giorgio, P.A., Cole, J.J. and Cimbleris, A., 1997, “Respiration rates in bacteria

exceed phytoplankton production in unproductive aquatic systems”. Nature, v.385, pp.

148-151.

Duarte, C.M. and Prairie, Y.T., 2005, “Prevalence of heterotrophy and atmospheric CO2

emissions from aquatic ecosystems”, Ecosystems, v.8, pp. 862-870.

40

Fenchel, T. and Finlay, B.J., 1995, “Ecology and evolution in anoxic worlds”. Oxford

University Press, 290p.

Freitas, F.R.S., Rhighetto, A.M. and Attayde, J.L., 2011, “Cargas de fósforo total e

material em suspensão em um reservatório do semiárido brasileiro”. Oecologia

Australis, v. 15, pp. 655 – 665.

Gu, B., Schelske, C.L. and Coveney, M.F., 2011, “Low carbon dioxide partial pressure

in a productive subtropical lake”. Aquatic Science, v. 73, pp. 317-330.

Hall., E.K., Schoolmaster Jr., D.R., Amado, A.M., Stets, E.G., Lennon, J.T., Domine, L.

and Cotner, J.B., 2016, “Scaling relationships among drivers of aquatic respiration in

temperate lakes: from the smallest to the largest freshwater ecosystems”. Inland Waters.

In Press.

Jespersen, A.M. and Christoffersen, K., 1987, “Measurements of chlorophylla from

phytoplankton using ethanol as extraction solvent”. Archiv fuer Hydrobiologie

AHYBA4, v.109, pp. 445-454.

Jeppesen, E., Meerhoff, M., Holmgren, K., et al., 2010, “Impacts of climate warming on

lake fish community structure and potential effects on ecosystems function”.

Hydrobiologia, v. 646, pp. 73-90.

Jeppesen, E., Brucet, S., Naselli-Flores, L., Papastergiadou, E., Stefanidis, K., Nõges,

T., Nõges, P., Attayde, J.L., Zohary, T., Coppens, J., Bucak, T., Menezes, R.F., Freitas,

F.R.S., Kernan, M., Sondergaard, M. And Beklioglu, M., 2014, “Ecological impacts of

global warming and water abstraction on lakes and reservoirs due to changes in water

level and related changes in salinity”. DOI: 10.1007/s10750-014-2169-x.

Jeppesen, E., Jensen, J.P., Sondergaard, M., Lauridsen, T. and Landkildehus, F., 2000,

“Trophic structure, species richness and biodiversity in Danish lakes: changes along a

phosphorus gradient”. Frershwater Biology, v. 45, pp. 201-218.

Jonsson, A., Meili, M., Bergstrom, A.K. and Jansson, M., 2001, “Whole-lake

mineralization of allochthonous and autochthonous organic carbon in a large humic lake

(Ortrasket, N.Sweden)”. v. 46, pp. 1691-1700.

41

Junger, P.C., Terra, I., Caliman,, A., Carneiro, L.S., Becker, V. and Amado, A.M.,

2015, “Tropical eutrophic semiarid reservoirs are net heterotrophic: Wet season drives

higuer CO2 emissions”. In Portuguese. Universidade Federal do Rio Grande do Norte.

Kosten, S., Roland, F., Da Motta Marques, D.M.L., Van Nes, E.H., Mazzeo, N.,

Sternberg, L.S.L., Scheffer, M. and Cole, J.J., 2010, “Climate-dependent CO2

emissions from lakes”. Golbal Biogeochemical Cycles, v. 24. GB2007.

Lazzaro, X., Bouvy, M., Ribeiro-Filho, R.A., Oliveira, V.S., Sales, L.T., Vasconcelos,

A.R.M. and Mata, M.R., 2003, “Do fish regulate phytoplankton in shallow eutrophic

Northeast Brazilian reservoirs?”. Freshwater Biology, v. 48, pp. 649-668.

Liu, X., Wu, Q., Chen, Y. and Dokulil, M.T., 2011, “Imbalance of plankton community

metabolism in eutrophic lake Taihu, China”. Journal of Great lakes Research, v.37, pp.

650 – 655.

Marotta H., Duarte, C.M., Sobek, S. and Enrich-Prast, A., 2009, “Large CO2

disequilibria in tropical lakes”. Global Biogeochemical Cycles, v.23.

Medeiros, L.C., Mattos, A., Lurling, M. and Becker, V., 2015, “Is the future blue-green

or brown? The effects of extreme events on phytoplankton dynamics in a semiarid man-

made lake”. Aquatic Ecology. DOI: 10.1007/s10452-015-9524-5.

Mendonça, R., Kosten, S., Sobek, S., Cole, J.J., Bastos, A.C., Albuquerque, A.L.,

Cardoso, S. J. and Roland, F., 2014, “Carbon sequestration in a large hydroelectric

reservoir: an integrative seismic approach”. Ecosystems, v.17, pp. 430-441.

Menezes, R.F., Attayde, J.L. and Vasconcelos, F.R., 2010, “Effects of omnivorous

filter-feeding fish and nutrient enrichment on the plankton community and water

transparency of a tropical reservoir”, Freshwater Biology, v. 55, pp. 767 – 779.

Murphy, J. and Rilley, J.P., 1962, “A modified single – solution method for

determination of phosphate in natural waters”. Analyt. Chim. Acta, v.27, pp. 31 – 36.

O`Sullivan, P.E. and Reynolds, C.S., 2003, “The lakes handbook”. Blackwell

publishing, 709p.

42

Pacheco, F.S., Roland, F. and Downing, J.A., 2013, “Eutrophication reverses whole-

lake carbon budgets”. Inland Waters, v.4, pp.41 – 48.

Porter, K.G. and Feig, Y.F., 1980, “The use of DAPI for identifying and counting

aquatic microflora”. Limnology and Oceanography, v. 25, pp. 943-948.

Richey, J.E., Melack, J.M., Aufdenkampe, A.K., Ballester, V.M. and Hess, L.L., 2002,

“Outgassing from Amazonian rivers and wetlands as a large tropical source of

atmosphere CO2”. Nature, v. 416, pp. 617-620.

Scheffer, M., 2004, “Ecology of shallow lakes”. ed. Springer, B.V., p. 378.

Scheffer, M., Portielje, R. and Zambrano, L., 2003, “Fish facilitate wave resuspension

of sediment”. Limnology and Oceanography, v. 48, pp. 1920-1926.

Sobek, S., et al. (2005), “Temperature independence of carbon dioxide supersaturation

in global lakes”. Global Biogeochem. Cycles, v. 19, GB2003.

Sousa, W., Attayde, J.L., Rocha, E.S. and Eskinazi-Ant’Anna, 2008, “The response of

zooplankton assemblages to variations in the water quality of four man-made lakes in

semi-arid northeastern Brasil”. Journal of plankton research, v. 30, pp. 699-708.

Starling, F., Lazzaro, X., Cavalcanti, C. and Moreira, R., 2002, “Contribution of

omnivorous tilapia to eutrophication of a shallow tropical reservoir: evidence from a

fish kill”. Freshwater Biology, v.47, pp. 2443 – 2452.

Tranvik, L., Downing, J. A., Cotner, J. B., Loiselle, S. A., Striegl, R. G., Ballatore, T. J.

et al., 2009, “ Lakes and reservoirs as regulators of carboncyclingandclimate”.

Limnology and Oceanography, v. 54, pp. 2298–2314.

Thornton, J. A. and Rast, W., 1993, “A test of hypotheses relating to the comparative

limnology and assessment of eutrophication in semiarid man-made lakes”. In

Straskraba, Y., Tundisi, J. G. and Duncan, A., ed, Comparative Reservoir Limnology

and Water Quality Management. Kluwer Academic Publishers, London, pp. 1–24.

43

Wahl, D.H., Wolfe, M.D., Santucci Jr., V.J. and Freedman, J.A., 2011, “Invasive carp

and prey community composition disrupt trophic cascades in eutrophic ponds”.


Wetzel, R.G. and Likens, G.E., 2000, “Limnological Analyses”. 3rd ed., Springer.

Xu, Y., Cai, Q., Shao, M., Han, X., and Cao, M., 2009, “Seasonal dynamics of

suspended solids in a giant subtropical reservoir (China) in relation to internal processes

and hydrological features”. Quartenary International, v. 208, pp. 138-144.

Zambrano, L., Scheffer, M. and Ramos, M.M., 2001, “Catastrophic response of lakes to

benthivorous fish introduction”. Oikos, v.94, pp. 344-350.

Zuur, A.F., Ieno, E.N. and Elphick, C.S., 2010, “A protocol for data exploration to

avoid common statistical problems”. Methods in Ecology and Evolution, v.1, pp. 3-14.

44








maximum depth (zMAX), secchi depth (secchi), pH in ESEC reservoir.

MEAN

STANDARD

DEVIATION MINIMUM MEDIAN MAXIMUM

DOC (mg.L-1

) 22.26 5.37 9.71 22.36 36.84

POC (mg.L-1

) 4.07 1.75 1.74 3.99 8.20

DN (mg.L-1

) 1.78 0.61 0.92 1.57 3.23

TN (mg.L-1

) 3.54 1.49 1.75 3.36 5.93

TP (µg.L-1

) 107.22 70.51 8.20 111.43 269.00

SRP (µg.L-1

) 45.43 64.59 1.00 15.36 237.56

GPP (µmol.h-1

.d-1

) 134.97 60.58 13.99 141.18 232.56

NEP (µmol.h-1

.d-1

) 46.56 66.48 -107.32 62.32 127.67

R (µmol.h-1

.d-1

) 97.08 31.06 32.17 99.95 193.33

PCO2 (µATM) 1615.33 950.70 67.43 1497.37 3873.94

BR (µmol.L-1

.O2.h-1

) 1.60 1.48 0.22 1.26 5.88

PR (µmol.L-1

.O2.h-1

) 2.80 1.15 0.26 2.76 4.71

DO (mg.L-1

) 5.63 1.69 2.00 5.50 10.04

TEMP (°C) 28.20 1.77 25.80 27.80 32.50

CHLA (µg.L-1

) 31.73 15.85 6.24 29.61 69.54

SFS (mg.L-1

) 20.17 17.68 2.50 13.21 56.67

SVS (mg.L-1

) 12.27 8.76 3.67 9.44 35.00

zMAX (m-1

) 2.73 0.58 1.80 2.75 3.80

SECCHI (m-1

) 0.33 0.22 0.08 0.30 0.98

pH 7.42 0.24 6.77 7.43 7.84

45








maximum depth (zMAX), secchi depth (secchi), pH in FARM reservoir.

MEAN

STANDARD

DEVIATION MINIMUM MEDIAN MAXIMUM

DOC (mg.L-1

) 41.08 17.90 18.33 37.77 73.36

POC (mg.L-1

) 11.93 4.29 2.00 12.23 21.61

DN (mg.L-1

) 3.78 2.59 1.27 2.75 10.76

TN (mg.L-1

) 6.58 3.47 1.53 6.61 15.89

TP (µg.L-1

) 142.64 105.90 15.00 118.75 317.40

SRP (µg.L-1

) 9.76 8.06 2.13 8.67 39.50

GPP (µmol.h-1

.d-1

) 492.29 320.61 35.54 407.11 1090.32

NEP (µmol.h-1

.d-1

) 296.79 276.04 -172.42 249.44 799.84

R (µmol.h-1

.d-1

) 200.17 66.11 84.49 207.96 320.21

PCO2 (µATM) 309.59 400.34 12.32 139.36 1714.64

BR (µmol.L-1

.O2.h-1

) 0.96 0.49 0.17 0.87 1.77

PR (µmol.L-1

.O2.h-1

) 5.30 2.52 1.05 4.79 9.23

DO (mg.L-1

) 4.64 1.89 1.35 4.35 8.90

TEMP (°C) 27.53 1.18 25.70 27.50 30.10

CHLA (µg.L-1

) 280.04 263.69 40.77 134.29 897.00

SFS (mg.L-1

) 4.72 5.45 0.50 3.00 26.00

SVS (mg.L-1

) 38.87 21.68 15.33 32.75 84.00

zMAX (m-1

) 3.40 0.77 2.10 3.30 4.80

SECCHI (m-1

) 0.29 0.14 0.10 0.30 0.60

pH 7.87 0.30 7.36 7.81 8.50

46

Table 3: Correlation coefficients between environmental variables and the first

components PCA axes.

VARIABLES Axis 1 Axis 2

CO2

-0.643

0.534

GPP

0.935

0.031

R

0.868

0.134

PR

0.788

0.121

BR

-0.251

0.547

DO

-0.558

-0.384

SFS

-0.363

0.806

SVS

0.914

0.118

zDEPHT

-0.088

-0.753

SECCHI

-0.392

-0.571

POC

0.744

-0.151

DN

0.844

0.189

TN

0.833

0.052

TP

0.517

0.543

SRP

-0.263

0.688

PH

0.606

-0.471

TEMP -0.126 0.458

47

Table 4: Test t for independent samples (groups), to test the differences between

variables of ESEC reservoir and FARM reservoir during two years of monitoring. The

variables were: dissolved organic carbon (DOC), particulate organic carbon (POC),

dissolved nytrogen (DN), total nytrogen (TN), total phosphorus (TP), soluble reactive

phosphorus (SRP), gross primary production (GPP), net ecosystem production (NEP),

ecosystem respiration (R), partial pressure of carbon dioxide (pCO2), bacterial

respiration (BR), total plankton respiration (PR), dissolved oxygen (DO), temperature

(TEMP), Chlorophyll - a concentration (Chla), suspended fixed solids (SFS), suspended

volatile solids (SVS), maximum depth (zMAX), secchi depth (secchi), pH. Inside the

table are t-value, degrees of freedom (df) and p-value (p).

t-value df p

DOC (mg.L-1

) -4.83 44 p<0.01*

POC (mg.L-1

) -8.14 44 p<0.01*

DN (mg.L-1

) -3.58 44 p<0.01*

TN (mg.L-1

) -3.83 44 p<0.01*

TP (mg.L-1

) -1.19 44 0.24

SRP (µg.L-1

) 2.70 44 p<0.01*

GPP (µmol.h-1

.d-1

) -5.28 44 p<0.01*

NEP (µmol.h-1

.d-1

) -4.26 44 p<0.01*

R (µmol.h-1

.d-1

) -6.73 44 p<0.01*

PCO2 (µATM) 6.61 44 p<0.01*

BR (µmol.L-1

.O2.h-1

) 2.05 44 0.04*

PR (µmol.L-1

.O2.h-1

) -4.33 44 p<0.01*

DO (mg.L-1

) 1.61 44 0.11

TEMP (°C) 1.39 44 0.17

CHLA (µg.L-1

) -4.52 44 p<0.01*

SFS (mg.L-1

) 4.20 44 p<0.01*

SVS (mg.L-1

) -5.41 44 p<0.01*

zMAX (m-1

) -3.19 44 p<0.01*

SECCHI (m-1

) 0.94 44 0.35

pH -5.59 44 p<0.01*

* p-value equal or lower than 0.05.

48


( SFS, GPP, BR TP) separeted that showed significant effect (p<0.1) to explained the

variance of CO2 at ESEC reservoir.

SIMPLE REGRESSION - ESEC RESERVOIR - DEPENDENT VARIABLE CO2

Variables R R2 Adjusted R

2 Parameter F P

SFS 0.681 0.465 0.441 0.012 19.128 <0.01

GPP 0.474 0.225 0.191 -0.031 7.308 0.01

BR 0.368 0.135 0.096 0.028 3.451 0.07

TP 0.381 0.145 0.107 0.000 3.756 0.06

49


(SVS, R, PR, SRP, zMAX) separeted that showed significant effect (p<0.1) to explained

the variance of CO2 at FARM reservoir.

SIMPLE REGRESSION - FARM RESERVOIR - DEPENDENT VARIABLE CO2

Variables R R2 Adjusted R

2 Parameter F P

SVS 0.445 0.198 0.160 -0.021 5.201 0.03

R 0.491 0.241 0.204 -0.081 6.654 0.02

PR 0.449 0.202 0.164 -0.002 5.322 0.03

SRP 0.392 0.154 0.114 -0.007 3.830 0.06

zMAX 0.368 0.136 0.095 0.000 3.307 0.08

50

Figure legends:

Figure 1: CO2 concentration during two years of monitoring in the ESEC reservoir and

FARM reservoir (September – 2012 to September – 2014). The black bars correspond

to the CO2 concentration in the ESEC reservoir. The white bars correspond to the CO2

concentration in the FARM reservoir. The red line correspond to the equilibrium

atmosphere concentration (CO2 = 390 µATM ). When the CO2 concentration > 390

µATM, the environment is emmiting CO2 to the atmosphere. When the CO2 < 390

µATM, the environment is synk by atmosphere.

Figure 2: Principal Component Analyse (PCA) scores of 18 environmental variables in

the ESEC reservoir (E – dark circles) and FARM reservoir (F – open circles).

Environmental variables monitored during two years (from September - 2012 to

September – 2014), particulate organic carbon (POC), dissolved nitrogen total (DN),

total nitrogen (TN), total phosphorus (TP), soluble reactive phosphorus (SRP), gross

primary production (GPP), ecosystems respiration ( R ), carbon dioxide concentration

(CO2), bacterial respiration (BR), planktonic respiration (PR), dissolved oxygen (DO),

temperature (TEMP), suspended fixed solids - inorganic (SFS), suspended volatile

solids - organic (SVS), maximum depth (zMAX), transparence of Secchi (secchi), PH.

Figure 3: Monitoring of limnologicalvariables during two years (from September of

2012 to September of 2014) of two reservoirs inserted in the semiarid region of

northeastern Brazil. The dark bar correspond the values of ESEC reservoir. The white

bar correspond the values of FARM reservoir. Insert in this figure are: (a) gross primary

production – GPP (µmol.h-1

.d-1

), (b) net ecosystem production – NEP (µmol.h-1

.d-1

), (c)

ecosystem respiration – R (µmol.h-1

.d-1

) and (d) Chlorophyll – a – Chla (µg.L-1

).

Figure 4: Suspended solids (mg.L-1

) concentration during two years of monitoring in the

ESEC reservoir and FARM reservoir (September – 2012 to September – 2014). The

black bars correspond to the SFS (suspended inorganic solids) and white bars

correspond to the SVS (suspended organic solids). (a) SFS and SVS concentration in

the ESEC reservoir. (b) SFS and SVS concentration in the FARM reservoir.

Figure 5: Monitoring of variables during two years (from September of 2012 to

September of 2014) of two reservoirs inserted in the semiarid region of northeastern

Brazil. The dark line correspond the values of ESEC reservoir. The gray line correspond

51

the values of FARM reservoir. The variables insert in this figure are: (a) bacterial

respiration – BR (µmol.O2.h-1

), (b) planktonic respiration – PR (µmol.O2.h-1

).

Figure 6: BR:PR ratio in the ESEC reservoir and FARM reservoir during two years of

monitoring (September – 2012 to September – 2014). The mid black of the bars

correspond to the ESEC values. The mid white of the bars correspond to the FARM

values.

Figure 7: (a) zMax (m-1

) and (b) transparency of secchi – secchi (m-1

), during two years

of monitoring in the ESEC reservoir and FARM reservoir (September – 2012 to

September – 2014). The dark line correspond the values of ESEC reservoir. The gray

line correspond the values of FARM reservoir.

52

Figure 1. Moura et al.

53


- ESEC RESERVOIR

- FARM RESERVOIR

54

Figure 3 Moura et al.

a b

c d

55


a

b

56


a

b

57


58


a

b

59

CAPÍTULO II

60

Benthivorous fish increase CO2 emission in a shallow semiarid

eutrophic reservoir.

Caroline Gabriela Bezerra de Moura1, Danhyhelton Douglas

2, Maria Marcolina

Cardoso2, Mariana Amaral da Costa

3, Fabiana Araújo

2, Pablo Rubim

2, Leonardo

Teixeira2, Jurandir Rodrigues

3, Marcos Paulo Figueiredo-Barros

4, José Luiz de Attayde

2

and André Megali Amado1.







3 – Departamento de Engenharia Civil



4 – Departamento de Ecologia

Núcleo de Pesquisas em Ecologia e Desenvolvimento Sócio Ambiental de Macaé

Universidade Federal do Rio de Janeiro – UFRJ – Brasil


Key – words: carbon flux, aquatic metabolism, nutrient translocation.

61

Abstract

Sediment resuspension by benthivorous fish (bioturbation) in shallow lakes can

stimulate phytoplankton biomass through nutrient translocation or inhibit it due to light

availability decrease. Thus, the bioturbation might affect planktonic metabolism in

antagonist ways and it is not yet clear how the bioturbation process affect the carbon

cycle in shallow lakes. Most inland waters (lakes, rivers, reservoirs) in the world are

estimated to function as heterotrophic ecosystems, especially in the tropical zone. In the

Brazilian semiarid region it was recently shown that 91% of the lakes are net

heterotrophic, even though most of them are eutrophic. Besides, a benthivorous fish

Prochildus brevis is a dominant fish of Brazilian semiarid region and might be a key

component of the carbon cycle in these ecosystems. The aim of this study was to

evaluate the effect of sediment resuspension by a benthivorous fish on the carbon

dioxide concentration (CO2) at a shallow tropical semiarid lake. We hypothesized that

the sediment resuspension induced by a benthivorous fish enhance the CO2 and the CO2

flux to the atmosphere. To test this hypothesis we performed a 2 x 2 factorial mesocosm

experiment designed in four treatments with and without fish and with fish having

access or not to the sediment. We found that the bioturbation enhanced phosphorous

availability and Chorophyll a (Chla) concentration (e.g. primary production), but also

enhanced the bacterial respiration rates and CO2 release from the sediment. The overall

effect resulted in higher CO2 emission to the atmosphere due to the fish bioturbation.

We concluded that the benthivorous fish Prochildus brevis bioturbation might increase

CO2 production in tropical shallow reservoirs, which can have important implications to

the carbon balance.

62

Introduction

The inland aquatic ecosystem plays important role to global carbon cycling since

they can fix, process and transport great amount of carbon to the ocean and to the

atmosphere (Cole et al., 2007; Tranvik et al., 2009). An aquatic ecosystem behaves

either as sink or source of carbon dioxide (CO2) to the atmosphere depending on its

dominant metabolic processes. When gross primary production (GPP) exceeds the net

ecosystem respiration (NER), the system turns to autotrophic state and it acts as a sink

of CO2. However, when the NER exceeds GPP the system moves to a heterotrophic

state acting as a source of CO2 to the atmosphere (Cole et al., 2000).

The organic matter origin (allochthonous or autochthonous organic matter), the

input of nutrients to the system and the food web structure are factors that may

influence how the aquatic system acts in carbon cycling (Schindler et al., 1997; Tranvik

et al., 2009). For instance, the presence of benthivorous fishes may represent a link

between benthonic and pelagic habitats with great effects in shallow lakes (Vanni et al.,

1997; Vanni, 2002). Thus, benthivorous fish may translocate nutrients from the

sediment to the water column, process known as bioturbation, which may stimulate

phytoplankton biomass (Vanni, 2002; Roozen et al., 2007). As a consequence, the high

phytoplankton biomass could exceed the respiration shifting the system to an

autotrophic state; thus acting as a sink of CO2. Indeed, Gu et al. (2011) observed that a

shallow subtropical eutrophic lake with a high input of nutrients and predominance of

planktivorous-benthivorous fish remained autotrophic for a long period, while

functioning as a sink of CO2 from the atmosphere.

On the other hand, the benthivorous fish bioturbation can also increase the

turbidity (e.g. resuspension of particles from the sediment) of ecosystems leading to a

decreased primary production (Wahl et al., 2011). Together with the nutrients release,

the fish action may also release great amounts of organic matter fuelling the

heterotrophic activity and the microbial food web (Cotner et al., 2000). Besides,

bioturbation may also directly release gases from the sediment, such as CO2 and

methane (CH4) (Figueireido-Barros et al., 2009). As a consequence, the system might

shift to a heterotrophic state, functioning as a source of CO2 to the atmosphere.

However, it is not yet clear in which environmental conditions the benthivorous fish

action contribute to the autotrophic or to the heterotrophic state.

63

The bacterioplankton is an important mediator in the processing of organic

matter in aquatic ecosystems. They can act as decomposers of dissolved organic matter

contributing to the CO2 production (Bergreen et al., 2011). Moreover, bacterioplankton

can act as a carbon source for higher trophic levels through the microbial food web

(O`Sullivan and Reynolds, 2004). Nutrients (such as nitrogen and phosphorous) and

dissolved organic carbon (DOC) availability increase usually stimulate both bacterial

mineralization (e.g. bacterial respiration; BR) and bacterial biomass production (BP)

(e.g. Farjalla et al., 2002). Furthermore, increased autochthonous DOC usually increases

BP, rather than BR (Farjalla et al. 2006). However, in tropical environments rates by

unit of biomass production are two times higher than in temperate regions, suggesting a

greater nutrient and organic matter turnover and higher CO2 production (Amado et al.,

2013), which might favor carbon mineralization than flow through the trophic chain.

Thus, the fish bioturbation might affect microbial metabolism in antagonistic ways, and

it is not straightforward to predict whether it can increase carbon mineralization

(releasing CO2) or carbon uptake in the microbial food web.

Eutrophication of lakes is currently a global issue (Smith and Schindler, 2009)

and is being accelerated by anthropogenic activities (Carpenter et al., 1998). The

predominance of benthivorous fish, especially in shallow lakes, can contribute to

eutrophication process through internal reloading of nutrients from the sediments

(Jeppesen et al., 2000; Lazzaro et al., 2003). In Brazil, most of the reservoirs inserted in

the semiarid region (northeast) are eutrophic (Bouvy et al., 2000; Souza et al., 2008)

and are usually dominated by benthivorous fishes, such as Prochilodus brevis (Gurgel

and Fernando, 1994; Nascimento et al., 2014), that might contribute to the trophic status

of these ecosystems. Thus, a question that rises is whether or not the presence of the

bethivorous fishes affects the carbon cycling by remobilizing nutrients and organic

matter from the sediments. We hypothesized that the bioturbation by benthivorous fish

Prochilodus brevis, enhance the CO2 emission in a tropical semiarid shallow reservoir.

To test this hypothesis we performed a 2x2 full factorial mesocosm experiment in a

shallow semiarid reservoir in Brazil manipulating the presence and absence of the

benthivorous fish and with or without access to the sediment.


Study area and experimental design

64

The current study was performed from 25 June to 25 July of 2013 in a small

(11ha) and shallow (max. depth 4m) reservoir situated at Seridó Ecological Station

(ESEC) in Serra Negra do Norte, Rio Grande do Norte, Brazil (06°34'852″N,

37°15'519″W). The reservoir is considered eutrophic with the chlorophyll-α (Chlɑ),

total phosphorus (TP) and total nitrogen (TN) respectively: 25 (µg.L-1

), 114 (µg.L-1

) and

3981(µg.L-1

) and concentration of carbon dioxide CO2; 2876 µATM.

To investigate the effects of the bioturbation of the bentivorous fish Prochilodus

brevis we manipulate its presence and absence, as well as the acess and no acess to

sediment twenty mesocosms (depth: 2m, area: 4m2, volume: 8m

3), made of an

aluminum frame surrounded by transparent plastic (0.45 mm of thickness) were used to

isolate the inside water from the reservoir. All mesocosms were open to atmosphere and

ten of the mesocosms had a galvanized wire mesh (pvc coated) attached close to the

bottom, blocking the fish access to the sediment. The side iron bars were buried in the

sediment around 20 cm to ensure a seal along the sediment. The experimental design

consisted of four treatments: sediment access without fish (A-F), without sediment

access and without fish (N-F), sediment access plus fish (A+F), without sediment access

with fish (N+F). The treatments were replicated five times and randomly assigned. The

treatments with fish were set by adding 4 individuals of Curimatã (Prochilodus brevis;

final density of 0.5 individual per cubic meter). The chosen density was similar to the

density in the reservoir (data not shown – removal fishes).

The experiments began immediately after the placement of mesocosms

structures. The experiment lasted for 30 days and the samplings during the experiment

were performed in the beginning, fifteen days and at the end of the experiment: day 1

(right before the fish addition), day 15 and day 30 (end of experiment). The variables

monitored during the experiment in the mesocosms were: partial pressure of carbon

dioxide (CO2), total plankton respiration (PR), bacterial respiration (BR), Chlorophyll -

a concentration (Chla), total organic carbon (TOC), total phosphorus (TP), total

nitrogen (TN), dissolved oxygen (DO), water transparence - Secchi depth (WT),

bacterial abundance (BA), heterotrophic flagellates abundance (HNF), total zooplankton

abundance (Total Zoo) and methane concentration (CH4) in the water.

Sampling

65

At first, in each mesocosm the water temperature and dissolved oxygen profile were

measured at water column using a portable oxygen meter (Instrutherm MO-900). We

also took water samples for CO2 and CH4 concentration determination. For CO2, water

samples were taken with polycarbonate bottle (100mL) and complete removal of

internal atmosphere or bubbles. Samples were taken to the field laboratory for

measurements (c.a. 30 min). Water samples for Ch4 determination were collected (20

mL) in the bottom of the water column (right above the sediment; depth=1.5m) of each

mesocosm and the lake (control) with Van D’orn bottle and were immediately injected

into 50mL glass capped vials at negative pressure and preserved with NaCl (20% of

volume as the final concentration). The secchi disk depth was measured as light

attenuation in each mesocosm. Finally, for each mesocosm we collected water samples

with a 1.5m long tube (through water column) at 5 random points into each mesocosm

and integrated in a plastic bucket (20L) to subsample for: PR, BR, Chla, TOC, TP, TN,

DO, BA, HNFA. The samples to zooplankton abundance was filtered through a

plankton net (64µm) of mesh size – 20L.

Analyses

The CO2 was estimated from the pH and alkalinity performed titration using

0.02N H2SO4, adjusting for temperature, ionic strength and air pressure (Cole et al.,

1994). Subsequently, the results were expressed as undersaturated or supersaturated

with CO2 relative to the atmosphere (considered here as being 390 uATM).


while BR rates were estimated as oxygen consumption in filtered through glass fiber

filtered (1.2µm average pore size; VWR INTERNATIONAL) water using a golden tip

oxygen microsensor connected to a picoammeter controlled by the MicOx software

(Unisence ©; Briand et al., 2004). The samples for PR and BR measurements were

incubated in exetainers (5.9mL; Labco®) with no internal atmosphere in 5 replicates for

each mesocosm in the dark and room temperature (25⁰C ± 1) for 24 hours.

TOC and TN concentration was measured by catalytic combustion in a Total

Organic Analyzer (TOC – V, Shimadzu – 2.0) with a TN analyzer attached (VNP

module). TOC was calculated from the sum of the dissolved organic carbon (DOC) and

66

particulated organic carbon (POC) (Wetzel and Likens, 2000). TP concentration was

measured with a spectrophotometer by the acid ascorbic method after persulphate

digestion (Murphy and Riley, 1962). Water samples were filtered through glass fiber

filter (VWR INTERNATIONAL – 1.2µm) for chlorophyll-α concentration, which was

extracted with ethanol 95% and measured by spectrophotometry (Jespersen and

Cristoffersen, 1987).



determined after nucleic-acid staining with Syto13 (Molecular Probe) at 2.5 µM (Del

Giorgio et al., 1996). Fluorescent latex beads (Polysciences, 1.5µm diameter) were

added to each sample for calibration of side scatter and green fluorescence signals, and

as an internal standard for the cytograms.




epifluorescence microscope (Porter and Feig., 1980). On average 400 individuals were


The zooplankton organisms were counted under a microscope in a 1 mL Sedwick-

Rafter chamber. Between three and five subsamples were counted for each sample

collected in the field until a minimum of 100 individuals of each taxonomic group had

been counted. Subsequently, the average of the subsamples was taken for each group of

organisms counted, this being multiplied by the sample volume (mL) and divided by the

subsample volume (1 mL) to estimate the total number of individuals in the sample.

Afterwards, the number of individuals in the sample was divided by the water volume

(L) sampled in the Field to calculate the original density (Ind. L-1

) of organisms in the

sample.

The samples for CH4 concentration were analyzed through gas chromatrography

using a Varian Star 3400 chromatrograph equipped with a POROPAK-Q column (as

described in Figueireido – Barros et al., 2009).

Statistical analysis

67

To analyze the effect of fish and its access to sediment over dependent variables

we used a two-way Anova. The values have been averaged (between day 15 and day 30)

following the proposed by Shaus and Vanni (2000). Prior to analyses, data were log

transformed to stabilize variances among treatments (homogeneity). Homogeneity of

variance was tested by Levene’s Test, and a significant level of α = 0.05 was assumed.

To understand the differential fish effect with and without sediment access, we estimate

the effect size for the fish accessing or not the sediment of each limnological variable

using the log ratio [ln(experiment/control)]. Two controls were considered: A-F as the

control of A+F, and N-F as the control of N+F. The effect size was evaluated crossing

all replicates of the experiment with replicates of the respective control creating 25

effect sizes for each dependent variable building up a histogram of distribution. The

effect size response is considered significant when the confidence interval did not cross

the zero. Positive results are shown when the differences between treatments were

higher than 0. Negative results are showed when the differences between treatments

were lower than 0. Then, the identified distribution was used in a bootstrap analysis

generating the confidence interval. The distribution was accessed by the “fitdistrplus”

package and the confidence interval by the boot package. We used the R software (R

Development Core Team, 2011) for all analysis.

Results

No pre-treatment variations were found among the mesocosms and they were all

considered as heterotrophic (CO2 sources to the atmosphere; CO2 = 2876±239 µATM)

and eutrophic (Chla = 46±9 µg.L-1

; TP = 54±39 µg.L-1

) prior to treatments addition.

The experimental results showed that the F+A treatment had higher (p < 0.05) CO2,

Chla, BR, PR, TP, and lower (p < 0.05) Secchi, C:P, N:P, D.O, CH4, than the other

treatments following the same pattern of the reservoir for most variables (Figure 2;

Figure 3).

The ANOVA shows a significant interaction between access (A) and fish

presence (F) for the following variables: Chla, BR, TOC, TN, TP, Secchi, D.O, C:P

ratio, N:P ratio and CH4 (Table 1). The significant individual effect of fish affected the

CO2,, TP, D.O, C:P, N:P ratios and CH4. The significant individual effect of access

affected the CO2, Chla, TP, Secchi, D.O, C:P ratio, N:P ratio (Table 1). BA and HNF

abundance and C:N ratio did not show any effect from treatments (Table 1).

68

The effect size results reveal a differential effect of the benthivorous fish when it

has access to the sediment. The fish with access have a positive effect on CO2, chl-a,

TP, TN BR, PR, TOC and a negative effect on water transparency, C:N, C:P, N:P, D.O,

and CH4 (Figure 2). However, when the access to the sediment was removed the fish did

not show any effect for the most of variables, except the negative effect over TOC and

TN. Both fish treatments did not affected neither the bacterial abundance nor HNF

abundance (Figure 2).

Discussion

Our results demonstrated that the occurrence of the benthivorous fish Prochilodus

brevis has major implications to the shallow semiarid reservoir carbon cycling. Under

our experimental conditions, the bioturbation of the benthivorous fish resulted in the

super-saturation in CO2 in mesocosms confirming our working hypothesis. Even though

the benthivorous fish bioturbation released phosphorous from the sediment and slightly

increased Chla concentration its strongest response resulted in predominance of CO2

super-saturation in the water column. These findings have important implications to the

ecosystem functioning, once the native fish enhance the coupling between sediment and

pelagic compartments (for instance promoting nutrients release from the sediment to the

water column) and intensify the interaction and carbon exchange between lake and

atmosphere. It is clear that the fish bioturbation enhanced both the carbon sink and

release from the reservoir, with the prevalence of the latter. There are two possible

mechanisms to explain the predominance of CO2 emission due to the bioturbation: (1)

turbidity increase and releasing of nutrients and organic matter from the sediment,

which may stimulate the heterotrophic microbial food web (Leal et al., 2003) and

organic matter mineralization instead of primary production and, (2) direct CO2 and

other gases, such as methane (CH4) release from the sediment to the water column and

also CH4 oxidation in the sediment enhacing even more the CO2 release.

Regarding the first mechanism mentioned above, it is clear that the bioturbation

increased the carbon bioavailability to bacteria, since Chla increased and C:P, C:N and

N:P ratios decreased (Figure 2), which would increase bacterial metabolism. In fact,

algal biomass is well known as a labile organic matter source to bacterial metabolism

(Stets and Cotner, 2008). Furthermore, N and P availability increase resulted in higher

bacterial metabolic rates in tropical ecosystems, with emphasis to biomass production

69

(Farjalla et al., 2002). Accordingly, P release from sediment resuspension in the deep

Lake Michigan resulted in increased heterotrophic bacterioplankton biomass

productivity (Cotner, 2000). In the current study we did not observe any increase in

bacterial biomass, but there were increased bacterial respiration rates in the treatment

that fish bioturbation took place (A+F; Table 1; Figure 2). As HNF biomass was also

not affected as a consequence of fish bioturbation (Table 1; Figure 2), we could suggest

that there was no predation control on bacterial biomass, but probably a metabolic

constraint triggering the respiration increase. Under high temperature, it is common to

observe bacterial respiration increase rather than biomass production due to nutrients

concentration increase (Berggren et al. 2010; Hall and Cotner, 2007).

In one hand, the P availability increase was not recognized as a factor to increase

HNF biomass in light and nutrients manipulation experiments, i.e. no bottom-up

regulation (Elser et al., 2003), in agreement to our results. On the other hand, it is

possible that the zooplankton biomass increase due to the fish bioturbation could have

top-down regulated the HNF biomass. Thus, the BR increase due to bioturbation was

probably a result of metabolic alterations in bacterial metabolism, while the PR increase

trend could have resulted from increased zooplankton biomass (Dantas et al., 2015).

Both processes could explain, at least in part, the CO2 increase due to bioturbation. The

high average temperature in the studied semiarid ecosystem (27ºC ± 2ºC) could be

limiting to bacterial biomass production (Amado et al., 2013), independent of P

availability, which could explain the pattern of increased BR recorded here.

It is worthy noticing that the BR rates were estimated here in pre-filtered water

samples, to exclude bacterivores and lager organisms and avoid super-estimation (e.g.

Biddanda et al., 2001). However, the filtration process may exclude bacteria that are

adhered in particles and thus, underestimate BR rates. Considering that the treatment

where fish have access to the sediment the bioturbation process re-suspended sediments

and bacteria to the water column and in the other treatments it did not, it is reasonable to

argue that BR rates in the bioturbation (A+F) treatment should be even higher then the

other treatments and be even more relevant to the net heterotrophy under these

conditions (Table 1).

70

Regarding the second mechanism mentioned above, the CO2 super-saturation due

to bioturbation could be related to direct or indirect CO2 efflux from the sediment. The

bioturbation process may directly release to the water column CO2 that was previously

trapped in the sediment (Kristensen et al., 2001). Furthermore, the bioturbation

oxygenates anaerobic layers of sediments shifting anaerobic to aerobic metabolism

enhancing the organic matter mineralization rates and, consequently increasing CO2

concentration (Banta et al., 1999; Kristensen et al., 2001). For instance, CH4 oxidation

is usually stimulated due to bioturbation through oxygenation in the sediment

(Figueiredo-Barros et al., 2009). In agreement, in our study, the treatments where fish

had access to the sediment, the CH4 and dissolved oxygen concentration in the water

were reduced, probably due to higher CH4 oxidation rates (Table 1; Figure 2). Thus, it is

reasonable to infer that the bioturbation promoted the CH4 oxidation corroborating the

hyphotesis of enhanced CO2 formation due to mineralization rates increase. These

mechanisms could explain part of the CO2 increase in treatments with fish accessing the

sediments in the current experiment (Bastviken et al., 2003). One could argue that the

presence of the fishes itself would result in increased CO2 concentration in the water

due to fish respiration. However, it could be considered negligible since changes in the

effect size for CO2 concentration was due to access to the sediment instead of presence

of fish (Figure 2).

In the treatments where the bioturbation did not take place (N-F, N+F, A-F) there

were recorded the lowest concentrations of CO2 (380 μATM ± 10) after 30 days of

experiment (Figure 3) and treatments were equilibrated with the atmosphere for CO2

concentration. Furthermore, in the same treatments, it was evident the reduction of

parameters that indicate eutrophication, such as Chla, water transparency and total

phosphorous (Table 1; Figure 3). It has been previously demonstrated that fish

manipulation, such as trophic cascade manipulation, in lakes could drive eutrophication

restoration and also carbon flux reversion (Schindler et al., 1997). Our results

demonstrate that the benthivorous habit of the dominant fish species might be one

important mechanism driving semiarid reservoirs to net heterotrophy (Junger et al.,

2015; Dantas et al., 2015). On the other hand, our results also suggest that the controlled

removal of the benthivorous fish may work as tool to minimize carbon efflux to

atmosphere in the semi-arid reservoirs, but also to revert eutrophication process, which

71

was addressed in details in other works (Dantas et al., 2015; Araújo et al., 2016 – In

press).

We concluded here that the benthivorous fish is an important component on the

carbon cycle in the tropical semi-arid reservoirs, once it stimulates the net heterotrophy

through sediment bioturbation. We proposed that this bioturbation fuels the bacterial

respiration in the water column with organic matter from the sediment and

phytoplankton exudates. Moreover, the direct CO2 release from the sediment, as well as

the aerobic oxidation of organic compounds in the sediment, such as the methane

oxidation enhancement due to bioturbation, should have contributed to the

predominance of CO2 emission. The results from our work is extremely relevant to the

carbon budget in the tropical semi-arid regions and to global carbon budgets because it

demonstrates the importance of the sediment and fish community composition to CO2

release. Considering that the benthivorous fish Prochilotus brevis is a dominant species

in several semi-arid aquatic ecosystems in Brazil and its bioturbation ability, this fish

might play a relevant role to the prevalence of CO2 super-saturation in most eutrophic

reservoirs in this region (Junger et al., 2015; Dantas et al, 2015). Moreover, the lower

nutrients concentration found in the treatments without fish, suggests that the

benthivorous fish biomanipulation (e.g. massive removing) might be a useful tool for

improving the water quality of reservoirs located in semi-arid regions, besides reducing

CO2 emissions to the atmosphere.

72

Reference



bacterial growth efficiency than temperate ones”. Frontiers in Microbiology, v.4, pp.1-

8.

Araújo et al., Becker, V. and Attayde, J.L., 2016, “Shallow lake restoration by the

combined effects of polyaluminium chloride addition and benthivorous fish removal: a

field mesocosm experiment”. Hydrobiologia. In press.

Banta, G.T., Holmer, M., Jensen, M.H. and Kristensen, E., 1999, “Effects of two

polychaete worms, on aerobic and anaerobic decomposition in a sandy marine

sediment”. v.19, pp.189-204.

Bastviken, D., Ejlertsson, J., Sundh, I. and Tranvik, L., 2003, “Methane as a source of

carbon and energy for lake pelagic food webs”.Ecology, v. 84, pp. 969-981.

Bergreen, M., Lapierre, J.F. and Del Giorgio, P., 2011, “Magnitude and regulation of

bacterioplankton respiratory quotient across freshwater environmental gradients”. The

ISME Journal, pp. 1-10.

Bergreen, M., Laudon, H., H, Jonsson, A. and Jansson, M., 2010, “Nutrient constraints

on metabolism affect the temperature regulation of aquatic bacterial growth efficiency”.

Microbial Ecology, v. 60, pp. 894-902.







Carpenter, S., Chair, Caraco, N.F., Correl, D.L., Howarth, R.W., Sharpley, A.N. and

Smith, V.H., 1998, “Nonpoint pollution of surface waters with phosphorus and

nitrogen”. Issues in Ecology, v.3, pp.1-12.

73











Dantas, D.D.F. (2015) Causas e consequências da onivoria de peixes em ecossistemas

aquáticos (In Portuguese). PhD Thesis. Federal University of Rio Grande do Norte, Rio

Grande do Norte, Brazil.

Dantas, F.C.C. (2014) Saturação em CO2 e regulação metabólica do bacterioplâncton

em ecossistemas aquáticos de baixa latitude (in Portuguese). Master Thesis. Federal

University of Rio Grande do Norte, Rio Grande do Norte, Brazil.

Del Giorgio, P.A., Bird, D.F., Prairie, Y.T. and Planas, D., 1996, “Flow cytometric



Del Giorgio, P.A., Cole, J.J. and Cimbleris, A., 1997, “Respiration rates in bacteria


148-151.

Elser, J.J., Kyle, M., Makino, W., Yoshida, T. and Urabe, J., 2003, “Ecological

stoichiometry in the microbial food web: a test of the light:nutrient hypothesis”. Aquatic

Microbial Ecology, v. 31, pp. 49-65.

Farjalla, V.F., Esteves, F.A., Bozelli, R.L. and Roland, F., 2002, “ Nutrient limitation of

bacterial production in clear water Amazonian ecosystems”, Hydrobiologia, v. 489, pp.

197-205.

74

Farjalla, V.F., Azevedo, D.A., Esteves, F.A., Bozelli, R.L., Roland, F. and Prast, A.,

2006, “Influence of hydrological pulse on bacterial growth and DOC uptake in a clear-

water Amazonian lake”. Microbial Ecology. v. 52, pp. 334-344.

Figueiredo – Barros, M.P., Caliman, A., Leal, J.J.F., Bozelli, R.L., Farjalla, V.F. and

Esteves, F.A., 2009, “Benthic bioturbator enhances CH4 fluxes among aquatic

compartments and atmosphere in experimental microcosms”. Can. J. Fish. Aquat. Sci.,

v. 66, pp. 1649-1657.



Gurgel, J.J.S. and Fernando, C.H., 1994, “Fisheries in Semi-Arid Northeast Brazil with

special reference to the role of Tilapias”. Int.Revue ges. Hydrobiol, v.79, pp.77-94.

Hall, E.K. and Cotner, J.B., 2007, “Interactive effect of temperature and resources on

carbon cycling by freshwater bacterioplankton communities”. Aquatic Microbial

Eclogy, v. 49, pp. 35-45.


phytoplankton using ethanol as extraction solvent”, Archiv fuer Hydrobiologie

AHYBA4, v.109, pp. 445-454.

Jeppesen, E., Jensen, J.P., Sondergaard, M., Lauridsen, T. and Landkildehus, F., 2000,

“Trophic structure, species richness and biodiversity in Danish lakes: changes along a

phosphorus gradient”. Frershwater Biology, v. 45, pp. 201-218.

Kristensen, E., 2001, “Impact of polychaetes (Nereis spp. and Arenicola marina) on

carbon biogeochemistry in coastal marine sediments”. Geochemical Transactions, v. 12.




Leal, J.J.F., Esteves, F.A., Farjalla, V.F. and Enrich-Prast, A., 2003, “Effect of

Campsurus notatus on NH4+, DOC flux, O2 uptake and bacterioplankton production in

experimental microcosm with sediment-water interface of an Amazonian lake impacted

by bauxite tailings”. Internat. Rev. Hydrobiol., v.88, pp.167-178.

75



Nascimento, W.S., Barros, N.H.C., Araújo, A.S., Gurgel, L.L., Canan, B., Molina, W.F.

and Chellapa, S., 2014, “Composição da ictiofauna das bacias hidrográficas do Rio

Grande do Norte, Brasil”. Biota Amazônica. v.4, pp. 126-131.



Rooze, F.C.J.M., Lurling, M., Vlek, Hanneke, Van Der Pouw Kraan, Ibelings, B.W. and

Scheffer, M., 2007, “Resuspension of algal cells by benthivorous fish boosts

phytoplankton biomass and alters community structure in shallow lakes”. Freshwater

Biology, v. 52, pp. 977-987.

Schindler, D.E., Carpenter S.R., Cole, J.J., Kitchell, J.F. and Pace, M.L., 1997,

“Influence of food web structure on carbon exchange between lakes and the

atmosphere”. Science, v. 277, pp. 248-251.

Schaus, M.H. and Vanni, M.J., 2000, “Effects of gizzard shad on phytoplankton and

nutrient dynamics: role of sediment feeding and fish size”. Ecology, v. 81, pp. 1701-

1019.

Smith, V.H. and Schindler, D.W., 2009, “Eutrophication science: where do we go from

here?”. Trends in Ecology and Evolution, v.24, pp. 201-207.




Stets, E.G. and Cotner, J.B., 2008, “The influence of dissolved organic carbon on

bacterial phosphorus uptake and bacteria-phytoplankton dynamics in two Minessota

lakes”. Limnology and Oceanography, v. 53, pp. 137-147.


ET AL., 2009, “ Lakes and reservoirs as regulators of carboncyclingandclimate”.


76

Vanni, M.J., 2002, “Nutrient Cycling by animals in freshwater ecosystems”. Annu. Rev.

Ecol. Syst., v. 33, pp. 341-370.

Vanni, M.J., Layne, C.D. and Arnott, S.E., 1997, “Top- down trophic interactions in

lakes: effects of fish on nutrient dynamics”. Ecology, v. 78, pp. 1-20.




Wetzel, R.G. and Likens, G.E., 2000. “ Limnological Analyses ”. 3rd

Ed. Springer-

Verlag. New York. 430p.

77



Variables A F A X F

F - ratio P F – ratio P F - ratio P

CO2 (µATM) 4.64 0.04** 7.19 0.01** 2.44 0.13

Chla (µg.L-1

) 10.08 <0.01** 2.24 0.15 11.25 <0.01**

BR (µmol.L-1

.h-1

) 0.43 0.51 2.96 0.1 7.78 0.01**

PR (µmol.L-1

.h-1

) 2.66 0.12 2.72 0.11 2.55 0.12

TOC (mg.L-1

) 0.37 0.54 0.24 0.62 3.61 0.07*

TN (µg.L-1

) 1.2 0.28 2.1 0.16 5.2 0.03**

TP (µg.L-1

) 37.05 <0.01** 70.76 <0.01** 26.37 <0.01**

Secchi depht (m) 47.59 <0.01** 1.83 0.19 54.28 <0.01**

DO (mg.L-1

) † 4.77 0.04** 4.93 0.04** 3.49 0.08*

C:N Ratio 0.14 0.71 0.24 0.6 0.009 0.92

C:P Ratio 16.18 <0.01** 25.08 <0.01** 19.75 <0.01**

N:P Ratio 14.81 <0.01** 19.12 <0.01** 10.29 <0.01**

BA (Cell.mL-1

) 0.006 0.93 0.57 0.45 0.21 0.64

HNF (Cell.mL-1

) † 1.17 0.29 0.95 0.34 1.04 0.32

CH4 (ppm) † 0.82 0.37 3.53 0.07* 3.33 0.08*

Total Zoo (Ind.L-1

) 0.778 0.394 13.639 0.003* 1.434 0.254

* p-value between 0.1 and 0.05.

** p-value equal or lower than 0.05.

†Only on the last sample date.

78

Figure legends:


in this study.

Figure 2: Effect size (mean and confidence interval) of benthivorous fish with and

without access to the sediment over response variables. A) CO2; B) Chlorophyll-a; C)

Bacterial respiration; D) Planktonic respiration; E) Total organ carbon; F) Total

nitrogen; G) Total phosphorous; H) C:N ratio ; I) C:P ratio; J) N:P ratio; K) Water

transparency; L) Methane (CH4); M) Dissolved oxygen; N) Bacterial abundance; O);

Flagellate abundance.


reservoir during the experiment. A) CO2; B) Chlorophyll-a; C) Bacterial respiration; D)

Planktonic respiration; E) Total organ carbon; F) Total nitrogen; G) Total phosphorous;

H) C:N ratio ; I) C:P ratio; J) N:P ratio; K) Secchi; L) Methane (CH4) – (Day 30); M)

Dissolved oxygen (Day 0; Day 30); N) Bacterial abundance; O); Flagellate abundance

(Day 0; Day 30). Gray line on CO2 graph means 390 µATM (boundary between

supersaturation and undersaturation).

Figure 4: A) Effect size (mean and confidence interval) of benthivorous fish with and

without access to the sediment over response CO2 variable; B) Mean values (±standard

deviation) of response variable of the treatments and reservoir during the experiment of

Total Zoo – (Day 30);

79


80


81


82


a b

83

CAPÍTULO III

84

Effects of the omnivorous fish Nile tilapia on the CO2 emission in

eutrophic lakes

Caroline Gabriela Bezerra de Moura1, Danhyhelton Douglas

2, Maria Marcolina

Cardoso2, Fabiana Araújo

3, Mariana Amaral da Costa

2, Pablo Rubim

2, José Luiz de

Attayde2

and André Megali Amado1.







3 - Departamento de Engenharia Civil




Key – words: carbon balance, omnivory, Nile tilapia.

85

Abstract

The Nile tilapia is an omnivorous fish from African ecosystems that invaded several

lakes and rivers in other continents and that affect ecosystem processes such as primary

production, sediment and nutrient ressuspension and trophic structure in tropical

freshwater ecosystems. All these process potentially affect the carbon balance in aquatic

ecosystems. Thus, the objective of this study is whether or not the Nile Tilapia affect

carbon dioxide (CO2) emission to the atmosphere and identify which mechanism are

involved in this process in a shallow reservoir. We tested the hypothesis that tilapia

attenuates the emission of CO2 into the atmosphere by increasing primary production.

For this, we conduct an experiment manipulating the presence and absence of Nile

tilapia and its sediment access to understand the mechanism by which this filter-feeding

fish affect the balance of CO2 in lakes. Our hypothesis was supported, once tilapia

decreased the CO2 concentration in the water. The main route to phytoplankton

stimulation and consequent CO2 sink in this study was the trophic cascade via reducing

zooplankton biomass and input of phosphorus via excretion by tilapia. The nutrients

input via sediment resuspension was not an important route to influence the CO2 flux in

this experiment.

86

Introduction

The inlad aquatic environments are important actors in the global carbon cycle

and may act as sources or synks of carbon dioxide (CO2) from the atmosphere (Cole et

al., 2007). When respiration rates are greater than the primary production rates, the

environment may function as CO2 source (heterotrophic). However, when the primary

production rates are greater than the system's respiration rates, aquatic environments can

function as CO2 synk (autotrophic) from the atmosphere (Cole et al., 2000).

Most lakes and reservoirs are characterized as CO2 sources to the atmosphere

(Cole et al. 1994; Duarte and Prairie, 2005; Cole et al, 2007; Tranvik et al., 2009). One

of the main factors that justify this behavior is the carbon input from allochthonous

dissolved organic matter and the resulting stimulation of the carbon mineralization by

microbial organisms, against the primary production (Cole et al., 1994; Dodds and Cole,

2007). It was recently shown that ecosystems with high primary production (considered

as eutrophic) may function as net autotrophic (Pacheco et al., 2013). In contrast, in the

tropics most reservoirs and lakes work as a source of CO2 to the atmosphere in even

greater intensity than the world`s average (Rickey et al, 2002;. Marotta et al, 2009;..

Barros et al, 2011). However, some recent studies showed that even in environments

with high primary productivity (eutrophic ecosystems), the net heterotrophy is

prevalente in semi-arid regions of low latitude (between 5 and 7 ° S) in northeastern

Brazil (Dantas et al, 2015.; Junger, et al., 2015). Some factors, such as high rates of

microbial respiration and high temperatures can justify the pattern of heterotrophy

demonstrated in this part of the world (Amado et al., 2013).

Nutrient concentrations (eg. nitrogen and phosphorus) and organic matter are

key factors in regulating the balance between primary production and decomposition. In

87

addition, the composition of organisms in an ecosystem and its trophic structure can

also be extremely relevant to their carbon balance (Schindler et al., 1997). For example,

the presence of piscivorous fish in lakes can increase the zooplankton biomass by

trophic cascade effect and therefore inhibit the phytoplankton biomass stimulating

consequently the emission of CO2 to the atmosphere. Furthermore, the predominance of

zooplanktivorous fish can increase the phytoplankton biomass and reduce the emission

of CO2 to the atmosphere (Schindler et al., 1997).

In ecosystems with predominance of omnivorous fish the trophic cascades

effects to CO2 balance can become more complex and less straightforward. Omnivorous

fish can control the zooplankton biomass thus contribute to increased production by

reducing the primary herbivores (Okum et al., 2008). On the other hand, the omnivorous

fish can directly reduce the phytoplankton biomass when using this community as the

main food resource, thus reducing the primary production (Torres et al., 2015).

Furthermore, the fish excretes can increase the phytoplankton biomass and primary

production in up to 23% (Vanni et al., 2006; Domine et al., 2009). Thus, it is difficult to

predict what is the response to the CO2 balance to the omnivorous fish action in aquatic

ecosystems, consequently.

The Nile tilapia is an exotic omnivorous fish wide distributed in the tropics,

which was introduced in semiarid reservoirs in northeastern Brazil in the decade of

1970 (Attayde et al., 2011). It originates from Africa and has some features that can

make it dominant in South America where it was introduced (Attayde et al., 2011). The

Nile Tilapia in an omnivorous filter-feeding fish, and eventual detritivorous. They build

ther nests in the bottom, which causes sediment ressuspension (bioturbation),

(Getachew and Fernando, 1989; Beveridge et al., 2000; Starling et al., 2002). Overall,

88

the Nile tilapia can affect the ecosystem metabolism, promoting eutrophication in

several tropical reservoirs where it was introduced (Starling et al, 2002; Lazzaro et al.,

2003).

According to the feeding habits and the nesting behavior of the Nile Tilapia, it is

not clear what is the net effect of this fish to aquatic ecosystems metabolism. Thus, we

hypothesized some possible mechanisms by which the Tilapia can affect the carbon

balance in aquatic ecosystems:

1) By zooplankton consumption, tilapia can increase phytoplankton biomass and

production and consequently increase the consumption of CO2 and reduce the emission

to the atmosphere.

2) Through the zooplankton consumption, tilapia can increase the biomass of bacteria

and flagellates, and consequently increase respiration rates and CO2 emissions.

3) Through the consumption of phytoplankton, tilapia can decrease primary production

and increase the emission of CO2 to the atmosphere.

4) Through sediment resuspension tilapia can increase the concentration of nutrients in

the water column, increasing the phytoplankton primary production and reducing the

emission of CO2 into the atmosphere.

5) Through the sediment resuspension tilapia can also increase the inorganic turbidity,

inhibiting phytoplankton primary production and increasing the emission of CO2 to the

atmosphere.

6) Tilapia can release nutrientes in the water through excretion increasing the primary

production and reducing the emission of CO2 to the atmosphere.

89

The aim of this study is evaluate wheather or not the Nile Tilapia affect the

carbon balance in reservoirs and understanding the mechanisms that play this role.

Since the presence of tilapia has been recognized as a factor that contributes to the

eutrophication (Starling et al, 2002;. Lazarro et al. 2003; Attayde et al, 2011), our

hypothesis is that the Tilapia reduces the CO2 concentration in the water due to primary

production increase. To achieve this goal we performed 2 X 2 factorial mesocosms

experiment with the presence and the absence of Tilapia and with or no access to the

sediment.


Study area and experimental design

The current study was performed from January 11th

to February 10th

in 2014 in a

shallow (mean depth = 4m) and small (11 ha) reservoir situated at Seridó Ecological

Station (ESEC) in Serra Negra do Norte, Rio Grande do Norte, Brazil (06°34'852″N,

37°15'519″W). The reservoir is considered eutrophic (37 (µg.L-1

), 229 (µg.L-1

), 2150

(µg.L-1

) for chlorophyll-α [Chla], total phosphorus [TP], total nitrogen [TN],

respectively) and net heterotrophic (CO2 = 1500 [µATM]) prior the experiment.

Twenty mesocosms (depht: 2m, area: 4m2, volume: 8m

3) made of an aluminum

frame surrounded by transparent plastic (0.45 mm of thickness) were used to isolate the

inside water from the reservoir. All mesocosms were open to atmosphere and ten of the

mesocosms had a galvanized wire mesh (pvc coated) attached in the bottom part,

blocking the fish access to the sediment. The side iron bars were buried in the sediment

around 20 cm to ensure a seal along the sediment (Figure 1). The experimental design

consisted of four treatments: sediment access without fish (A-F), sediment access plus

90

fish (A+F), without sediment access without fish (N-F), without sediment access plus

fish (N+F). The treatments were replicated five times and randomly assigned. The

treatments with fish were set by adding 3 individuals of Nile tilapia (Oreochromis

niloticus; final density of 0,37 individual per cubic meter), accordingly to densities of

this fish reported in natural lakes and rivers of the Brazilian semi-arid (Gurgel and

Fernando, 1994). The fishes were caught at the same reservoir right before experiment

begins (ESEC reservoir).

The treatments setup was performed immediately after the placement of

mesocosms structures. The experiment lasted for 30 days and the samplings during the

experiment were performed in the beginning, fifteen days and at the end of the

experiment: day 1 (right before the fish addition), day 15 and day 30 (end of

experiment). The parameters monitored during the experiment in the mesocosms were:

concentration of carbon dioxide (CO2), total plankton respiration (PR), bacterial

respiration (BR), Chlorophyll - α concentration (Chla), total organic carbon (TOC),

total phosphorus (TP), total nitrogen (TN), dissolved oxygen (DO), water transparence

(Secchi depth), total suspended solid (TSS), suspended volatile solid (SVS), suspended

fixed solid (SFS), bacterial abundance (BA), heterotrophic flagellates abundance (HNF)

and zooplankton abundance (Total Zoo).

Sampling

At first, in each mesocosm the water temperature and dissolved oxygen profile were

measured at water column using a portable oxygen meter (Instrutherm MO-900). We

also took water samples for CO2 concentration determination through the alkalinity

method. For CO2, water samples were carefully taken in the sub-surface using a

polycarbonate bottle (100mL) ensuring the complete removal of internal atmosphere or

91

bubbles. Samples were taken to the field laboratory for measurements (c.a. 30 min). The

secchi disk depth was measured as light attenuation in each mesocosm. Finally, for each

mesocosm we collected water samples with a 1.5m long tube (through water column) at

5 random points into each mesocosm and integrated in a plastic bucket (20L) to

subsample for: PR, BR, Chla, TOC, TP, TN, DO, TSS, SFS, SVS, BA, HNF. Twenty

liters of water were filtered through a plankton net (64µm of mesh size) to estimate the

zooplankton abundance.

Analyses

The CO2 was estimated from the pH and alkalinity performed titration using

0.02N H2SO4, adjusting for temperature, ionic strength and air pressure (as described in

Cole et al., 1994). Subsequently, the results were expressed as undersaturated or

supersaturated with CO2 relative to the atmosphere (considered here as being 390

uATM). A mesocosm pilot experiment performed in june of 2010 (unpublished data),

showed that CO2 concentrations decreased with time and become close to the

atmospheric balance range (390 µATM) independent of treatment in the 30 day of

experiment. This same pattern was observed in this study, and thus, we assume it as a

experimental setup artifact (e.g. changing the effects of wind on water turbulence, etc.).


while BR rates were estimated as oxygen consumption in filtered through glass fiber

filtered (1.2µm average pore size; VWR INTERNATIONAL) water using a golden tip

oxygen microsensor connected to a picoamperimeter controlled by the MicOx software

(Unisence ©; Briand et al., 2004). The samples for PR and BR measurements were

incubated in exetainers (5.9mL; Labco®) with no internal atmosphere in 5 replicates for

each mesocosm in the dark at room temperature (25°C ± 1) for 24 hours.

92

TOC and TN concentration were measured by catalytic combustion in a Total

Organic Analyzer (TOC – V, Shimadzu – 2.0) with a TN analyzer attached (VNP

module). TOC was calculated from the sum of the dissolved organic carbon (DOC) and

particulated organic carbon (POC) (Wetzel and Likens, 2000). TP concentration was

measured with a spectrophotometer by the acid ascorbic method after persulphate

digestion (Murphy and Riley, 1962). Water samples were filtered through glass fiber

filter (VWR INTERNATIONAL – 1.2µm) for chlorophyll-α concentration, which was

extracted with ethanol 95% and measured by spectrophotometry (Jespersen and

Cristoffersen, 1987).



determined after nucleic-acid staining with Syto13 (Molecular Probe; final

concentration 2.5 µM; del Giorgio et al., 1996). Fluorescent latex beads (Polysciences,

1.5µm diameter) were added to each sample for calibration of side scatter and green

fluorescence signals, and as an internal standard for the cytograms.




epifluorescence microscope (Porter and Feig., 1980). On average 400 individuals were


The zooplankton organisms were counted under a microscope in a 1 mL Sedwick-

Rafter chamber. Between three and five subsamples were counted for each sample

collected in the field until a minimum of 100 individuals of each taxonomic group had

been counted. Subsequently, the average of the subsamples was taken for each group of

93

organisms counted, this being multiplied by the sample volume (mL) and divided by the

subsample volume (1 mL) to estimate the total number of individuals in the sample.

Afterwards, the number of individuals in the sample was divided by the water volume

(L) sampled in the Field to calculate the original density (Ind. L-1

) of organisms in the

sample.

Total and inorganic suspended solids were determined by weight after drying the

filters overnight at 100 ºC and ignition of filters at 500ºC for three hours, respectively

(APHA, 1998). The organic suspended solids were measured by the difference between

total suspended solids and inorganic suspended solids (APHA, 1998). The fixed

suspended solids (SFS) were used as a proxy of inorganic turbidity and the volatile

suspended solids (SVS) as a proxy of organic turbidity.

We quantified nutrient excretion rates of the fishes at the end of the experiment.

We randomly removed one single individual from each mesocosm of the juvenile fish

treatments and put them separately in plastic bags filled with distilled water to release

nutrients for one hour (as described by Schaus et al., 1997, with modifications). Water

samples were then collected from each plastic bag and used to quantify phosphorus

concentration excreted by Nile tilapia.

Statistical analysis

To analyze the effect of fish and its access to sediment over dependent variables

we used the two – way ANOVA. The values have been averaged (between day 15 and

day 30) following the proposed by Shaus and Vanni (2000). Prior to analyses, data were

log transformed to stabilize variances among treatments (homogeneity). Homogeneity

of variance was tested by Levene’s Test, and a significant level of α = 0.05 was

assumed. To understand the differential fish effect with and without sediment access,

94

we estimate the effect size using the log ratio [ln(experiment/control)]. Two controls

were stablisehd: A-F as the control for A+F, and N-F as the control for N+F. The effect

size was evaluated crossing all replicates of the experiment with replicates of the

respective control creating 25 effect sizes for each dependent variable building up a

histogram of distribution. The effect size is considered significant when the confidence

interval does not cross the zero in the Y axis. Positive results are showed when the

differences between treatments were higher than 0. Negative results are showed when

the differences between treatments were lower than 0. Then, the identified distribution

was used in a bootstrap analysis generating the confidence interval. The distribution was

accessed by the “fitdistrplus” package and the confidence interval by the boot package.

We used the R software (R Development Core Team, 2011) for all analysis.

Results

No pre-treatment variations were found among the mesocosms prior to the fish

addition. In the beginning (Day 1), the water in each mesocosm and in the reservoir was

supersaturated in CO2 (mean=13.000µATM; 9.000µATM, respectively) (Figure 3).

Along the experiment, all mesocosms reduced the CO2 concentration (mean=4.000

µATM), as well as the water in the reservoir (2.000µATM).

Higher values of chlorophyll-a, TOC, TN, SFS, C:P ratio, N:P ratio and lower

values of TP were found for the treatments with fish (Figure 3). The two-way ANOVA

results showed significant effects of the tilapia presence in the CO2, Chla, Total Zoo,

BR, TOC, C:P and N:P ratio (Table 1). However, these effects were independent of

sediment access. Furthermore, the access to the sediment, independently of the fish

presence, affected significantly the DO, TP, WT, C:P ratio, N:P ratio, SFS and SVS

95

(Table 1). There were no interaction results between sediment access and fish presence

for the studied variables (Table 1).

The analysis of effect size showed that the presence of the Tilapia significantly

decreased the CO2, total zooplankton abundance and TP independent of sediment

access. On the other hand, the presence of Tilapia increased the chla, PR, TOC, C:P,

N:P, DO. BR, WT and BA increased when fish did not have access to the sediment. The

SFS and TN increased when fish had access to the sediment (Figure 2).

The excretion rates estimated showed that each individual of Tilapia released on

average of 200µg.L-1

h-1

of phosphorous in the water.

Discussion

In the current work, we investigated the role of the Nile Tilapia to the CO2

balance in a shallow semi-arid reservoir. Our hypothesis that the Nile tilapia decreases

CO2 concentration in the water was confirmed. This reduction of CO2 was directly

related to increased primary production and higher phytoplankton biomass (high Chla

concentrations)

in the treatments where the tilapia individuals were present. We

described six mechanisms that possibly drive the relationship between tilapia and the

CO2 balance in the water column (see the introduction section), but our results suggest

that two of those mechanisms should be able to explain the recorded pattern: (a) trophic

cascade reducing the zooplankton biomass, which would reduce the grazing pressure on

phytoplankton community and, (b) phytoplankton primary production enhancement due

to nutrients release from fish excretes.

The top down effect of Tilapia can be one of the causes the decrease of CO2 in

this experiment, and corroborate the hypothesis number 1 showed in the introduction in

96

this work. The Tilapia decreased total zooplankton biomass (Figure 4) and consequently

increased the phytoplankton primary production and biomass (Figure 3). Thus, the

increase of the phytoplankton is likely to be an effect of trophic cascade as the presence

of the fish affected reciprocal predator-prey abundance and metabolism across two links

in the food web (Pace et al. 1999).

Another hypothesis in the introduction was that the top-down effect of Tilapia

would have a negative effect on zooplankton and this positively affect the bacterial

community and flagellates (Pace and Cole, 1996; Jurgens and Matz, 2002), and thus a

higher contribution in CO2 emissions. However, we find no effect on flagellate

community or in the bacteria. Perhaps because it is an omnivorous fish in which the

effect of trophic cascade was not strong enough (Okum et al., 2008) to be observed at

lower trophic levels. Thus, the hypothesis number 2 was no corroborated.

The hypothesis number 3 states that Nile Tilapia through the consumption of

phytoplankton, can decrease primary production (Torres et al., 2015) and consequently

increase the emission of CO2 to the atmosphere. However, in this work the Nile Tilápia

stimulate the phytoplankton biomass as previously described in the literature (Starling et

al., 2002; Lazzaro et al., 2003; Okum et al., 2008). Thus, this hypothesis was no

corroborated.

Despite the initial reduction in CO2 emissions in the presence of the filter-

feeding fish, the water in mesocosms and in the reservoir tended to heterotrophy even in

the eutrophic conditions. The remarkable high suspension inorganic solids of the

reservoir have the potential to inhibit the phytoplankton growth (Wahl et al., 2011). As

we show, the treatments with fish had higher inorganic suspended solids than in the fish

absence. Furthermore, the SFS seems to have an important role in fostering

97

heterotrophy in lakes (Moura et al. – Chapter 2). The sediment resuspension can

increase organic and inorganic nutrients, and can stimulate the heterotrophic production,

as well as decrease the autotrophic production by decrease the light (Cotner et al., 2000;

Alongi et al., 2003). Besides, the nutrients release from the sediment that would

stimulate the primary production could be an alternative mechanism to explain the

effect of tilapia to the CO2 balance. In fact, these mechanisms were ruled out in the

current study since the tilapia access to the sediment (no interaction effect) did not

present significant response to the CO2 or phytoplankton biomass (Table 1; Figure 2).

However, this mechanism should not be ruled out in the lake conditions, since the short

duration of the mesocosms and experimental conditions did not allow the fish to

perform the nesting behavior, which results in the bioturbation. Thus, under the

experimental conditions our hyphotesis number 4 and 5 were no corroborated.

The stimulation of filter-feeding fishes to phytoplankton biomass has been

demonstrated and occurs through the increase in the nutrient supply via excretion

(Domine et al., 2009). The estimated excretion rates showed that each individual of

Tilapia released an average of 200µg.L-1

h-1

of TP. However, we did not record any P

increase in the water column, probably because P was immediately absorbed by the

plankton communities and incorporated in the biomass of phytoplankton or/and

bacterioplankton (Domine et al., 2009; Fonte et al., 2011). To ensure this mechanism it

would be necessary to perform microcosm experiments to follow P excretion going to

the primary producers. However, our results showed that the Nile Tilapia can release

great amount of nutrients (eg. TP) in the water through excretion, and it was coincident

to the increasing in the primary production and reducing the emission of CO2 to the

atmosphere. Thus, we can not rule this mechanism (hypothesis number 6). It probably

98

occurs concomitantly with the top-down control of the zooplankton and both factors

converge to the autotrophic metabolism response of the mesocosms.

Finally, our work showed that the Nile Tilapia trigger mechanisms to promote

the CO2 sink from the atmosphere, by stimulating the phytoplankton biomass and

primary production. Despite the fact that the Nile Tilapia is an exotic omninovorous

fish and cause some problems to the quality of water (e.g. eutrophication) (Starling et

al., 2002; Attayde et al., 2011) it work in positive feedback to carbon cycle promoting

the CO2 sink from the atmosphere.

99

Reference

Alongi, D.M., Chong, V.C., Dixon, P., Sasekumar, A. and Tirendi, F., 2003, “The

influence of fish cage aquaculture on pelagic carbon flow and water chemistry in tidally

dominated mangrove estuaries of peninsular Malaysia”. Marine Environmental

Research, v. 55, pp. 313-333.




8.

APHA, 1998, “Standard Methods for the Examination of Water and Wastewater”, 20th

eds. American Public Health Association, Washington DC.

Attayde, J.L., Brasil, J. and Menescal, R.A., 2011, “Impacts of introducing Nile tilapia

on the fisheries of a tropical reservoir in North-eastern Brazil”, Fisheries Management

and Ecology, v.18, pp. 437 – 443.

Barros, N., Cole, J.J., Tranvik, L.J., Prairie, Y.T., Bastviken, D., Huszar, V.L.M., del

Giorgio, P. and Roland, F., 2011, “Carbon emission from hydroelectric reservoirs linked

to reservoir age and latitude”. DOI: 10.1038/NGEO1211.

Beveridge, M.C.M. and Baird, D.J., 2000, “Diet, feeding and digestive physiology. In

Beveridge, M.C.M. and McAndrew, B.J. (eds), Tilapias: Biology and Exploitation.

Kluwer Academic Publishers, Belgium, pp. 59 – 87.

Biddanda, B., Ogdahl, M. and Cotner, J., 2001, “Dominance of bacterial metabolism in

oligotrophic relative to eutrophic waters”. Limnology and Oceanography, v.46, pp. 730-

739.

100




Carpenter, S.R. and Kitchell, J.F., 1988, “Consumer control of lake productivity”.

Bioscience, v.38, pp. 764 - 769.


supersaturation in the surface waters of lakes”, Science, v.265, pp. 1568-1570.




Cotner, J.B.,2000, “Intense winter heterotrophic production stimulated by benthic

resuspension”, Limnology and Oceanography, v.45, pp. 1672 – 1676.

de Moura, C.G.B., 2015, “Mecanismos de emissão de CO2 em reservatórios do

semiárido brasileiro”. Capítulo 2, Thesis.

Del Giorgio, P.A., Bird, D.F., Prairie, Y.T. and Planas, D., 1996, “Flow cytometric



del Giorgio, P.A., Cole, J.J. and Cimbleris, A., 1997, “Respiration rates in bacteria


148-151.




101

Domine, L.M., Vanni, M.J. and Renwick, W.H., 2009 “New and regenerated primary

production in a productive reservoir ecosystem”. Can. J. Fish. Aquat. Sci., v.67, pp.

278-287.



Figueredo, C.C. and Giani, A., 2005, “Ecological interactions between Nile Tilapia

(Oreochromis niloticus, L.) and the phytoplankton community of the Furnas Reservoir

(Brasil)”. Freshwater Biology, v. 50, pp. 1391-1403.

Fonte, E.S., Carneiro, L.S., Caliman, A., Bozelli, R.L., Esteves, F.A. and Farjalla, V.F.,

2011 “Effects of resources and food wed structure on bacterioplankton production ina

tropical humic lagoon”. Journal of Plankton Research, v.33, pp. 1596-1605.

Getachew, T. and Fernando, C.H., 1989, “The food habitats of an herbivorous fish

(oreochromis niloticus Linn.) in Lake Awasa, Ethiopia”. Hydrobiologia, v. 174, pp.

195-200.

Gurgel, J.J.S. and Fernando, C.H., 1994, “Fisheries in Semi-Arid Northeast Brazil with

special reference to the role of Tilapias”. Int.Revue ges. Hydrobiol, v.79, pp.77-94.


phytoplankton using ethanol as extraction solvent”, Archiv fuer Hydrobiologie

AHYBA4, v.109, pp. 445-454.




102

Jurgens, K. and Matz, C., 2002, “Predation as a shaping force for the phenotypic nad

genotypic composition of planktonic bactéria”. Aontonie van Leeuwenhoek, v.81, pp.

413-434.








Okum, N., Brasil, J., Attayde, J.L. and Costa, I.A.S., 2008, “Omnivory does not prevent

trophic cascades in pelagic food webs”. Freshwater Biology, v.53, pp. 129-138.

Pace, M.L. and Cole, J.J., 1996, “Regulation of bacteria by resources and predation

tested in whole-lake experiments. Limnology and Oceanography, v.41, pp. 1448-1460.

Pace, M.L., Cole, J.J., Carpenter, S.R. and Kitchell, J.F., 1999, “Trophic cascades

revealed in diverse ecosystems”. TREE, v.14.





Rickey, J.E., Melack, J.M., Aufdenkampe, A.K., Ballester, V.M. and Hess, L.L., 2002,


atmospheric CO2”. Nature, v. 416, pp. 617-620.

103

Schaus, M.H., Vanni, M.J., Wissing, T.E., Bremigan, M.T., Garvey, J.E. and Stein,

R.A., 1997, “Nitrogen and phosphorus excretion by detritivorous gizzard shad in a

reservoir ecosystem”. Limnology and Oceanography, v. 42, pp. 1386-1397.

Schaus, M.H. and Vanni, M.J., 2000, “Effects of gizzard shad on phytoplankton and

nutrient dynamics: role of sediment feeding and fish size”. Ecology, v.81, pp. 1701-

1719.







Torres, G.S., Silva, L.H.S., Rangel, L.M., Attayde, J.L. and Huszar, V.L.M., 2015,

“Cyanobacteria are controlled by omnivorous filter-feeding fish (Nile Tilapia) in a

tropical eutrophic reservoir. Hydrobiologia. DOI: 10.1007/s10750-015-2106-y.




Valderrama, J.C.1981. The simultaneous analysis of total N and total P in natural

waters, Mar.Chem. v.10, 109-122.

Vanni, M.J., Bowling, A.M., Dickman, E.M., Hale, R.S., Higgins, K.A., Horgan, M.J.,

Knoll, L.B., Renwick, W.H., and Stein, R.A. 2006, “Nutrient cycling by omnivorous

fish supports an increasing proportion of lake primary production as ecosystem

productivity increases. Ecology, v. 87, pp. 1696–1709.

104



Hydrobiologia, v.678, pp. 49 – 63.

Wetzel, R.G. and Likens, G.E., 2000, “Limnological Analyses, Springer-Verlag.

105



Variables A F A X F

F - ratio P F – ratio P F - ratio P

CO2 (µATM) 2.35 0.144 5.54 0.031* 1.15 0.297

Chla (µg.L-1

) 0.01 0.972 20.91 0.000* 0.06 0.801

BR (µmol.L-1

.h-1

) 0.32 0.576 4.74 0.045* 1.95 0.182

PR (µmol.L-1

.h-1

) 3.81 0.068 0.58 0.456 0.02 0.887

TOC (mg.L-1

) 0.62 0.443 4.97 0.040* 0.35 0.559

TN (µg.L-1

) 0.25 0.623 1.31 0.268 0.53 0.473

TP (µg.L-1

) 15.98 0.001* 3.31 0.087 0.19 0.666

Secchi deapht (m) 24.64 0.000* 0.914 0.353 1.132 0.303

DO (mg.L-1

) 6.09 0.025 2.36 0.143 0.18 0.669

C:N Ratio 0.02 0.886 0 0.943 0.79 0.386

C:P Ratio 7.81 0.012* 6.68 0.019* 0.42 0.526

N:P Ratio 10.67 0.004* 6.63 0.020* 0.04 0.834

SFS 9.41 0.007* 0.37 0.547 2.13 0.163

SVS 6.56 0.020* 0.072 0.79 0.372 0.55

BA (Cell.mL-1

) † 0.08 0.785 0.47 0.5 3.12 0.096

HNF (Cell.mL-1

) † 2.38 0.142 0.41 0.528 0 0.965

Total Zoo (Ind.L-1

) 0.778 0.394 13.639 0.003* 1.434 0.254

* p-value equal or lower than 0.05.

†Only on the last sample date.

106

Figure legends:


in this study.

Figure 2: Effect size (mean and confidence interval) of fish with and without access to

the sediment over response variables. Treatments legends: effect of fish without

sediment access ( ); effect of fish with sediment access ( ). A) pCO2; B) Chl-a; C)



ratio; L) Dissolved oxygen; M) Bacterial abundance; N) Flagellate abundance; O)

SFS;P)SVS.


reservoir during the experiment. Treatments legends: without sediment access and

without fish ( ), without sediment access but with fish ( ), sediment access without

fish ( ), sediment access with fish ( ), and reservoir ( ). A) pCO2; B) Chl-a; C)



ratio; L) Dissolved oxygen; M) Bacterial abundance; N); Flagellate abundance O) SFS.

Figure 4: A) Effect size (mean and confidence interval) of fish with and without access

to the sediment over response of Total Zoo variable; B) Mean values (±standard

deviation) of response Total Zoo variable of the treatments and reservoir during the

experiment.

107



108


109


a b

110

Considerações Finais:

Os resultados desta tese nos mostraram que peixes com diferentes hábitos

alimentares podem influenciar o balanço de carbono em reservatórios do semiárido do

nordeste brasileiro (Capítulo I; Capítulo II; Capitulo III). Demonstramos através de

experimentos de mesocosmos realizados em reservatório que peixes bentívoros

(detritívoros) aumentam a heterotrofia e emissão de CO2 para atmosfera, através da

ressuspensão de matéria orgânica e nutrientes presos ao sedimento, que estimulam as

taxas de respiração planctônica e microbiana, assim como os processos de metanotrofia

(Capítulo II). Por outro lado, peixes onívoros como a Tilápia do Nilo, favorecem a

diminução da emissão de CO2 para a atmosfera, através do estímulo da biomassa

fitoplanctônica ocasionado principalmente via cascata trófica pela diminuição da

biomassa de zooplâncton (Capítulo III). Além disso, reservatórios que apresentam uma

dominância de sólidos inorgânicos em suspensão pode indicar que o ambiente está

emitindo CO2 para a atmosfera. Em contrapartida, reservatórios que apresentam uma

dominância de sólidos orgânicos em suspensão pode indicar que o ambiente esteja

apreendendo CO2 da atmosfera (Capítulo I). Podemos concluir, que alguns fatores como

a dominância de sólidos em suspensão pode ser um indicativo da função do ecossistema

aquático frente ao balanço de carbono. Além disso, peixes com diferentes hábitos

alimentares podem influenciar o balanço de carbono de lagos.

111

Reference




8.

Attayde, J.L., Brasil, J. and Menescal, R.A., 2011, “Impacts of introducing Nile tilapia

on the fisheries of a tropical reservoir in North-eastern Brazil”, Fisheries Management

and Ecology, v.18, pp. 437 – 443.

Azam, F., Fenchel, T., Field, J,D., Gray, J.S., Meyer-Reil, L.A. and Thingstad, F., 1983,

“The ecological role of water – column microbes in the sea”, Mar. Ecol. Prog. Ser.,

v.10, pp. 257-263.

Barbosa, J.E.L., Medeiros, E.S.F., Brasil, J., Cordeiro, R.S., Crispim, M.C.B. and da

Silva, G.H.G., 2012, “Aquatic systems in semiarid Brazil: limnology and management”.

Acta Limnologica Brasiliensia.

Barros, N., Cole, J.J., Tranvik, L.J., Prairie, Y.T., Bastviken, D., Huszar, V.L.M., del

Giorgio, P. and Roland, F., 2011, “Carbon emission from hydroelectric reservoirs linked

to reservoir age and latitude”. DOI: 10.1038/NGEO1211.

Beveridge, M.C.M. and Baird, D.J., 2000, “Diet, feeding and digestive physiology. In

Beveridge, M.C.M. and McAndrew, B.J. (eds), Tilapias: Biology and Exploitation.

Kluwer Academic Publishers, Belgium, pp. 59 – 87.

112

Bilotta, G.S. and Brazier, R.E., 2008, “Understanding the influence of suspended solids

on water quality and aquatic biota”. Water Research, v.42, pp. 2849-2861.




Bouvy, M., Nascimento, S.M., Molica, R.J.R., Ferreira, A., Huszar, V. and Azevedo,

S.M.F.O., 2003, “Limnological features in Tapacurá reservoir (northeast Brasil) during

a severe drought.”. Hydrobiologia, v.493, pp. 115-130.

Braga, G.G., Becker, V., de Oliveira, J.N.P., Mendonça Júnior, J.R., Bezerra, A.F.M.,

Torres, L.M., Galvão, A.M.F. and Mattos, A., 2015, “Influence of extended drought on

water quality in tropical reservoirs in a semiarid region”. Acta Limnologica Brasiliensia,

v.27, pp. 15 – 23.











113

Costa, M.R.A.C., Attayde, J.L. and Becker, V., 2016, “Effects of water level reduction

on the dynamics of phytoplankton functional groups in tropical semiarid shallow lake.”.

Hydrobiologia.




Domine, L.M., Vanni, M.J. and Renwick, W.H., “New and regenerated primary

production in a productive reservoir ecosystem”. Can. J. Fish. Aquat. Sci., v.67, pp.

278-287.



Figueiredo – Barros, M.P., Caliman, A., Leal, J.J.F., Bozelli, R.L., Farjalla, V.F. and

Esteves, F.A., 2009, “Benthic bioturbator enhances CH4 fluxes among aquatic

compartments and atmosphere in experimental microcosms”. Can. J. Fish. Aquat. Sci.,

v. 66, pp. 1649-1657.

Freitas, F.R.S., Rhighetto, A.M. and Attayde, J.L., 2011, “Cargas de fósforo total e

material em suspensão em um reservatório do semiárido brasileiro”. Oecologia

Australis, v. 15, pp. 655 – 665.

González-Bergonzoni, I., Meerhoff, M., Davidson, T.A., Teixeira-de-Melo, F.,

Baattrup-Pederson, A. and Jeppesen, E., 2012, “Meta-analysis shows a consistent and

114

strong latitudinal pattern in fish omnivory across ecosystems”. Ecosystems. DOI:

10.1007/s10021-012-9524-4.



Jeppesen, E., Meerhoff, M., Holmgren, K., et al., 2010, “Impacts of climate warming on

lake fish community structure and potential effects on ecosystems function”.





Kosten, S., Roland, F., Da Motta Marques, D.M.L., Van Nes, E.H., Mazzeo, N.,

Sternberg, L.S.L., Scheffer, M. and Cole, J.J., 2010, “Climate-dependent CO2

emissions from lakes”. Golbal Biogeochemical Cycles, v. 24. GB2007.




Liu, X., Wu, Q., Chen, Y. and Dokulil, M.T., 2011, “Imbalance of plankton community

metabolism in eutrophic lake Taihu, China”. Journal of Great lakes Research, v.37, pp.

650 – 655.

Marengo, J.A., et al., 2010, “Future change of climate in South America in the late

twenty-first century: intercomparison of scenarios from threen regional climate

models.”.Clim.Dyn., v.35, pp. 1073-1097.

115



Marotta, H., Duarte, C.M., Guimarães-Souza, B.A. and Enrich-Prast, A., 2012,

“Synergistic control of CO2 emissions by fish and nutrients in a humic tropical lake”.

Oecologia, v. 168, pp. 839-847.

Medeiros, L.C., Mattos, A., Lurling, M. and Becker, V., 2015, “Is the future blue-green

or brown? The effects of extreme events on phytoplankton dynamics in a semiarid man-

made lake”. Aquatic Ecology. DOI: 10.1007/s10452-015-9524-5.

Mendonça, R., Kosten, S., Sobek, S., Cole, J.J., Bastos, A.C., Albuquerque, A.L.,

Cardoso, S. J. and Roland, F., 2014, “Carbon sequestration in a large hydroelectric

reservoir: an integrative seismic approach”. Ecosystems, v.17, pp. 430-441.

Menezes, R.F., Attayde, J.L. and Vasconcelos, F.R., 2010, “Effects of omnivorous

filter-feeding fish and nutrient enrichment on the plankton community and water

transparency of a tropical reservoir”, Freshwater Biology, v. 55, pp. 767 – 779.

Moss, B., et al., 2011, “Allied attack: climate change and eutrophication.”. Inland

Waters. v. 1, pp. 101-105.

Odum, H.T., 1956, “Primary production in flowing waters.”Limnol Oceanogr, v.1, pp.

112-7.



116

Rickey, J.E., Melack, J.M., Aufdenkampe, A.K., Ballester, V.M. and Hess, L.L., 2002,


atmospheric CO2”. Nature, v. 416, pp. 617-620.

Roland, F., Huzar, V.L.M., Farjalla, V.F., Enrich-Prast, A., Amado, A.M. and Ometto,

J.P.H.B., 2012, “Climate change in Brazil: perspective on the biogeochemistry of inland

waters.”. Braz.J.Biol., v. 72, pp. 709-722.

Sarmento, H., Amado, A.M., and Descy, J.P., 2013, “Climate change in tropical fresh

waters (comment on the paper`Plankton dynamics under different climatic conditions

in space and time` by De Senerpont Domis et al., 2013”. DOI: 10.1111/fwb.12140.

Scheffer, M., Hosper, S.H., Meijer, M-L, Moss, B. and Jeppesen, E., 1993, “Alternative

equilibria in shallow lakes”. TREE, v.8.

Scheffer, M., 2004, “Ecology of shallow lakes”. ed. Springer, B.V., p. 378.




Smith, V.H. and Schindler, D.W., 2009, “Eutrophication science: where do we go from

here?”. TREE, v. 24.

Smith, E.M. and Kemp, M., 2003, “planktonic and bacterial respiration along an

estuarine gradient: responses to carbon and nutrient enrichment”, Aquatic Microbial

Ecology. v. 30, pp. 251-261.

117

Soeken-Gittinger, L.A., Stoeckel, J.A. and Havel, J.E., 2009, “Differing effects of

suspended sediments on the performance of native and exotic Daphnia.”, v.54, pp. 495-

504.










Vanni, M.J., Layne, C.D. and Arnott, S.E., 1997, “Top- down trophic interactions in

lakes: effects of fish on nutrient dynamics”. Ecology, v. 78, pp. 1-20.




Xu, Y., Cai, Q., Shao, M., Han, X., and Cao, M., 2009, “Seasonal dynamics of

suspended solids in a giant subtropical reservoir (China) in relation to internal processes

and hydrological features”. Quartenary International, v. 208, pp. 138-144.

Documents

Mecanismos de emissão de CO em reservatórios do semiárido ... · Agradeço à Carlos Eduardo Alencar (Cadú), pelo companheirismo e ajuda em momentos decisivos deste trabalho e