82 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

Altitudinal distribution of the functional feeding groups of aquatic

macroinvertebrates using an ecological network in Andean streams

Juan Pablo Serna1*; https://orcid.org/0000-0002-2805-8218

David Fernandez1; https://orcid.org/0000-0003-4898-4144

Fabio Velez1; https://orcid.org/0000-0001-6348-6405

Julián Ruiz1; http://orcid.org/0000-0001-5863-9761

Broder Breckling2

Néstor Aguirre1; https://orcid.org/0000-0002-0847-7335

1. Facultad de Ingeniería, Universidad de Antioquia, Medellín, Colombia; jpablo.serna@udea.edu.co (Correspondence*),

david.fernandez@udea.edu.co, fabio.velez@udea.edu.co, julian.ruiz@udea.edu.co, nestor.aguirre@udea.edu.co

2. University of Bremen, General and Theoretical Ecology, broder@uni-bremen.de

Received 07-V-2021. Corrected 18-XI-2021. Accepted 31-I-2022.

ABSTRACT

Introduction: Analysis of functional feeding groups (FFG) in aquatic macroinvertebrates is important in

understanding the structure, function, and dynamics of ecological processes in ecosystems. Modularity refers

to the degree of compartmentalization of food webs and varies between -1 and 1. A network with a modular-

ity value close to 1 is resilient to disturbances and can be interpreted as an indicating factor for the stability of

communities.

Objective: In this study, we analyzed the trophic structure of benthic macroinvertebrates in La Nitrera stream,

the San Juan River, and the Cauca River in the Colombian Andes.

Methods: The study was supported by ecological networking techniques using Gephi software. We studied nine

sites in dry, rainy, and transition seasons in 2017 and 2018, monitoring changes in the altitude gradient. At each

of the sites, the organisms were captured and determined, and physicochemical and hydraulic information was

obtained.

Results: The variance component analysis allowed to explain the variability of the data by relating the following

environmental variables: FFG, diversity, richness, modularity, season, and time. Simple multifactorial ANOVA

indicated that significant changes in FFG were associated with altitude, and modularity to time. The allocation

of the FFG was done by stomach analysis and secondary information.

Conclusion: The transition season had the highest modularity, possibly due to the recolonization of some

biotopes caused by the decrease in the velocity of water currents. La Nitrera and San Juan presented higher

values than the Cauca, which may indicate that the altitudinal change and velocity of water currents affects the

compartmentalization of the network.

Key words: modularity; altitudinal gradient; stomach analysis; network analysis; invertebrate trophic structure.

https://doi.org/10.15517/rev.biol.trop..v70i1.46904

AQUATIC INVERTEBRATES

Network analysis is an important element

in understanding the structure, function, and

dynamics of ecological systems. These systems

are complex because they involve relationships

and interactions that are not easily quantifi-

able. However, progress has been made in

understanding some interactions whose iden-

tity may change in space and time, such as

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

food webs, patch dynamics, and population

fluctuations (Anand et al., 2010). Although

networks have provided fundamental represen-

tations of ecological complexity, they need to

be extended to systematically and simultane-

ously capture these multifaceted interactions

(Pilosof et al., 2017).

The importance of aquatic macroinverte-

brates as indicators of environmental conditions

lies in the benthic life form of these organ-

isms, their longer life cycles compared other

aquatic organisms like bacteria and algae, and

their specialized adaptation to specific envi-

ronmental characteristics (Castillo-Figueroa et

al., 2018; Roldán-Pérez, 2016). These aspects

become a response to the assessment of the

space-time conditions of the environment in

which they are found (National Academy of

Sciences, 1988; Rosenberg & Resh, 1993). The

energy and nutrient transfer in aquatic food

webs depends mainly on the diversity, qual-

ity, and quantity of resources available in the

system, as well as on the species composition,

the number of individuals and their trophic

relationships (Cummins et al., 2005).

Food webs are important in the ecological

processes that occur in river ecosystems. As

such, information on food web relations serve

as a relevant part of the scientific basis for the

planning and management of river ecosystems

(Tamaris-Turizo et al., 2018). In the case of

aquatic macroinvertebrate communities, these

trophic relationships relate to life cycles, habi-

tat choice, behavior, predation, and other fac-

tors (Chará-Serna et al., 2010).

The complexity involved in analyzing

these interactions, especially the species clas-

sification and taxonomy in tropical rivers in

Colombia, has prompted studies in fields such

as functional ecology (Schmera et al., 2013;

Schmera et al. 2017), whereby functional traits

are assigned that allow the measurement of

morphological, physiological or phenological

characteristics at the individual level (Violle

et al., 2007). In this respect, some authors

have made advances in the analysis, study and

functional classification of the aquatic insects

feeding groups in the Neotropics (Chará-Serna

et al., 2010; Ramírez & Gutiérrez-Fonseca,

2014; Tamaris-Turizo et al., 2018; Tomanova

et al., 2006). This has improved the understand-

ing of the predator-prey relationship function

among organisms, and thereby the determina-

tion of the balance of the communities that

inhabit a biotope.

In Colombia, the altitudinal variation

influences the physicobiotic structure of the

rivers because at short distances in the Colom-

bian Andean it is possible to have changes in

the biotopes and species composition. These

changes are also accompanied by rheological

variability due to the annual cycles of rainfall

and dry seasons (Latrubesse et al., 2005). For

example, in the rainy season the rivers are

abundant and turbid, while in the dry season

the rivers significantly reduce their flow and

increase their transparency of water (Gutiér-

rez, 2018). Thus, spatial variability – meters

elevation - and variability in time, i.e., annual

rainfall cycle, influence the structure and func-

tioning of aquatic macroinvertebrate.

According to Ramírez and Gutiérrez-Fon-

seca (2014), aquatic macroinvertebrates can be

classified into six major categories. Scrapers

(Sc) generally live attached to rocks and other

substrates and feed on periphyton (algae, bac-

teria, and fungi). Piercers (Pc) feed mainly on

vascular plants by cutting the tissue and con-

suming its fluid. Shredders (Sh) chew up plant

and wood debris, breaking down large particles

into smaller ones that can be assimilated by

other organisms, and generally facilitating the

organic matter decomposition process. Collec-

tors (GC) can screen small particles, although,

having smaller oral devices, they also feed on

material that can be re-suspended in the water

column. Filter feeders (Ft) remove particles

from the water column; some may consume

bacteria and suspended organic particles by

filtering. Finally, predators (Pr) are those that

can consume other organisms, some of which

have mouth structures like jaws.

In ecology, the idea of modularity is wide-

ly used (Olesen et al., 2007; Ruiz-Toro et al.,

2020). One application of it is to measure

the degree of compartmentalization of food

84 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

webs, which can be used to indicate the stabil-

ity relations of communities. A network with

high modularity would be considered as more

resilient to external disturbances or changes

because influences to a limited number of the

parts would less intensely propagate through-

out the network (Gauzens et al., 2015; Stouffer

& Bascompte, 2011). The idea that a modular

organization would be beneficial for the local

stability of ecological communities and their

ability to recover from small perturbations

(Grilli et al., 2016).

The question that this paper seeks to

answer is whether the altitudinal gradient and

temporal variation influence the distribution

and resistance of aquatic macroinvertebrate

communities in a river basin. Therefore, this

research aims to analyze the distribution of

aquatic macroinvertebrates according to the

altitudinal gradient in two rivers and a stream

in the same basin in Antioquia, Colombia.

This is done using ecological networks analy-

sis and modularity as a response variable, to

determine whether there are differences in the

resistance characteristics of macroinvertebrate

communities according to the typology of each

river or stream.

MATERIALS AND METHODS

Study area and site characteristics: The

study area is located in the department of

Antioquia in Colombia (Fig. 1) and includes

three different catchment areas: a micro-basin

where the La Nitrera stream is located; an inter-

mediate basin whose main drainage is the San

Juan, between the municipality of Jardin and its

mouth on the Cauca River in the place known

as Peñalisa; and a section of the Cauca River

forming a macro-basin, between the village of

Bolombolo and the municipality of Caucasia in

the North of the department.

Three sampling stations were located at

different altitudinal gradients at each study

site. At each of the sites, the physicochemi-

cal, hydraulic, and biological variables were

measured. La Nitrera stream is in the South-

west of the department of Antioquia, in the

municipality of Concordia. Its source is 2 190

elevation meters, and it is one of the main

tributaries of the reservoir located in La Nitrera

protected area that supplies water to the munic-

ipality of Concordia. This natural park is clas-

sified as low mountain rainforest with a highly

rugged topography (Morales-Quintero et al.,

2019; Ruiz-Toro et al., 2020).

The San Juan River basin is in the South-

west region of the Department of Antioquia,

with an area of approximately 1 400 km2, and

elevations between 534 and 3 920 meters eleva-

tion San Juan River, within the classification of

types of streams, can be classified as a river or

mountain stream. The Cauca River originates

in the Colombian Massif between the Western

and Central Andes Mountains and has a drain-

age area of 59 000 km2, which represents 5 %

of the national territory (Puertas et al., 2011).

The fluvial system of the Cauca River

runs 1 350 km from its source in the Colom-

bian massif (Cauca Department) to its mouth

in the “Brazo de Loba” (Bolivar Department).

The basin is polluted by mining, agro-indus-

trial waste and wastewater treatment plants

and hydroelectric activities, resulting in large

amounts of sediments with heavy metals, xeno-

biotics, and cyanides along its path (López,

2013; Torres & Pinilla, 2011).Table 1 presents

the geographical information on latitude, longi-

tude, and altitude.

Field sampling collection: The informa-

tion for site description was gathered at each

of the nine stations (Table 1). Hydraulic and

physicochemical data, as well as information

on aquatic macroinvertebrates, were sampled.

Each of these variables was repeatedly record-

ed. The fieldwork was carried out over dry

(February, July), rainy (April, September) and

transition (August, September) seasons.

Environmental and biological variables:

The parameters measured in the field were:

dissolved oxygen, electrical conductivity, pH,

turbidity, and water temperature. WTW 3330

multiparametric cells were used for this pur-

pose. The aquatic macroinvertebrates were

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

Fig. 1. Sampling site locations with the three streams: La Nitrera stream (NS1, NS2, NS3), San Juan River (SJS1, SJS2,

SJS3) and Cauca River (CS1, CS2, CS3).

TABLE 1

Geographic information

Site Longitude (North) Latitude (West) Altitude (m)

NS1 6°2.129’ 75°56.026’ 2 220

NS2 6°2.139’ 75°55.669’ 2 183

NS3 6°2.119’ 75°55.951’ 2 096

SJS1 5°35’ 14’’ 75°48’ 44’’ 2 134

SJS2 5°35’ 58’’ 75°49’ 03’’ 1 892

SJS3 5°55’ 56’’ 75°51’ 34’’ 556

CS1 5°58’ 31’’ 75°50’14’’ 555

CS2 6°2.13’ 75°56.026’ 453

CS3 6°2.14’ 75°55.669’ 50

collected with a Surber net of 0.09 m2 area

placed against the current. Large material such

as stones was extracted and the material in

the quadrant was removed by hand, ensuring

that the sediment and organisms remained in

the net. This process was carried out for one

minute in one of the margins of each stream.

Also, a semiquantitative sampling using a

triangular net was carried out in a 10 m sec-

tion for five minutes and the sampling effort

86 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

in each station was the same. The extracted

material was separated in situ in plastic trays

and later stored in plastic bottles with 70 %

alcohol duly labeled with date, place, type of

substrate, replica, responsible, to later be taken

to the Laboratory of Sanitary Hydrobiology of

the University of Antioquia.

All these measurements were taken in

duplicate. Different substrates were chosen

according to the station: leaf and rock for La

Nitrera and rock for San Juan and Cauca. The

extracted material was separated in situ in plas-

tic trays and later stored in plastic bottles with

70 % alcohol, labeled with date, place, type

of substrate, replicate, and person responsible.

These were then transported to the sanitary

hydrobiology laboratory of the University of

Antioquia. In the laboratory, all samples of

the organisms were determined to the low-

est possible taxonomic level using taxonomic

keys (Álvarez & Daza, 2005; Boltovskoy et

al., 1995; Cummins et al., 2005; Domínguez

& Fernández, 2009; Govedich & Moser, 2015;

Lasso et al., 2018; Petersen, 2002; Ramírez &

Gutiérrez-Fonseca, 2014; Roldán, 1988, 2003;

Thorp & Covich, 2009). A BST-606 stereomi-

croscope was used for this purpose.

Stomach analysis of functional feeding

groups: The protocol proposed by Muñoz et

al. (2009) in the trophic relationships chapter

for the river ecosystem was used to analyze the

stomach contents of aquatic macroinvertebrates

(Elosegui & Sabater, 2009). The digestive

analysis was performed for the most represen-

tative morphotypes (Table 2). From these three

specimens were selected. This was to identify

the food preference of each organism reported

in the literature.

For this analysis, the digestive material

closest to the mouth was extracted with the

help of dissection forceps and deposited on

a slide before a drop of glycerol (85 %) was

added. The sample was then homogenized with

the assistance of a coverslip, and the observa-

tion was made under the inverted microscope,

making an inventory of the material present

in each plate (Muñoz et al., 2009). The obser-

vations were made with an inverted Boeco

BIB100 microscope using the 40X (0.05 mm)

objective, except for the Phylloicus, where the

10X objective was used.

Biomass determination: The biomass

was determined in the sanitary hydrobiology

laboratory by relative dry weight following

the protocol of Puerta et al. (2009) his pur-

pose, semi-quantitative sampling of La Nitrera

stream was performed, integrating the biotopes

(rock and leaf litter), which were then arranged

in porcelain crucibles. Initially, the sample

was filtered for 30 minutes, removing as much

alcohol as possible. The organisms were then

separated according to their functional feed-

ing group and placed in an MF-2001 electric

muffle furnace at 105 ºC for two hours. The

fixed solids determination (inorganic matter)

was then performed, first calcining the sample

at 550 ºC for half an hour, then weighing the

sample on a Shimadzu TX323L analytical

balance with a sensitivity of 0.001 mg. This

procedure was performed by replication. In this

way, the volatile solids were determined via the

TABLE 2

Diversity indices in the three streams

Index

Nitrera San Juan Cauca

Mean

)

Standard

deviation

(σ)

Variance

coefficient

(VC)

Mean

)

Standard

deviation

(σ)

Variance

coefficient

(VC)

Mean

)

Standard

deviation

(σ)

Variance

coefficient

(VC)

Richness 9.541 2.604 27.29 6.65 2.158 32.46 4.5 3.326 73.92

Diversity 1.812 0.283 5.66 1.329 0.387 29.18 0.979 0.738 75.41

Evenness 0.820 0.080 9.816 0.723 0.122 16.87 0.773 0.211 27.34

Dominance 0.198 0.080 40.47 0.354 0.164 46.25 0.407 0.348 85.41

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

difference in weight, indicating the available

biomass that can be used as energy for higher-

order organisms (Ruiz-Toro et al., 2020).

Modularity Analysis using ecological

network: The modularity value was obtained

through the software Gephi 0.9.2. This met-

ric indicates the feeding relationships of the

aquatic macroinvertebrates present in three

rivers under altitudinal gradient distributions.

This modularity value allows the determination

of the clustering intensity degree of a network.

This measures the quality of the resulting parti-

tions and can have a scalar value between -1

and 1, with positive values indicating the pos-

sible presence of community structure. If the

value is negative, it indicates a bad grouping

(Blondel et al., 2008).

Gephi 0.9.2 software was used as an eas-

ily accessible tool for the representation and

analysis of ecological networks (Bastian et

al., 2009; Martin, 2015). This software was

applied to the data of the functional feeding

groups of the semi-quantitative sampling. The

networks were made up of two components:

the list of nodes that made up the network

(functional feeding groups and morphospecies

abundances) and a list of interactions between

the nodes (biomass).

Analysis and processing of data:

Descriptive analyses were performed using

inferential statistics relating to the physico-

chemical, hydro morphological, and biological

variables for each environment, considering

averages, maxima, and minima, to visualize

trends and extremes in each of the variables.

The modularity associated with trophic interac-

tions was used as a response variable. In this

way, through multifactorial ANOVA of two

factors, it was possible to determine whether

there were significant differences associated

with time or site of sampling. Finally, the

diversity indices were determined through the

Biodiversity Pro software.

RESULTS

Abundance and ecological structure of

aquatic macroinvertebrates: A total of 7 702

benthic macroinvertebrates were collected and

identified, comprising 150 taxa. Smicridea. (N

= 1 096) was the most abundant taxon in all

streams, followed by Thraulodes sp. (N = 971)

and Anacroneuria. (N = 576). In La Nitrera,

the most abundant taxa were Smicridea and

Phylloicus (N = 935 and 431). In the San Juan

River, Thraulodes (N = 546) Traverella (N =

325) had the highest abundances in the Cauca

River. Table 2 presents in terms of richness, La

Nitrera had the highest mean value per station

 = 9.5) followed by San Juan (x

 = 6.6) and

Cauca (x

 = 4.5).

Analysis of stomach content and FFG

assignment: Stomach contents of the most

representative morphotypes were analyzed and

grouped according to their feeding groups’

preferences. The material found was classified

according to the nutrition types according to

Cheshire et al. (2005). Classifications were

as follows: FPOM (fine particulate organic

matter, < 50 µm); CPOM (coarse particulate

organic matter, 50 µm - 1 mm); OR (organisms’

remains); PA (periphytic algae); and PT (plant

tissue, > 1 mm).

Table 3 presents comparative information

with the FFG allocation found through the

diet presented in the stomach analysis of some

representative taxa from tropical rivers in this

study, and that reported by previous authors.

12 taxa were analyzed. 5 taxa had FPOM

in their stomachs; 7 taxa had CPOM; 8 had

plant tissue; 4 had traces of organisms, and

only one had periphytic algae. This made it

possible to assign a feeding function to each

of them. The findings for nearly 92 % of the

groups are consistent with reports by authors

such as (Ramírez & Gutiérrez-Fonseca, 2014).

Table 4 presents the summary of the mul-

tifactorial ANOVA relating the abundances of

88 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

the functional groups to the hydrological sea-

son and altitude factors, as well as the interac-

tion between these factors.

These values show that statistically sig-

nificant differences occur between FFG and

Altitude. This last factor is directly related

to the sampling site. The season does not

appear to have a significant effect on the abun-

dance of the functional feeding groups in the

three streams.

Fig. 2 illustrates the spatial changes along

the different basins studied. Pie charts, in dif-

ferent shades of gray, show the composition

in percentages of the abundance of the func-

tional feeding groups in the semi-quantitative

sampling for each of the monitored stations.

The numerical value in each segment of the

pie chart corresponds to richness which, in the

case of Nitrera, is higher than for the other two

rivers. However, no differences are observed

between the sampling stations in La Nitrera.

Meanwhile, for the San Juan River, there is a

variation of the FFG in each period. Filter feed-

ers were abundant at the time, while collectors

were abundant in the middle and lower parts.

The diagrams presented in Fig. 3 shows

the configuration of the ecological network

through the relationships of the FFG with the

modularity value for each of the stations in

September 2018.

In the case of La Nitrera, the number of

nodes is higher due to the presence of a greater

number of morphotypes compared to the other

rivers. The Shredder have then greatest contri-

bution of biomass in La Nitrera (Fig. 3A, Fig.

3B, Fig. 3C), while in the San Juan River it is

from collectors (Fig. 3D, Fig. 3E, Fig. 3F). In

the Cauca River, the contribution of biomass

varies between stations. At the station (Fig.

3G), the main contribution is from scrapers,

TABLE 3

Diet and Feeding Functional Group of some taxa identified in this research and reported in the literature

Taxa Food diet FFG in the current study FFG reported in the literature

Traverella FPOM, OR CG, Pr CG

Baetodes FPOM, PT Ft, Sc Sc

Prebaetodes FPOM CG CG

Lepthoyphes CPOM, PT CG, Sh CG

Hetaerina FPOM, CPOM, OR, PA, PT Pr Pr

Smicridea PT Ft Ft

Phylloicus CPOM, PT Sh Sh

Anacroneuria CPOM, OR, PT Pr Pr

Chironomidae CPOM CG, Sh CG

Simulium FPOM, PT Ft Ft

Physa CPOM, PT Sc Sc

Gammarus CPOM, OR Pr Sh

CPOM: Coarse particulate organic matter, FPOM: Fine particulate organic matter OR: Organism remins, PA: Periphytic

algae PT: Plant tissue, CG: Collector gatherers, Pr: Predator, Ft: Filter feeder, Sc: Scrapers, Sh: Shredders.

TABLE 4

Multifactorial variance analysis for the functional feeding groups

MAIN EFFECTS Ft CG Pr Sc Sh

FP Value FP Value FP Value FP Value FP Value

A: Season 0.49 0.6186 0.83 0.4444 0.48 0.6243 1.94 0.1583 0.76 0.4733

B: Altitude 4.70 0.0005 3.39 0.0051 5.14 0.0002 5.45 0.0001 4.75 0.0005

AxB 0.74 0.7394 0.24 0.9983 0.37 0.9809 1.88 0.0561 0.98 0.5002

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

Fig. 2. Abundance and Richness distribution of the FFG following the altitudinal gradient in dry, transition and rainy seasons

in 2017 and 2018.

90 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

especially the Thiaridae family; at (Fig. 3H)

it is from filter feeders although other groups

contribute, while at CS3 (Fig. 3I) filter feeders

were found as the only biomass contributors.

The analysis of variance components

(AVC) using modularity as a dependent vari-

able showed that the altitude is the factor that

contributes most to the variance, with 65.49

%, followed by the season, with 27.57 %.

Hence, a multifactorial ANOVA was carried

out to establish whether there are statistical

differences for each principal or interaction

effect in each watercourse studied. Table 5

presents multifactorial ANOVA indicating sig-

nificant statistics differences between season

and La Nitrera modularity.

Fig. 4 shows the variation of the modular-

ity values in the different hydrological periods.

Fig. 4A and Fig. 4B corresponding to July,

Fig. 4C and Fig. 4D corresponding to August.

Finally, Fig. 4E and Fig. 4F corresponding to

September in 2017 and 2018 respectively. The

Fig. 3. Trophic relationship diagrams for September 2018 at the different stations for the three riverine scenarios using

ecological networks. A. Nitrera S1 September 2018 B. Nitrera S2 September 2018 C. Nitrera S3 September 2018 D. San

Juan S1 September 2018 E. San Juan S2 September 2018 F. San Juan S3 September 2018 G. Cauca S1 September 2018 H.

Cauca S2 September 2018 I. Cauca S3 September 2018. Each color indicates the number of modules

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

bars indicate the modularity value for each sta-

tion. The shades of grey change according to

the season. Spatial and temporal changes can

be seen. The range of modularity in each figure

varies from 0 to 1. If the value is closer to 1

then the degree of compartmentalization of the

food web is greater, by contrast negative values

indicate there are not a community structure

base in the net. Nevertheless, all stations pres-

ent a positive value. If the polygon remains

spatially stable then it will close completely,

which means that the modularity is homoge-

neous throughout the changes in elevation.

The most compartmentalized months are

July 2018 (Fig. 4B) and September 2018 (Fig.

4F). It is important to bear in mind that, by

August 2018 (Fig. 4D), the sample of organ-

isms could not be taken because of a flood

in the river at the SJS1 station. August 2018

for SJS1, CS3 stations and July, August, and

September of 2017 for CS3 station were not

possible to calculate modularity value due to

the low or no number of organisms.

DISCUSSION

To the functional feeding groups analy-

sis, the stomach determination show that fine

organic matter was predominant, followed by

organic matter in most of the digestive sys-

tems. There were some remains of organisms,

generally mouthparts and antennae of Diptera

sp. Additionally, there were plant structures

(leaf litter, parts of internodes, seeds, shells,

fibers, etc.). These results are in line with

the statements of several authors that high-

light the importance of FPOM as a dominant

TABLE 5

Multifactorial variance analysis for Modularity

MAIN EFFECTS La Nitrera San Juan Cauca

FValue P FValue P FValue P

A: Season 3.15 0.0720 4.14 0.0490 2.64 0.1124

B: Altitude 3.24 0.0678 6.18 0.0179 0.01 0.9912

AxB 0.52 0.7207 3.37 0.0542 0.08 0.9856

Fig. 4. Modularity value hydrologic periods: dry ( ), transition ( ), rainy ( ) in the three streams

in 2017 and 2018. A. July de 2017 B. July de 2018 C. August de 2017 D. August de 2018 E. September de 2017 F.

September de 2018.

92 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

food for most aquatic invertebrates (Ferreira

et al., 2015; Moore et al., 2004; Tamaris-

Turizo et al., 2018).

In the lower altitude areas, the richness

of species is significantly reduced due to the

reduction in slopes and the topology of the riv-

ers, which results in a greater amount of depo-

sition, which does not promote the availability

of substrates. In general terms, La Nitrera pres-

ents a great abundance of FFG, mainly in the

rainy season, due to the high heterogeneity of

microhabitats (Tamaris-Turizo et al., 2018). Ft

and CG are the most abundant in this stream

in 2017, and Pr in 2018. The results indicate

that, for the San Juan River, there is no sig-

nificant variation between the FFGs during the

hydrological seasons. The predominant FFGs

are CG in the rainy season and Pr in the dry

season. A high variation of FFGs based on the

gradient was expected. However, lower rich-

ness was observed in the Cauca River, which is

mainly due to aspects reported by other authors

such as the geomorphology of the river, the

short length of the main channel and the steep

slopes of the river (Tamaris-Turizo et al., 2018;

Webster & Meyer, 1997). Another point to

consider is the impact of human activities as

the gradient decreases, which have a significant

effect on the composition of aquatic organisms.

Baumgartner and Robinson (2017) mention

due to natural gradients (stream order, substra-

tum, altitude, etc.) and the dendritic structure of

river networks cumulative environmental fac-

tors affecting biotic assemblages are expected

in human dominated sites.

The changes in the distribution of aquatic

macroinvertebrates according to their FFG

showed a statistically significant variation

due to altitude, which is associated with the

sampling altitude gradient. If the gradient is

lowered, the abundance of these FFGs also

decreases. This variation occurs due to factors

such as biotopes colonization availability and

increases in water turbidity and water tempera-

ture (Dallas, 2007). The utilization of the model

of ecological networks using the modular struc-

ture showed that the stations that presented

greater FFG, that is to say, a greater diversity

of biotopes related to the number of ecological

functions, may be able to buffer the propaga-

tion of disturbances, determining the resistance

of these networks (Gauzens et al., 2015).

In the study, a variety of community

responses were observed, depending on the

type of stream and the intensity of hydrological

alteration (Serna López et al., 2020). During

the dry and rainy periods, the food webs had

a smaller number of organisms, which can be

seen in the modularity and compartmentaliza-

tion of the webs. These aspects show how

tributaries can influence longitudinal recovery

trends (Mellado-Díaz et al., 2019). If a net-

work has a high modularity degree, then this

is reflected in the compartmentalization degree

of the food webs. High modularity could indi-

cate the network is resilient to external distur-

bances or changes.

As presented in the transition period, the

potential recovery of aquatic communities

along longitudinal gradients in the basin is

maintained. The bars indicated that the transi-

tion period presented the highest modularity,

which is due to the recolonization of some

chorotypes resulting from the decrease in the

bed speed. La Nitrera stream and the San Juan

River presented higher values than the Cauca

River, which may indicate that the altitudinal

change affects the ecological network, pre-

senting a greater degree of compartmental-

ization due to the number of nodes. Finally,

this study represents a contribution to the

analysis of aquatic macroinvertebrate commu-

nities through functional feeding groups, using

the modularity of ecological networks along

an altitude gradient in rivers in Colombia.

Although further studies are required, it can be

shown that the habitat conditions and changes

in the configuration of the watercourses alter

the stability of these relationships and reduce

the species richness in low areas < 50 m.

Ethical statement: the authors declare

that they all agree with this publication and

made significant contributions; that there is

no conflict of interest of any kind; and that

we followed all pertinent ethical and legal

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

procedures and requirements. All financial

sources are fully and clearly stated in the ack-

nowledgements section. A signed document

has been filed in the journal archives.

ACKNOWLEDGMENTS

This article was written within the frame-

work of the doctoral thesis and with the support

of Minciencias, convention 727 of 2015, and in

cooperation with the research groups GeoLim-

na and Gepar of the faculty of engineering of

the university of Antioquia. Some of the results

obtained are thanks to work carried during the

internship at the university of Vechta, Ger-

many. This paper has not conflict of interests.

RESUMEN

Distribución altitudinal de los grupos funcionales

de alimentación de macroinvertebrados acuáticos

utilizando una red ecológica en arroyos andinos

Introducción: El análisis de grupos funcionales de alimen-

tación (GFA) en macroinvertebrados acuáticos es impor-

tante para comprender la estructura, función y dinámica de

los ecosistemas de procesos ecológicos. La modularidad se

refiere al grado de compartimentación de las redes alimen-

tarias y varía entre -1 y 1. Una red con un valor de modu-

laridad cercano a 1 es resistente a las alteraciones y puede

interpretarse como un factor indicativo para la estabilidad

de las comunidades.

Objetivo: En este estudio se analizó la estructura trófica

de los macroinvertebrados bentónicos, un elemento impor-

tante en la calidad ambiental, en el arroyo La Nitrera, el río

San Juan y el río Cauca.

Métodos: El estudio contó con el apoyo de técnicas de

redes ecológicas utilizando el software Gephi. En 2017

y 2018, estudiamos nueve sitios en estaciones secas,

lluviosas y de transición, monitoreando cambios en el

gradiente de altitud. En cada uno de los sitios se capturaron

y determinaron los organismos y se recogió información

fisicoquímica e hidráulica.

Resultados: El análisis de componentes de varianza per-

mitió explicar la variabilidad de los datos relacionando las

siguientes variables ambientales: GFA, diversidad, rique-

za, modularidad, estación y tiempo. La ANOVA simple

multifactorial indicó que existen cambios significativos en

los GFA en relación con la altitud, y la modularidad con

el tiempo. La asignación de los GFA se realizó mediante

análisis estomacal e información secundaria.

Conclusión: La temporada de transición tuvo la mayor

modularidad, posiblemente debido a la recolonización de

algunos biotopos provocada por la disminución de la velo-

cidad del cauce. La Nitrera y San Juan presentaron valores

superiores a los del Cauca, lo que puede indicar que el

cambio altitudinal y la velocidad de las corrientes de agua

influyen en la compartimentación de la red.

Palabras clave: modularidad; gradiente altitudinal; aná-

lisis estomacal; análisis de redes; estructura trófica de

invertebrados.

REFERENCES

Álvarez, L. F., & Daza, E. M. (2005). Metodologia para

la utilización de los macroinvertebrados acuáticos

como indicadores de la calidad del agua. http://repo-

sitory.humboldt.org.co/handle/20.500.11761/31357

Anand, M., González, A., Guichard, F., Kolasa, J., &

Parrott, L. (2010). Ecological systems as complex

systems: Challenges for an emerging science. Diver-

sity, 2(3), 395–410. https://doi.org/10.3390/d2030395

Bastian, M., Heymann, S., & Jacomy, M. (2009). Gephi:

An Open Source Software for Exploring and Mani-

pulating Networks. Third International AAAI Confe-

rence on Weblogs and Social Media, 361–362. https://

doi.org/10.1136/qshc.2004.010033

Baumgartner, S. D., & Robinson, C. T. (2017). Chan-

ges in macroinvertebrate trophic structure along a

land-use gradient within a lowland stream network.

Aquatic Sciences, 79(2). https://doi.org/10.1007/

s00027-016-0506-z

Blondel, V. D., Guillaume, J. L., Lambiotte, R., &

Lefebvre, E. (2008). Fast unfolding of commu-

nities in large networks. Journal of Statistical

Mechanics: Theory and Experiment. https://doi.

org/10.1088/1742-5468/2008/10/P10008

Boltovskoy, D., Izaguirre, I., & Correa, N. (1995). Feeding

selectivity of Corbicula fluminea (Bivalvia) on natu-

ral phytoplankton. Hydrobiologia, 312(3), 171–182.

https://doi.org/10.1007/BF00015510

Castillo-Figueroa, D., Garzón-Salamanca, L. L., & Alba-

rracín-Caro, J. F. (2018). Aquatic macroinvertebra-

tes as water quality bioindicators in colombia: A

systematic review. Neotropical Biology and Con-

servation, 13(3), 235–248. https://doi.org/10.4013/

nbc.2018.133.06

Chará-Serna, A. M., Chará, J. D., Zúñiga, M. del C., Pedra-

za, G. X., & Giraldo, L. P. (2010). Trophic classifi-

cation of aquatic insects in eight sheltered streams of

the Colombian coffee ecoregion. Universitas Scien-

tiarum, 15(1), 27–36. https://doi.org/10.11144/jave-

riana.SC15-1.tcoa

Cheshire, K., Boyero, L., & Pearson, R. G. (2005). Food

webs in tropical Australian streams: Shredders are not

94 Revista de Biología Tropical, ISSN: 2215-2075 Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

scarce. Freshwater Biology, 50(5), 748–769. https://

doi.org/10.1111/j.1365-2427.2005.01355.x

Cummins, K., Merritt, R., & Andrade, P. (2005). The

use of invertebrate functional groups to charac-

terize ecosystem attributes in selected streams

and rivers in south Brazil. Studies on Neotropical

Fauna and Environment, 40(1), 69–89. https://doi.

org/10.1080/01650520400025720

Dallas, H. F. (2007). The influence of biotope availa-

bility on macroinvertebrate assemblages in South

African rivers: Implications for aquatic bioas-

sessment. Freshwater Biology, 52(2). https://doi.

org/10.1111/j.1365-2427.2006.01684.x

Domínguez, E., & Fernández, H. (2009). Macroinvertebra-

dos bentónicos sudamericanos. Fundación Miguel

Lillo.

Elosegui, E., & Sabater, S. (2009). Conceptos y técnicas en

ecología fluvial. Fundación BBVA.

Ferreira, W. R., Ligeiro, R., Macedo, D. R., Hughes, R.

M., Kaufmann, P. R., Oliveira, L. G., & Callisto, M.

(2015). Is the diet of a typical shredder related to the

physical habitat of headwater streams in the Brazi-

lian Cerrado? Annales de Limnologie-International

Journal of Limnology, 51(2), 115–127 https://doi.

org/10.1051/limn/2015004

Gauzens, B., Thébault, E., Lacroix, G., & Legendre, S.

(2015). Trophic groups and modules: Two levels of

group detection in food webs. Journal of the Royal

Society Interface, 12(106), 20141176. https://doi.

org/10.1098/rsif.2014.1176

Govedich, F. R., & Moser, W. E. (2015). Clitellata: Hiru-

dinida and Acanthobdellida. Thorp and Covich’s

Freshwater Invertebrates, 565–588. https://doi.

org/10.1016/B978-0-12-385026-3.00023-1

Grilli, J., Rogers, T., & Allesina, S. (2016). Modulari-

ty and stability in ecological communities. Nature

Communications, 7(1), 1–10. https://doi.org/10.1038/

ncomms12031

Gutiérrez, J. P. (2018). Riverbank filtration in highly turbid

rivers. TU Delft University. https://doi.org/10.4233/

uuid:a6fe2afd-1f34-4b03-9687-1a5b627b64c2

Lasso, C. A., Morales-Betancourt, M. A., Vera-Ardila, M.

L., & Linares, E. L. (2018). Moluscos dulceacuí-

colas de Colombia. Instituto Humboldt. https://doi.

org/10.21068/a2018n03

Latrubesse, E. M., Stevaux, J. C., & Sinha, R. (2005).

Tropical rivers. Geomorphology, 70(3-4), 187–206.

https://doi.org/10.1016/j.geomorph.2005.02.005

López, P. A. (2013). Propuesta de prevención y manejo

de la contaminación por mercurio en la región de

La Mojana. Journal of Chemical Information and

Modeling, 53, 1689–1699. https://doi.org/10.1017/

CBO9781107415324.004

Martin, G. (2015). Gephi -Introduction to network analysis

and visualization. Digital Humanities. http://www.

martingrandjean.ch/gephi-introduction/

Mellado-Díaz, A., Sánchez-González, J. R., Guareschi, S.,

Magdaleno, F., & Toro Velasco, M. (2019). Explo-

ring longitudinal trends and recovery gradients in

macroinvertebrate communities and biomonitoring

tools along regulated rivers. Science of the Total

Environment, 695, 133774. https://doi.org/10.1016/j.

scitotenv.2019.133774

Morales-Quintero, D. E., Velez-Macías, F. de J., & Guerre-

ro-Hoyos, B. (2019). Caracterización hidrogeológica

y geomorfológica de la cuenca y embalse del parque

natural la Nitrera. Revista de La Facultad de Cien-

cias, 8(1), 73–99. https://doi.org/10.15446/rev.fac.

cienc.v8n1.71399

Muñoz, I., Romaní, A. M., Rodrigues-Capítulo, A., Gonzá-

les Esteban, J., & Garcia-Berthou, E. (2009). Relacio-

nes tróficas en el ecosistema fluvial. En A Elosegui

(Ed.), Conceptos y Técnicas En Ecología Fluvial (pp.

1–22). Fundación BBVA.

National Academy of Sciences. (1988). In E. O. Wilson

(Ed.), Biodiversity. The National Academies Press.

https://doi.org/https://doi.org/10.17226/989

Olesen, J. M., Bascompte, J., Dupont, Y. L., & Jordano,

P. (2007). The modularity of pollination networks.

Proceedings of the National Academy of Sciences of

the United States of America, 104(50), 19891–19896.

https://doi.org/10.1073/pnas.0706375104

Petersen, H. (2002). General aspects of collembolan eco-

logy at the turn of the millennium. Pedobiologia,

46(3–4), 246–260.

Pilosof, S., Porter, M. A., Pascual, M., & Kéfi, S. (2017).

The multilayer nature of ecological networks. Natu-

re Ecology and Evolution, 1(4), 1–9. https://doi.

org/10.1038/s41559-017-0101

Puerta, B., Molina, F., & Rodriguez, D. (2009). Manual

de laboratorio de procesos biológicos. Reimpresos-

Universidad de Antioquia.

Ramírez, A., & Gutiérrez-Fonseca, P. E. (2014). Functional

feeding groups of aquatic insect families in Latin

America: A critical analysis and review of existing

literature. Revista de Biología Tropical, 62, 155–167.

https://doi.org/10.15517/rbt.v62i0.15785

Roldán, G. (1988). Guía para el estudio de los macroinver-

tebrados acuáticos del departamento de Antioquia.

Universidad de Antioquia.

Roldán, G. (2003). Bioindicación de la calidad del agua en

Colombia (Uso del método BMWP/Col). Universidad

de Antioquia.

Revista de Biología Tropical, ISSN: 2215-2075, Vol. 70: 82-95, January-December 2022 (Published Feb. 10, 2022)

Roldán-Pérez, G. (2016). Los macroinvertebrados como

bioindicadores de la calidad del agua: cuatro décadas

de desarrollo en Colombia y Latinoamerica. Revista

de La Academia Colombiana de Ciencias Exac-

tas, Físicas y Naturales, 40(155), 254. https://doi.

org/10.18257/raccefyn.335

Rosenberg, D., & Resh, V. (1993). Freshwater biomoni-

toring and benthic macroinvertebrates. Chapman

& Hall.

Ruiz-Toro, J., Aguirre-Ramírez, N., Serna-López, J.,

Hernández-Atilano, E., & Vélez-Macías, F. (2020).

Energía calórica, biomasa y estructura de los

macroinvertebrados acuáticos en la reserva La Nitre-

ra, Concordia, Antioquia, Colombia. Acta Biológica

Colombiana, 25(1), 8. http://dx.doi.org/10.15446/

abc.v25n1.76435

Schmera, D., Eros, T., & Heino, J. (2013). Habitat filtering

determines spatial variation of macroinvertebrate

community traits in northern headwater streams.

Community Ecology, 14(1), 77–88. https://doi.

org/10.1556/ComEc.14.2013.1.9

Schmera, D., Heino, J., Podani, J., Erős, T., & Dolédec, S.

(2017). Functional diversity: a review of methodo-

logy and current knowledge in freshwater macroin-

vertebrate research. Hydrobiologia, 787(1), 27–44.

https://doi.org/10.1007/s10750-016-2974-5

Serna López, J. P., Fernández Mc Cann, D. S., Vélez

Macías, F. de J., & Aguirre Ramírez, N. J. (2020).

An image processing method for recognition of

four aquatic macroinvertebrates genera in freshwater

environments in the Andean region of Colombia.

Environmental Monitoring and Assessment, 192(10),

1–11. https://doi.org/10.1007/s10661-020-08545-2

Stouffer, D. B., & Bascompte, J. (2011). Compartmentali-

zation increases food-web persistence. Proceedings

of the National Academy of Sciences of the United

States of America, 108(9), 3648–3652. https://doi.

org/10.1073/pnas.1014353108

Tamaris-Turizo, C. E., Pinilla, G. A., & Muñoz, I. (2018).

Trophic network of aquatic macroinvertebrates along

an altitudinal gradient in a Neotropical mountain

river. Revista Brasileira de Entomologia, 62(3),

180–187. https://doi.org/10.1016/j.rbe.2018.07.003

Thorp, J. H., & Covich, A. P. (2009). Ecology and Classi-

fication of North American Freshwater Invertebra-

tes (3rd Ed.). https://doi.org/https://doi.org/10.1016/

C2009-0-02669-5

Tomanova, S., Goitia, E., & Helešic, J. (2006). Trophic

levels and functional feeding groups of macroinverte-

brates in neotropical streams. Hydrobiologia, 556(1),

251–264. https://doi.org/10.1007/s10750-005-1255-5

Torres, J. F., & Pinilla, G. A. (2011). Revisión de las carac-

terísticas limnológicas de los sistemas acuáticos de

la región de la Mojana. Facultad de Ingeniería, 70.

Violle, C., Navas, M. L., Vile, D., Kazakou, E., Fortunel,

C., Hummel, I., & Garnier, E. (2007). Let the concept

of trait be functional! Oikos, 116(5), 882–892. https://

doi.org/10.1111/j.0030-1299.2007.15559.x

Webster, J. R., & Meyer, J. L. (1997). Organic matter bud-

gets for streams: A synthesis. Journal of the North

American Benthological Society, 16(1), 141–161

https://doi.org/10.2307/1468247