ArticlePDF Available

A new predictive model (MINASPACS) for spatially extensive biological assessments in southeastern Brazilian streams

March 2025
Water Biology and Security

DOI:10.1016/j.watbs.2025.100386

License
CC BY 4.0

Authors:

Maria João Feio

University of Coimbra & Marine and Envrionmental Sciences Centre

Robert MASON Hughes

Show all 5 authorsHide

Freshwater ecosystems are threatened by flow regulation, sedimentation, habitat degradation, non-native species, and water pollution. These disturbances have led to global losses of biodiversity and habitats. Therefore, it is essential to evaluate the ecological condition of freshwater ecosystems to promote effective management practices. Quantitative predictive models based on multivariate analyses of taxa richness are recognized ecological tools that can facilitate the monitoring and managing of freshwater ecosystems worldwide. However, few studies have used this approach to assess tropical rivers and streams. By evaluating predictive models, we can assess their usefulness for determining water-body taxonomic richness. We built a RIVPACS-type model based on macro-invertebrate assemblage (MINASPACS), for spatially extensive taxa richness assessments of Minas Gerais state streams, southeast Brazil. As a second objective, we assessed the sensitivity of the MINASPACS to human-induced disturbances affecting Minas Gerais streams through the relative risk (RR) approach. The MINASPACS model was trained with biological and environmental data from 78 reference sites and showed good accuracy (R 2 > 0.6, SD O/E ¼ 0.16). We found that percent of urban infrastructure, percent of catchment anthropogenic land use, Turbidity, Total Nitrogen, and Total Phosphorus represented significant risks to the taxa richness of Minas Gerais streams. Because of its accuracy, sensitivity, and use of map-level predictor variables, our model provides a clear, simple, and defensible measure of stream macroinvertebrate taxa richness across diverse biomes.

Content uploaded by Maria João Feio

Content may be subject to copyright.

Research Article

A new predictive model (MINASPACS) for spatially extensive biological

assessments in southeastern Brazilian streams

Pedro Fialho Cordeiro

, Maria Jo~

ao Feio

, Marcos Callisto

, Robert M. Hughes

, Diego

Rodrigues Macedo

Programa de P

os-Graduaç~

ao Em An

alise e Modelagem de Sistemas Ambientais, Instituto de Geoci^

encias, Universidade Federal de Minas Gerais, Av. Ant^

onio Carlos 6627,

CEP 31270-901, Belo Horizonte, MG, Brazil

University of Coimbra, Faculty of Sciences and Technology, Department of Life Sciences, Marine and Environmental Sciences Centre, Associate Laboratory ARNET,

Coimbra, Portugal

Laborat

orio de Ecologia de Bentos, Departamento de Gen

etica, Ecologia e Evoluç~

ao, Instituto de Ci^

encias Biol

ogicas, Universidade Federal de Minas Gerais, Av. Ant^

onio

Carlos 6627, CP 486, CEP 31270-901, Belo Horizonte, MG, Brazil

Amnis Opes Institute, 2895 SE Glenn, Corvallis, OR, 97333, USA

Department of Fisheries, Wildlife, &Conservation Sciences, 104 Nash Hall, Oregon State University, 97331, USA

Departamento de Geograﬁa, Instituto de Geoci^

encias, Universidade Federal de Minas Gerais, Av. Ant^

onio Carlos 6627, CEP 31270-901, Belo Horizonte, MG, Brazil

ARTICLE INFO

Keywords:

RIVPACS

Relative risk approach

Benthic macroinvertebrates

Streams

Freshwaters

Taxa richness

ABSTRACT

Freshwater ecosystems are threatened by ﬂow regulation, sedimentation, habitat degradation, non-native species,

and water pollution. These disturbances have led to global losses of biodiversity and habitats. Therefore, it is

essential to evaluate the ecological condition of freshwater ecosystems to promote effective management prac-

tices. Quantitative predictive models based on multivariate analyses of taxa richness are recognized ecological

tools that can facilitate the monitoring and managing of freshwater ecosystems worldwide. However, few studies

have used this approach to assess tropical rivers and streams. By evaluating predictive models, we can assess their

usefulness for determining water-body taxonomic richness. We built a RIVPACS-type model based on macro-

invertebrate assemblage (MINASPACS), for spatially extensive taxa richness assessments of Minas Gerais state

streams, southeast Brazil. As a second objective, we assessed the sensitivity of the MINASPACS to human-induced

disturbances affecting Minas Gerais streams through the relative risk (RR) approach. The MINASPACS model was

trained with biological and environmental data from 78 reference sites and showed good accuracy (R

>0.6, SD

O/E ¼0.16). We found that percent of urban infrastructure, percent of catchment anthropogenic land use,

Turbidity, Total Nitrogen, and Total Phosphorus represented signiﬁcant risks to the taxa richness of Minas Gerais

streams. Because of its accuracy, sensitivity, and use of map-level predictor variables, our model provides a clear,

simple, and defensible measure of stream macroinvertebrate taxa richness across diverse biomes.

1. Introduction

Freshwaters are one of the most threatened ecosystems worldwide

(Lake, 2000;Reid et al., 2019). Flow regulation and longitudinal barriers

(Dudgeon, 2010), sedimentation and habitat degradation (Sano et al.,

2019), non-native species, and water pollution have led to biodiversity

and habitat losses (Feio et al., 2014). Given this scenario, assessing the

ecological condition of freshwater ecosystems is critical for addressing

cost-efﬁcient management practices (Silva et al., 2017;Paulsen et al.,

2020). Several methodological approaches based on the use of aquatic

organisms as bioindicators have been used in the biological assessment of

freshwater ecosystems in North America, Europe and Australia. The

major approaches are multimetric indices (e.g., Buss et al., 2015;Karr,

1999;Ruaro et al., 2020), relative risk (RR) and relative extent (RE)

approaches (e.g., Van Sickle and Paulsen, 2008), and predictive taxa

richness models (e.g., Clarke et al., 2003;Feio et al., 2014;Reynoldson

et al., 1997;Wright, 1995). However, these approaches are not

completely explored in other continents, such as Asia (e.g., Blakely et al.,

* Corresponding author.

E-mail addresses: pedroﬁalhoc@gmail.com (P.F. Cordeiro), mjf@ci.uc.pt (M.J. Feio), mcallisto13@gmail.com (M. Callisto), hughes.bob@amnisopes.com

(R.M. Hughes), diegorm@ufmg.br (D.R. Macedo).

Peer review under the responsibility of Editorial Ofﬁce of Water Biology and Security.

Contents lists available at ScienceDirect

Water Biology and Security

journal homepage: www.keaipublishing.com/en/journals/water-biology-and-security

https://doi.org/10.1016/j.watbs.2025.100386

Received 17 November 2024; Received in revised form 25 February 2025; Accepted 15 March 2025

(http://creativecommons.org/licenses/by/4.0/).

Water Biology and Security xxx (xxxx) xxx

Please cite this article as: Cordeiro, P.F. et al., A new predictive model (MINASPACS) for spatially extensive biological assessments in southeastern

Brazilian streams, Water Biology and Security, https://doi.org/10.1016/j.watbs.2025.100386

2014;Chen et al., 2019) and South America (e.g., Aida Campos et al.,

2021;Martins et al., 2020;Oliveira et al., 2011;Silva et al., 2018).

Globally, benthic macroinvertebrates are the most-used biological as-

semblages for assessing lotic ecological condition (Feio et al., 2022).

Both the USA and Europe have legal frameworks that support the use

of biotic indicators to assess ecological conditions at local and regional

spatial extents (Aguiar et al., 2011;Barbour et al., 1999;Poikane et al.,

2016;Feio et al., 2021). In contrast, neotropical countries lack speciﬁc

legislation or guidelines for biological assessment, reﬂected in the rela-

tively few studies addressing the development and application of meth-

odological approaches based on aquatic organisms compared to the USA

and Europe (Ruaro and Gubiani, 2013;Feio et al., 2021). In Europe,

benthic macroinvertebrate assemblages are used in 26 national assess-

ment systems for rivers, 13 assessment systems for very large rivers, and

21 assessment systems for lakes (Vitecek et al., 2021). For streams and

rivers, ecological status is often assessed using sensitivity, trait, and

abundance metrics integrated into multimetric and biotic indices (Birk

et al., 2012). Similarly, the U.S. Environmental Protection Agency uses

macroinvertebrate multimetric indices for assessing the condition of all

lakes (Mitchell et al., 2025), rivers and streams (Herlihy et al., 2020)in

the conterminous USA.

Predictive taxonomic richness models based on multivariate analyses

are also used to facilitate the monitoring and management of freshwater

ecosystems (Feio and Poquet, 2011;Wright, 1995). The River Inverte-

brate Prediction and Classiﬁcation System (RIVPACS) (Wright, 1995)

was the ﬁrst model of this kind and was developed for the United

Kingdom, based partly on the Reference Condition Approach (Hughes

et al., 1986;Reynoldson et al., 1997;Stoddard et al., 2006).

RIVPACS-type models make site-speciﬁc predictions of the benthic

macroinvertebrate taxa richness expected without anthropogenic dis-

turbances. Those predictions are based on empirical relationships be-

tween individual taxon capture probabilities and natural environmental

features (e.g., latitude, channel slope, alkalinity, elevation) derived from

data collected from a reference site network (Hargett et al., 2007).

Since its ﬁrst version, RIVPACS has evolved into a nationwide bio-

assessment tool in the UK (Wright, 1995). It also was adapted to assess

stream macroinvertebrate richness in Australia (AUSRIVAS, Smith et al.,

1999), Canada (Reynoldson et al., 1997), Sweden (Johnson and Sandin,

2001), the USA (Van Sickle et al., 2005), the Czech Republic (Koke

s et al.,

2006), and Portugal (Feio et al., 2009). However, it has not been

implemented as a nationwide bioassessment tool except in the UK and

USA (USEPA, 2016). One of the main disadvantages of multivariate

predictive models compared to multimetric indices is their complex

modeling, which discourages potential users (Reynoldson et al., 1997).

Nonetheless, predictive models can provide more intuitive outputs than

simple multimetric and biotic indices because of their standardization to

site-speciﬁc conditions (Hargett et al., 2007). Each method has its ad-

vantages and disadvantages, and combining both approaches can

enhance the assessment of river ecological conditions across broad

geographic extents (Paulsen et al., 2020;USEPA, 2016). In the USA, for

example, the RIVPACS-type approach combined with multimetric indices

and a probability survey concluded that 58% of the stream length in the

conterminous USA had lost >20% of its common macroinvertebrate taxa

and only 28% of stream length was in good condition (USEPA, 2016).

Similarly, predictive multimetric indices have been implemented for ﬁsh

assemblages in Europe and the USA (Pont et al., 2006,2009) and mac-

roinvertebrate assemblages in Bolivia (Moya et al., 2011) and China

(Chen et al., 2019;Gao et al., 2020).

Several aspects of the RIVPACS approach were incorporated into the

prescribed methods of the European Water Framework Directive (WFD;

2000/60/EC) for assessing the ecological quality and ecological status of

European surface waters (Clarke et al., 2003). However, few studies have

used this approach for tropical rivers and streams, except in a single

catchment (e.g., Moreno et al., 2009) or 3 reservoirs (e.g., Molozzi et al.,

2012). In Brazil, Minas Gerais state (586,528 km

) recently implemented

quality classes for water body ecological condition as a component of

biomonitoring programs (Normative Deliberation COPAM/CERH-MG nº

008/2022; COPAM, 2022). Biotic indices (e.g., Junqueira et al., 2018)

and multimetric indices (e.g., Martins et al., 2021;Silva et al., 2018,

2017) were the most commonly used bioassessment tools, whereas pre-

dictive taxa richness and multimetric modeling remain largely

underexplored.

Predictive modeling could be extremely useful in large tropical

countries such as Brazil and contribute to fulﬁlling the recent ofﬁcial

requirements (Buss et al., 2015). Therefore, we built a RIVIPACS-type

model (MINASPACS) for making spatially extensive taxonomic richness

assessments of Minas Gerais streams. We also assessed the MINASPACS's

sensitivity to anthropogenic pressures through the relative risk approach

(Herlihy et al., 2020;Silva et al., 2018).

2. Materials and methods

2.1. Study area and environmental characterization

We compiled data from research projects conducted between 2003

and 2019 from a total of 381 sites sampled by the Laboratory of Ecology

of Benthos-Universidade Federal de Minas Gerais and Serviço Nacional

de Aprendizagem Industrial - Centro de Inovaç~

ao e Tecnologia. Sites

included 348 in Minas Gerais (Agra et al., 2019;Callisto et al., 2021;

Dala-Corte et al., 2020;Feio et al., 2013;Garuana et al., 2020;Linares

et al., 2021;Macedo et al., 2022;Martins et al., 2020;Silva et al., 2017;

Ferreira et al., 2017). We also compiled data from 20 stream sites in Goi

and 13 in S~

ao Paulo states (Callisto et al., 2019), covering a total area of

40,106 km

in 8 hydrological units (Fig. 1).

All sites were in the Cerrado (neotropical savanna) biome, which is

characterized by a well-deﬁned dry season from May to September

(Hunke et al., 2015), with an average precipitation from 800 to 2000

mm, and average annual temperature between 18 and 28 C(Silva et al.,

2017). The potential natural vegetation consists of dispersed trees and

shrubs, grass (Urbanetz et al., 2013), rocky ﬁeld vegetation (campos

rupestres) in the Espinhaço Range (Campos et al., 2017), and heteroge-

neous riparian forests. The major land uses are mechanized agricultural

cash crops, charcoal production, grazing, and urbanization (Macedo

et al., 2014). The topography encompasses extensive areas of ﬂat to

gently rolling relief (Grecchi et al., 2014) in the western region, and the

mountainous Espinhaço Range in the east (Campos et al., 2017). Each site

was characterized according to its lithological group (Supplementary

Information 1, Table S1), climate (50-year climatic reference, from

Worldclim Project - https://worldclim.org/), and river basin character-

istics (Table 1). Each river basin's land use proportions (six classes) were

estimated from a Geographic Information System (GIS). Those land use

data were obtained from Collection 5 of the MapBiomas online platform,

with a spatial resolution of 30 m (Souza et al., 2020)(Table 1). To

evaluate the representativeness of our reference dataset relative to the

full range of environmental conditions within the study area, we con-

ducted a comparative analysis of geospatial variables used in the

modeling. This comparison was performed using boxplots, considering

data from all 87 reference catchments and 87 randomly selected Cerrado

catchments.

2.2. Biological samples and water quality data

Each sample consisted of a composite of 3–20 Surber (30 30 cm,

500 mm mesh) or D-net (30 cm aperture, 500 mm mesh, and 0.09 m

)

samples in the most representative habitats, then aggregated into one

composite sample per site. Biological data were collected during the dry

season (May to September). Only the record exhibiting the highest

taxonomic richness was retained for sites sampled on multiple occasions.

In all 11 projects, family-level taxonomy was used, which is the most

common identiﬁcation level in Asia, Africa, and South America countries

(Eriksen et al., 2021). All individuals collected were identiﬁed with the

aid of taxonomic keys (Hamada et al., 2014;Merritt and Cummins, 1996;

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

Mugnai et al., 2009;P

erez, 1988;P

es et al., 2005;Wiggins, 1996). We

also compiled water quality data for each site. Water samples were

collected during the dry season, iced, and analyzed in the laboratory for

Total Phosphorus (mg/L), Total Nitrogen (mg/L), and Turbidity (NTU)

following APHA (2005) methods.

Because taxonomic richness typically increases with the number of

samples or individuals counted (Cao et al., 2002;Li et al., 2001,2014;

Silva et al., 2016), we performed a Spearman's rank correlation to eval-

uate the relationship between taxa richness and sampling effort. The

correlation between taxa richness and sampling effort was weak (r¼

0.04; p¼0.748), indicating that richness was weakly related to sam-

pling effort and therefore minimally inﬂuenced our results.

2.3. Reference site selection

The RIVPACS approach compares test site assemblages to those of

reference sites. Screening sites to establish reference data is necessary to

avoid the confounding effects of alterations to macroinvertebrate as-

semblages caused by anthropogenic disturbance instead of differences

resulting from natural abiotic characteristics, such as geology or climate

(Stoddard et al., 2006;Whittier et al., 2007). Therefore, we used a

ﬁltering procedure (e.g., Herlihy et al., 2008) to identify near-natural

sites for predictive model development based on land use in the catch-

ment and water-quality (Table 2). If a site failed any one of the ﬁlters, it

was not considered as a reference site. Those criteria were similar to

those established in other Cerrado studies to ensure that our

reference-site selection was comparable to theirs (e.g., Agra et al., 2019;

Garuana et al., 2020;Silva et al., 2017).

2.4. Predictive model construction (MINASPACS) and validation

The model training used biological and environmental data from

reference sites. Sites with <200 individuals were excluded from further

analyses, as were rare taxa with <5% occurrence. We a priori transformed

macroinvertebrate abundances by a fourth root transformation to reduce

the importance of different sampling efforts and the 17-year data span.

Ten natural predictive variables (following Mendes et al., 2014) were

selected as candidate predictor variables and were transformed to ensure

normality and homoscedasticity: latitude (log

xþ1

); longitude (log

xþ1

);

annual mean temperature (C); annual temperature range (C); annual

Fig. 1. Study area showing the hydrological units (sensu Seaber et al., 1987) (gray polygons) and sampled sites (black dots and green triangles). (1) Volta Grande, (2)

ao Sim~

ao, (3) Nova Ponte, (4)Tr

es Marias, (5) Cajuru, (6) das Velhas, (7) Pandeiros, and (8) Peti hydrologic units.

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

mean precipitation (Sqrt) (mm); altitude (log

xþ1

) (m); mean catchment

slope (log

xþ1

) ( %); distance from site to source (m); and lithological

synthesis. Following Ferreira et al. (2017), we grouped lithotypes to limit

the number of classes based on rock response similarities to erosion,

weathering, and leaching. These variables were previously used in

macroinvertebrate predictive models because they are not easily inﬂu-

enced by anthropogenic activities and are known to reﬂect the natural

distribution of river macroinvertebrate assemblages (e.g., Feio et al.,

2007;Hargett et al., 2007;Pardo et al., 2014).

To build the MINASPACS model, we used the AQUAWEB online

software (http://aquaweb.uc.pt/). This tool follows the approach

described in Van Sickle et al. (2006), which was previously used and

validated with large macroinvertebrate assemblage datasets (e.g., Aguiar

et al., 2011;Mendes et al., 2014). Building a RIVPACS-type model con-

tains several steps (Feio and Poquet, 2011). Brieﬂy, the reference dataset

was deﬁned by a priori reference criteria representing the natural envi-

ronmental variability present in the study area. Next, we classiﬁed the

reference sites according to their macroinvertebrate assemblage

composition via clustering (Unweighted Pair Group Method with arith-

metic mean, UPGMA) based on Bray–Curtis similarity and supported by

Non-metric Multidimensional Scaling (nMDS) (Mendes et al., 2014).

Groups required at least 10 reference sites to ensure that good predictions

and comparisons between Observed and Expected (under reference

conditions) generated reliable predictions (Wright, 1995).

We used Discriminant Function Analysis (DFA) to determine which of

the 10 candidate environmental variables best discriminated the mac-

roinvertebrate groups and ranked them using F-tests and Wilks' lambda

tests (Mendes et al., 2014) to maximize differences among reference

macroinvertebrate groups. Then, each taxon occurrence probability at a

site was calculated. The frequency of occurrence for each taxon in a

reference macroinvertebrate group was averaged and weighted based on

the site's probability of being assigned to that group through discriminant

analysis. From this, the number of Observed taxa (O) at a site was divided

by the sum of occurrence probabilities >50% of Expected taxa (E) to

obtain O/E

ratios. O/E

ratios measure deviation from reference

conditions, in which O/E

should be approximately 1. In other words, it

is the ratio between the Observed taxa in a site sample versus the Ex-

pected taxa predicted at that site with a probability of occurrence 0.50.

Model performance was assessed from O/E

mean values (MN) and

standard deviation (SD) for calibration sites. The MNO/E (mean value of

O/E

) measures model bias, and if its value is equal to one, the pre-

dictive model is unbiased. The SDO/E indicates model precision; a model

being more precise the lower its SDO/E (Mendes et al., 2014;Van Sickle

et al., 2005). We also targeted the model with a high F-statistic and low

Wilks' λvalue for model selection. We chose the best model by its O/E

regression (R

0.5; p<0.05), intersection close to the origin (a range of

1.5 to 1.5 is acceptable), SDO/E <0.2, and slope near 1 (acceptable

range of 0.85–1.15) (Linke et al., 2005;Mendes et al., 2014;Van Sickle

et al., 2005).

We classiﬁed site scores into 5 quality classes (Table 3). Well-

established multipressure-impact relationships for macroinvertebrate

assemblages are not available because of signiﬁcant knowledge gaps in

Brazil (Buss et al., 2015;Birk et al., 2012). Therefore, we set the

boundary between high and good classes at the 25th percentile of the

calibration site O/E

ratios, and the boundaries below were arbitrarily

divided into 4 equal classes, following the approaches used in European

methods, according to the Water Framework Directive for rivers (Mendes

et al., 2014). The use of 5 classes was also previously used in other

models developed in Brazil (e.g., Jovem-Azev^

edo et al., 2020;Moreno

et al., 2009), Europe (e.g., Feio et al., 2009;Feio et al., 2014;Mendes

et al., 2014), and the USA (e.g., Hargett et al., 2007;Paulsen et al., 2020).

Finally, we used SIMPER statistical procedures (similarity/distance per-

centages, fourth root transformation, Bray-Curtis coefﬁcient, Primer 6) to

determine the most representative families (up to 90% of cumulative

percentage) of each macroinvertebrate group created by the MINASPACS

model.

2.5. Sensitivity to disturbances—Assessing relative risk (RR)

We used seven variables to evaluate the sensitivity of MINASPACS to

anthropogenic disturbances. Total Phosphorus, and Total Nitrogen are

proxies for eutrophication. Land use (% urban,% pasture,% agriculture,%

total anthropogenic) and Turbidity represent multiple pressures (e.g.,

pollution and hydromorphological alterations) (Poikane et al., 2020).

We addressed all possible situations of having good or poor macro-

invertebrate O/E

values given good/low or poor/high disturbance

Table 1

Candidate natural predictor variables, pressure variables, and their sources.

Variable Source

NATURAL PREDICTIVE VARIABLES

Latitude (decimal degrees) Obtained through GIS

Longitude (decimal degrees) Obtained through GIS

Annual Mean Temperature

(C)

Obtained through GIS; Worldclim Project (htt

ps://worldclim.org/)

Annual Mean Precipitation

(mm)

Obtained through GIS; Worldclim Project (htt

ps://worldclim.org/)

Annual Temperature Range

(C)

Obtained through GIS; Worldclim Project (htt

ps://worldclim.org/)

Altitude (m) Obtained through GIS; Shuttle Radar Topographic

Mission–SRTM

Mean catchment slope (%) Obtained through GIS; Shuttle Radar Topographic

Mission–SRTM

Distance to site's source (m)

Obtained through GIS

Lithological synthesis (1–8)

Ferreira et al., (2017)

PRESSURE VARIABLES

Forest (%) Obtained through GIS; Souza et al., (2020)

Savanna (%) Obtained through GIS; Souza et al. (2020)

Pasture (%) Obtained through GIS; Souza et al. (2020)

Agriculture (%) Obtained through GIS; Souza et al. (2020)

Urban infrastructure (%) Obtained through GIS; Souza et al. (2020)

Total anthropogenic land

use (%)

Obtained through GIS; Souza et al. (2020)

Total Phosphorus (mg/L) Compiled data

Total Nitrogen (mg/L) Compiled data

Turbidity (NTU) Compiled data

Represents the largest distance of a site to its spring.

(1) Siliceous rocks; (2) Pelitic rocks; (3) Metamorphic rocks; (4) Carbonate

rocks; (5) Volcanic rocks; (6) Alkaline rocks; (7) Laterized sediments; and (8)

Unconsolidated sediments.

Table 2

Reference site ﬁltering criteria.

Filter Reference value/criteria Source

Catchment

land use

<25% total anthropogenic area;

absence of urban infrastructure.

MapBiomas (2021). (htt

ps://mapbiomas.org/)

Water quality TP <0.1 mg/L, TN <0.2 mg/L,

Turbidity <100 NTU (Silva et al.,

2017).

CONAMA Resolution nº357/

2005, Class II

, lotic

environments

Water intended for human consumption after simpliﬁed treatment and for

protection of aquatic communities (CONAMA, 2005).

Table 3

Site quality classes from the MINASPACS predictive model, with corresponding

taxa loss.

Class O/E

score Taxa loss

Upper Lower

High 1.268 0.936 <6% taxa loss

Good 0.935 0.702 7–30% taxa loss

Moderate 0.701 0.468 31–53% taxa loss

Poor 0.467 0.234 54–76% taxa loss

Very poor 0.233 0 >77% taxa loss

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

conditions (Table 4). The RR is a conditional probability representing the

likelihood that low/poor O/E

values are associated with high anthro-

pogenic disturbance scores and is calculated as follows (Equation (1)):

RR ¼Pr ðO=EjDhÞ

Pr ðO=EjDlÞ(1)

The numerator is the probability of ﬁnding poor biological conditions

(O/E

score >50% taxa loss) given high anthropogenic disturbance

scores (Dh), and the denominator is the probability of ﬁnding poor bio-

logical conditions given low anthropogenic disturbance scores (Dl) (Silva

et al., 2018;Van Sickle and Paulsen, 2008). RR scores equal to 1 denote

the absence of an association between the biological indicator (O/E

score) and environmental variables linked to anthropogenic alterations

(Van Sickle and Paulsen, 2008). For a RR >1, we interpret the value as

how many times more likely a poor O/E

value would occur given

poor/high disturbance conditions relative to good/low disturbance

conditions. We calculated 95% conﬁdence intervals for RR estimations

using conditional probabilities (Altman, 1991) and considered the RR

signiﬁcant when the lower 95% conﬁdence interval was >1.

3. Results

3.1. Macroinvertebrate taxa and reference site selection

We identiﬁed 97 taxa across the 381 sites. The most abundant taxa

were Chironomidae (41.99%), Simuliidae (10.36%), Elmidae (7.25%),

Baetidae (6.87%), and Oligochaeta (6.47%).

Eighty-seven sites met the selection criteria for minimally disturbed

reference sites and were used for MINASPACS model construction.

Reference and test sites occurred at similar elevations (510.59–1455.69

m and 411.00–1419.56 m a.s.l., respectively), temperature ranges

(16.89–23.75 C and 17.40–24.06 C), and annual precipitation

(1019.04–1675.43 mm and 972.99–1670.89 mm) (Table 5). However,

test sites had larger catchment areas, Total Phosphorus, Total Nitrogen,

Turbidity, and % total anthropogenic land uses than reference sites

(Table 4). More than 70% of our reference sites were in the S~

ao Francisco

River basin, with small catchment areas (<100 km

); only three sites had

a catchment area >1000 km

The boxplot analysis indicated a notable correspondence between the

environmental conditions of the 87 reference catchments and the 87

randomly selected catchments for mean altitude, annual temperature

range, and mean catchment slope (Supplementary Information 1,

Fig. S1). Although the 87 randomly selected catchments encompassed a

wider range of environmental conditions, the mean values, as well as the

25th and 75th percentiles, minimum, and maximum values, exhibited

minimal relative or absolute differences between the two groups (e.g.,

mean altitude: reference ¼725 m, random ¼723 m; mean catchment

slope: reference ¼4.7%, random ¼7.0%; annual temperature range:

reference ¼18.7 C, random ¼18.4 C). Conversely, annual mean pre-

cipitation and annual mean temperature showed greater differences.

However, those differences remained relatively small in absolute terms

for median values (e.g., annual mean precipitation: <200 mm; annual

mean temperature: <1C).

3.2. MINASPACS construction and validation

The MINASPACS model was built with 78 minimally disturbed cali-

bration sites, and 9 sites were used for validation. We deﬁned four

reference macroinvertebrate groups from the cluster analysis of the 78

calibration sites. All reference groups contained at least 14 reference

sites. The most representative families in each group were Psephenidae

and Pleidae for Group 1, Hydrobiosidae for Group 2, and Mega-

podagrionidae and Lutrochidae for Group 3. Group 4 had no exclusive

representative taxon (Supplementary Information 1,Table S2).

Four variables (mean catchment slope, lithological synthesis, annual

mean temperature, and annual mean precipitation) were selected for the

ﬁnal predictive O/E

model. Wilks' λwas low (0.148), and F-stat was

high (16.561), indicating the model's high discriminatory ability. This is

supported by the high accuracy evidenced by the MNO/E ¼1.001, SDO/

E¼0.16, and the O/E

regression was within acceptable values (R

0.608; slope ¼1.016; intersection ¼0.134). The 9 validation sites had

similar O/E

values (MNO/E ¼1.03; SDO/E ¼0.13), which suggested

good evaluation potential for test sites (Fig. 2).

3.3. Relative risk assessment and site disturbance classiﬁcation

Across the entire Cerrado biome, O/E

values ranged from poor/very

poor conditions (31% of test sites) to good/high (52% of test sites), with

17% in moderate condition (Fig. 3). Four anthropogenic disturbance

variables had signiﬁcant negative rank correlations with O/E

scores: %

urban infrastructure (r¼0.43; p<0.05), % total anthropogenic land

use (r¼0.22; p<0.05), Total Nitrogen (r¼0.20; p<0.05), and

Turbidity (r¼0.35; p<0.05). O/E

values were also negatively, but

weakly, correlated with Total Phosphorus (r¼0.09; p<0.05) and %

pasture (r¼0.04; p<0.05) (Supplementary Information 1,Table S3).

We found that MINASPACS could detect the inﬂuence of all seven

disturbance variables, but variable importance differed across the Cer-

rado biome and amongst river basins (Fig. 4). In the Cerrado and the S~

Francisco basin, all but % catchment agriculture signiﬁcantly inﬂuenced

the risk of low O/E

scores. On the other hand, only Total Nitrogen and

% urban infrastructure signiﬁcantly inﬂuenced the risk of low O/E

scores in the Alto Paran

a basin.

4. Discussion

Our predictive modeling results corroborate previous work from

Europe (e.g., Davy-Bowker et al., 2006;Mendes et al., 2014), North

America (e.g., Hargett et al., 2007;Hawkins et al., 2000), Asia (Chen

et al., 2019), and South America (e.g., Jovem-Azev^

edo et al., 2020;Moya

et al., 2011) that show predictive models can provide a powerful tool to

assess the biological condition of aquatic ecosystems. The relative risk

approach conﬁrmed the sensitivity of the MINASPACS to key distur-

bances affecting Minas Gerais lotic macroinvertebrate taxa richness.

4.1. MINASPACS model construction and validation

MINASPACS was built with variables exclusively obtained through

geospatial tools, which supports model development being a useful

approach for managers in terms of cost and time for data collection

(Hargett et al., 2007). The selection of model predictive variables

considered both statistical measures and the experience of model

development in other countries such as Great Britain (e.g., Wright, 1995),

Australia (Smith et al., 1999), the USA (Hawkins et al., 2000), and

Table 4

Land use and water quality criteria for distinguishing poor versus good anthro-

pogenic disturbance conditions.

Disturbances Pressure

condition

Source

Good Poor

% Agriculture <60 60 Silva et al. (2017)

% Pasture <60 60 Silva et al. (2017)

% Urban infrastructure 0 >0Lorenz et al., 2004

% Total anthropogenic land use

<25 25 Lorenz et al., 2004

Turbidity (NTU) 100 >100 CONAMA 357/2005, class II

Total Nitrogen (mg/L) 0.2 >0.2 Silva et al. (2017)

Total Phosphorus (mg/L) 0.1 >0.1 CONAMA 357/2005, class II

Class II water quality corresponds to water intended for human consumption

after simpliﬁed treatment and protection of aquatic communities (CONAMA,

2005).

All anthropogenic land uses combined (pasture, agriculture, monoculture,

mining, industrial area, and urban infrastructure).

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

Portugal (Feio et al., 2007,2009,2012). The four discriminant variables

ﬁnally selected for our model (mean catchment slope, lithological syn-

thesis, annual mean temperature, and annual precipitation) are aligned

with the results of other studies (e.g., Feio et al., 2007;Hargett et al.,

2007;Mendes et al., 2014). Variables such as temperature, which often

determines macroinvertebrate assemblage composition, is inﬂuenced by

latitude and elevation. At the same time, the geologic origin from which

bed materials are derived is also an important determinant of biotic

structure (Hawkins et al., 2000).

However, local environmental variables may enhance the accuracy

and precision of the model. Relevant variables may include annual

runoff, alkalinity, width, depth, and ﬂow regime (Davy-Bowker et al.,

2006;Moya et al., 2011; Pont et al., 2009). Martins et al. (2018) showed

that the taxonomic richness and composition of macroinvertebrate as-

semblages in Minas Gerais are positively affected by local variables, such

as leaf packs on the streambed, which serve as food and shelter against

predators (Ligeiro et al., 2020). In neotropical ecosystems, Macedo et al.

(2014) and Moya et al. (2011) showed that variables related to stream

size (wetted width, bank full width, and wetted area) are positively

correlated with macroinvertebrate richness. Thus, future work should

aim at combining regional and local variables to enhance the accuracy

and precision of the model.

MINASPACS was built using four macroinvertebrate reference

groups, which were well-discriminated by the environmental variables,

covering 40,106 km

. More groups would result in fewer sites per group,

reducing the model performance in the test site assessments. For

instance, Feio et al. (2007) developed and validated a multivariate model

for the Mondego catchment (6670 km

) using two reference groups,

whereas the RIVPACS study covering the UK (approx. 240,000 km

) used

35 groups. However, the grouping process is one of the most subjective

components of the modeling and should be reviewed in the future if new

high-quality reference sites are added (Feio et al., 2007).

Like Sudaryanti et al. (2001) noted in a previous study, family-level

identiﬁcation of macroinvertebrates was efﬁcient for our model

Table 5

Means and ranges of values for descriptive environmental variables at reference and test sites.

Variable Reference sites (n¼87) Test sites (n¼294)

Mean Min Max Mean Min Max

Annual Mean Temperature (C) 19.96 16.89 23.75 21.30 17.40 24.06

Annual Mean Precipitation (mm) 1,424.68 1,019.04 1,675.43 1,440.90 972.99 1,670.89

Mean catchment slope (%) 15.5 2.8 48.5 10.2 2.2 46.1

Altitude (m) 943.11 510.59 1,455.69 737.84 411.00 1,419.56

Distance to source (m) 12,429.75 68.71 85,906.92 29,075.95 46.00 706,308.64

Catchment area (km

) 132.28 0.03 1,809.83 606.20 0.00 27,923.96

% Forest 29.01 0.00 100.00 16.09 0.00 75.01

% Savanna 21.71 0.00 83.36 9.50 0.00 79.68

% Pasture 3.97 0.00 21.66 30.03 0.00 86.44

% Agriculture 1.77 0.00 19.35 5.20 0.00 37.45

% Urban infrastructure 0.00 0.00 0.00 6.48 0.00 100.00

% Anthropogenic use 8.32 0.00 24.26 60.63 0.00 100.00

Total Phosphorus (mg/L) 0.01 0.00 0.10 0.28 0.00 11.66

Total Nitrogen (mg/L) 0.08 0.03 0.20 0.25 0.00 16.30

Turbidity (NTU) 6.35 0.10 61.00 15.79 0.30 433.00

Taxa richness 29 16 41 19 0 43

O/E

score 1.01 0.47 1.27 0.62 0.00 1.26

Fig. 2. O/E

regression for the MINASPACS model built with biological and environmental data from 78 reference sites.

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

construction. This is an appropriate taxonomic resolution in many trop-

ical regions with high diversity but limited taxonomic knowledge (Godoy

et al., 2019), and it is the taxonomic level used in most Brazilian benthic

macroinvertebrate studies and in biomonitoring programs of govern-

mental agencies and other decision-makers. Species and genus-level

identiﬁcation are not currently feasible in Brazil because of insufﬁcient

knowledge of the autecology and life histories of most taxa and, there-

fore, a lack of taxonomic keys for many groups (Buss and Vitorino, 2010).

Moreover, family-level resolution is widely used in Europe for predictive

model development (e.g., Mendes et al., 2014) and other approaches,

such as General Degradation Indices (GDIs), including the Biological

Monitoring Working Party (BMWP) and the Average Score Per Taxon

(ASPT) indices (Vitecek et al., 2021). In Brazil, predictive models were

successfully developed using family-level taxonomic resolution (e.g.,

Jovem-Azev^

edo et al., 2020;Moreno et al., 2009). Nevertheless, Molozzi

et al. (2012) highlighted the importance of using genus-level Chirono-

midae (Diptera) in reservoir assessment because different genera have

different sensitivities to organic and metal contaminants. Conversely, a

higher taxonomic resolution requires taxonomic expertise and is more

time-consuming and costly (Feio et al., 2006;Vadas et al., 2022;Valen-

te-Neto et al., 2021). These are critical concerns in Brazil and elsewhere

(Buss et al., 2015).

Although the environmental conditions of the 87 reference catch-

ments and the 87 randomly selected catchments were similar (Fig. 2;

Table 4), we attribute the observed differences to the random selection

process. The random catchments were distributed throughout the Cer-

rado, including lower-lying regions, which are generally not well-

preserved catchments (Silva et al., 2018). In contrast, the reference

sites were predominantly located in the higher parts of river basins.

Therefore, if only preserved basins were considered in the random group,

the differences would probably be even smaller. Furthermore, because

Minas Gerais is heterogeneous in terms of its physical environment and

macroinvertebrate diversity, future studies should aim to increase the

number of reference sites and expand the sampling area to adjacent ba-

sins and states for more precise and accurate assessments. However,

minimally disturbed areas are generally scarce in Minas Gerais, much of

Europe (Borgwardt et al., 2019;Lorenz et al., 2004;Oliveira et al., 2016),

and large parts of the USA (Herlihy et al., 2020;Whittier et al., 2007).

Fig. 3. O/E

classiﬁcation for all sites.

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

4.2. Relative risk assessment, O/E classiﬁcation, and ecosystem

management

The MINASPACS detected the inﬂuence of all 7 anthropogenic dis-

turbances considered in this study and urban infrastructure posed the

most signiﬁcant risk to biological conditions. Land use is a proxy for

multiple pressures, including pollution, hydromorphological alterations,

and other pressures (Ligeiro et al., 2013;Poikane et al., 2020). In the

Cerrado assessment, high turbidity was a good indicator (high relative

risk) of poor biological condition, and Total Nitrogen posed a signiﬁcant

risk across all river basins. The latter reﬂects the application of nitrogen

fertilizers in croplands (Signor et al., 2013) and pasture (Pereira et al.,

2022) that commonly occurs in Minas Gerais. Both Turbidity and Total

Nitrogen are indicators of anthropogenic activities leading to sedimen-

tation and eutrophication (Silva et al., 2018). However, Total Phosphorus

and Total Nitrogen were insigniﬁcant stressors in some cases simply

because their criteria did not exceed those established by the Brazilian

water resources legislation (CONAMA, 2005).

Most of our sites were in the das Velhas River basin, which is highly

impaired in the Belo Horizonte Metropolitan Region, the capital of Minas

Gerais with a human population of 2.7 million. In the past two decades,

hydrologic modiﬁcations, channelization, sedimentation, nutrient load-

ings, heavy metals contamination, and microplastics have potentially

affected macroinvertebrate assemblages in the basin (Feio et al., 2013).

Impairment was prevalent for rivers and streams in the lowlands because

the population of Minas Gerais mainly occurs in lowland river sections.

Areas with high densities of test sites showing poor or very poor O/E

values were predominantly located in the plains and near major urban

centers. In Minas Gerais and throughout Brazil, rehabilitation in-

terventions focused mainly on sewage collection systems, rehabilitation

of riverbanks through engineering and bioengineering techniques,

rehabilitation of riparian vegetation, and ﬂood mitigation (Feio et al.,

2021). The lack of rehabilitation projects in Global South countries is

partly explained by the lack of ﬁnancial resources and by the lack of a

legal framework aimed at conserving and rehabilitating streams (Feio

et al., 2021;Macedo et al., 2022). Additionally, there is a lack of

empirical data on relevant geographical and long-term studies required

for assessing rehabilitation success. This challenge is evident not only in

Brazil (Buss et al., 2015) but also in Europe (Hering et al., 2010) and the

USA (Hughes et al., 2014).

Using different O/E

classes to represent good and poor biological

conditions and different land use classes and nutrient criteria to represent

reference conditions would likely affect the biological assessments (Feio

et al., 2014;Herlihy et al., 2020). Decreased or increased macro-

invertebrate richness in small streams could occur with much lower

nutrient concentrations than the values set in legislation or regulations,

at least in naturally nutrient-poor regions (Davies and Jackson, 2006;

Firmiano et al., 2017;Herlihy et al., 2020). Nonetheless, grading sites

into classes of “ecological status”for surface waters is now a requirement

in Minas Gerais (COPAM, 2022). The MINASPACS O/E

ratio may be a

useful metric for assessing the condition of macroinvertebrate assem-

blages as a constituent of the ecological status of rivers. Also, new models

based on this approach can be built to develop biological assessment

methods statewide, thereby yielding improved scientiﬁc generalizations,

assessments, and regulation (Stoddard et al., 2008;Vadas et al., 2022).

Lastly, employing tiered aquatic life uses and biological condition

gradient approaches would offer ecologically more meaningful classiﬁ-

cations of site condition (Davies and Jackson, 2006;Hughes et al., 2022;

Yoder et al., 2005).

4.3. Uncertainties and limitations

Sampling effort is fundamental to aquatic bioassessment because it

provides the base data on which water bodies are classiﬁed (Birk et al.,

2012). In this context, consistent sampling effort is crucial for

richness-based assessments, because reduced sampling effort increases

the risk of errors in interpreting results and leads to an underestimation

of taxa richness at a given site (Cao et al., 2002;Li et al., 2001,2014;

Silva et al., 2016). Future studies should consider greater sampling effort

and standardized procedures at both basin and site extents than typically

employed in stream ecological surveys when the goal is to assess taxo-

nomic richness. This would help maximize the representation of diversity

across different environmental contexts (Li et al., 2001;Silva et al.,

2016).

Fig. 4. Estimated relative risk (RR) and 95% conﬁdence intervals for poor/very poor O/E

scores linked to anthropogenic disturbances. RR conﬁdence intervals

below 1 (dashed red line) indicate insigniﬁcant association.

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

Additionally, by using genus-level taxonomy, the model could be

more capable of detecting impairment at test sites. This is because for

regions with many genera and species per family, important information

on species-speciﬁc taxon-habitat relationships could easily be lost by

adopting family-level taxonomic resolution because of the differing

ecological requirements of different species and genera within a family

(Hawkins et al., 2000;Lenat and Resh, 2001;Molozzi et al., 2012).

We did not analyze the effects of other pollutants, such as heavy

metals, acidiﬁcation, or hydromorphological degradation, all of which

can limit the presence and development of sensitive organisms (de Mello

et al., 2023). Another critical consideration is that using land use as a

proxy for multiple pressures does not always clarify which land-use

pressures are most responsible for an observed response (Poikane

et al., 2020). Ecological thresholds or tipping points are a useful

ecological concept but are rarely detectable in multi-pressure environ-

ments (Birk et al., 2012;Groffman et al., 2006), particularly in tropical

regions where the coarser taxonomic resolution may constrain the

detection of ecological impairment at test sites (Buss and Vitorino, 2010).

5. Summary

We developed and validated the MINASPACS predictive model based

on macroinvertebrates that can fulﬁll all scientiﬁc aspects required for

classiﬁcation systems under the present Minas Gerais water resource

legislation. Our model responded to seven disturbances impairing stream

ecosystems in southeast Brazil, providing a clear, simple, and defensible

measure of macroinvertebrate taxa richness of streams in a diverse

landscape. We demonstrated the feasibility of implementing a RIVPACS-

like approach, and we hope improved models based on this approach can

be built to develop biological assessments throughout Brazil.

CRediT authorship contribution statement

Pedro Fialho Cordeiro: Writing –original draft, Methodology,

Investigation, Formal analysis, Data curation, Conceptualization. Maria

Jo~

ao Feio: Writing –review &editing, Supervision, Methodology,

Conceptualization. Marcos Callisto: Writing –review &editing, Vali-

dation, Conceptualization. Robert M. Hughes: Writing –review &

editing, Supervision, Conceptualization. Diego Rodrigues Macedo:

Writing –review &editing, Validation, Supervision, Project adminis-

tration, Methodology, Investigation, Data curation, Conceptualization.

Funding

This research was supported by Companhia Energ

etica de Minas

Gerais (CEMIG) through Programa Cemig-Peixe Vivo, Programa de Pes-

quisa e Desenvolvimento Aneel (Cemig GT-479; GT- 487; GT-550; GT-

599), and Cemig/Fapemig (APQ-01961-15, APQ-00261-22, and CRA

3147); Fundaç~

ao de Amparo 

a Pesquisa do Estado de Minas Gerais –

FAPEMIG (APQ-01432-17 and APQ-261-22); Conselho Nacional de

Desenvolvimento Cientíﬁco e Tecnol

ogico (CNPq 311002/2023-4 to

DRM, and 304060/2020-8 to MC); MC is a Resident Professor at the

Institute of Advanced Transdisciplinary Studies (IEAT/UFMG); Fundaç~

para a Ci^

encia e Tecnologia through MARE strategic project (UIDB/

04292/2020), Associate Laboratory ARNET Project (LA/P/0069/2020),

and CEEC principal investigator to MJF; Coordenaç~

ao de Aperfeiçoa-

mento de Pessoal de Nível Superior (CAPES) Finance Code 001;

Fulbright-Brazil to RMH; and Projeto Manuelz~

ao/UFMG.

Declaration of competing interest

As a Co-editor-in-Chief of Water Biology and Security, Robert M.

Hughes examines the English and content of some manuscripts, but he

was not involved in the manuscript editorial or peer reviews or the de-

cision to publish this article. All authors declare no known competing

ﬁnancial interests or personal relationships that could potentially

inﬂuence the work reported in this paper.

Acknowledgments

We thank the ﬁeld crews for generating the large amount of data used

in this study, mainly colleagues from the Laborat

orio de Ecologia de

Bentos (Universidade Federal de Minas Gerais –UFMG), Laborat

orio de

Ecologia de Peixes (Universidade Federal de Lavras–UFLA), and Centro

de Inovaç~

ao e Tecnologia SENAI (Serviço Nacional de Aprendizagem

Industrial).

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://

doi.org/10.1016/j.watbs.2025.100386.

References

Agra, J.U.M., Ligeiro, R., Macedo, D.R., Hughes, R.M., Callisto, M., 2019. Ecoregions and

stream types help us understand ecological variability in Neotropical reference

streams. Mar. Freshw. Res. 70, 594–602. https://doi.org/10.1071/MF18309.

Aguiar, F.C., Feio, M.J., Ferreira, M.T., 2011. Choosing the best method for stream

bioassessment using macrophyte communities: indices and predictive models. Ecol.

Indic. 11, 379–388. https://doi.org/10.1016/j.ecolind.2010.06.006.

Aida Campos, C., Kennard, M.J., Gonçalves Júnior, J.F., 2021. Diatom and

Macroinvertebrate assemblages to inform management of Brazilian savanna's

watersheds. Ecol. Indic. 128. https://doi.org/10.1016/j.ecolind.2021.107834.

Altman, D.G., 1991. Practical statistics for medical research. Aust. N. Z. J. Surg. 61,

963–964. https://doi.org/10.1111/j.1445-2197.1991.tb00019.x.

APHA, 2005. American public health association. Standard Methods for the Examination

of Water and Wastewater, twenty-ﬁrst ed. American Water Works Association/Water

Environment Federation.

Barbour, M.T., Stribling, J.B., Snyder, B.D., Gerritsen, J., 1999. Rapid bioassessment

protocols for use in streams and wadeable rivers. EPA 841/B-99/002. Ofﬁce of Water,

U.S. Environmental Protection Agency, Washington, DC.

Birk, S., Bonne, W., Borja, A., Brucet, S., Courrat, A., Poikane, S., Solimini, A., Van De

Bund, W., Zampoukas, N., Hering, D., 2012. Three hundred ways to assess Europe's

surface waters: an almost complete overview of biological methods to implement the

Water Framework Directive. Ecol. Indic. 18, 31–41. https://doi.org/10.1016/

j.ecolind.2011.10.009.

Blakely, T.J., Eikaas, H.S., Harding, J.S., 2014. The SingScore: a macroinvertebrate biotic

index for assessing the health of Sinms and canals. Rafﬂes Bull. Zool. 62, 540–548.

Borgwardt, F., Leitner, P., Graf, W., Birk, S., 2019. Ex uno plures –Deﬁning different types

of very large rivers in Europe to foster solid aquatic bio-assessment. Ecol. Indic. 107,

105599. https://doi.org/10.1016/j.ecolind.2019.105599.

Buss, D.F., Carlisle, D.M., Chon, T.-S., Culp, J., Harding, J.S., Keizer-Vlek, H.E.,

Robinson, W.A., Strachan, S., Thirion, C., Hughes, R.M., 2015. Stream biomonitoring

using macroinvertebrates around the globe: a comparison of large-scale programs.

Environ. Monit. Assess. 187, 4132. https://doi.org/10.1007/s10661-014-4132-8.

Buss, D.F., Vitorino, A.S., 2010. Rapid Bioassessment Protocols using benthic

macroinvertebrates in Brazil: evaluation of taxonomic sufﬁciency. J. North Am.

Benthol. Soc. 29, 562–571. https://doi.org/10.1899/09-095.1.

Callisto, M., Linares, M.S., Kiffer, W.P., Hughes, R.M., Moretti, M.S., Macedo, D.R.,

Solar, R., 2021. Beta diversity of aquatic macroinvertebrate assemblages associated

with leaf patches in neotropical montane streams. Ecol. Evol. 11, 2551–2560.

https://doi.org/10.1002/ece3.7215.

Callisto, M., Macedo, D.R., Linares, M.S., Hughes, R.M., 2019. Multi-status and multi-

spatial scale assessment of landscape effects on benthic macroinvertebrates in the

Neotropical Savanna. Adv. Underst. Landsc. Inﬂu. Freshw. Habitats Biol. Assem

275–302.

Campos, L., Guedes, M.L.S., Acevedo-Rodríguez, P., Roque, N., 2017. Contributions to the

ﬂoristic and vegetation knowledge of Espinhaço septentrional, bahia, Brazil. Rev.

Bras. Bot. 40, 427–437. https://doi.org/10.1007/s40415-016-0347-y.

Cao, Y., Larsen, D.P., Hughes, R.M., Angermeier, P.L., Patton, T.M., 2002. Sampling effort

affects multivariate comparisons of stream assemblages. J. North Am. Benthol. Soc.

21, 701–714. https://doi.org/10.2307/1468440.

Chen, K., Razzaque, A., Hughes, R.M., Olson, J.R., Wei, H., Wang, B., 2019. Incorporating

functional traits to enhance multimetric index performance and assess land use

gradients. Sci. Total Environ. 691, 1005–1015. https://doi.org/10.1016/

j.scitotenv.2019.07.047.

Clarke, R.T., Wright, J.F., Furse, M.T., 2003. RIVPACS models for predicting the expected

macroinvertebrate fauna and assessing the ecological quality of rivers. Ecol. Model.

160, 219–233. https://doi.org/10.1016/S0304-3800(02)00255-7.

CONAMA, 2005. Conselho Nacional Do Meio Ambiente. Resoluç~

ao CONAMA N357/

2005. Di

ario Oﬁcial da República Federativa do Brasil, Brasília, DF.

COPAM, 2022. Conselho Estadual de Política Ambiental. Resoluç~

ao Normativa Conjunta

COPAM/CERH-MG N

008/2022. Di

ario Executivo, Minas Gerais.

Dala-Corte, R.B., Melo, A.S., Siqueira, T., Bini, L.M., Martins, R.T., Cunico, A.M.,

Pes, A.M., Magalh~

aes, A.L.B., Godoy, B.S., Leal, C.G., Monteiro-Júnior, C.S.,

Stenert, C., Castro, D.M.P., Macedo, D.R., Lima, D., Gubiani,



E.A., Massariol, F.C.,

Teresa, F.B., Becker, F.G., Souza, F.N., Valente-Neto, F., de Souza, F.L., Salles, F.F.,

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

Brej~

ao, G.L., Brito, J.G., Vitule, J.R.S., Simi~

ao-Ferreira, J., Dias-Silva, K.,

Albuquerque, L., Juen, L., Maltchik, L., Casatti, L., Montag, L., Rodrigues, M.E.,

Callisto, M., Nogueira, M.A.M., dos Santos, M.R., Hamada, N., Pamplin, P.A.Z.,

Pompeu, P.S., Leit~

ao, R.P., Ruaro, R., Mariano, R., Couceiro, S.R.M., Abilhoa, V.,

Oliveira, V.C., Shimano, Y., Moretto, Y., Suarez, Y.R., Roque, F.de O., 2020.

Thresholds of freshwater biodiversity in response to riparian vegetation loss in the

Neotropical region. J. Appl. Ecol. 1–12. https://doi.org/10.1111/1365-2664.13657.

Davies, S.P., Jackson, S.K., 2006. The Biological Condition Gradient: a descriptive model

for interpreting change in aquatic ecosystems. Ecol. Appl. 16, 1251–1266.

Davy-Bowker, J., Clarke, R.T., Johnson, R.K., Kokes, J., Murphy, J.F., Zahr

adkov

a, S.,

2006. A comparison of the European Water Framework Directive physical typology

and RIVPACS-type models as alternative methods of establishing reference conditions

for benthic macroinvertebrates. Hydrobiologia 566, 91–105. https://doi.org/

10.1007/s10750-006-0068-5.

de Mello, K., Taniwaki, R.H., Macedo, D.R., Randhir, T.O., 2023. Biomonitoring for

Watershed Protection from a Multiscale Land-Use Perspective. diversity 15, 1–20.

https://doi.org/10.3390/d15050636.

Dudgeon, D., 2010. Prospects for sustaining freshwater biodiversity in the 21st century:

linking ecosystem structure and function. Curr. Opin. Environ. Sustain. 2, 422–430.

https://doi.org/10.1016/j.cosust.2010.09.001.

Eriksen, T.E., Brittain, J.E., Søli, G., Jacobsen, D., Goethals, P., Friberg, N., 2021. A global

perspective on the application of riverine macroinvertebrates as biological indicators

in Africa, South-Central America, Mexico and Southern Asia. Ecol. Indic. 126.

https://doi.org/10.1016/j.ecolind.2021.107609.

Feio, M.J., Aguiar, F.C., Almeida, S.F.P., Ferreira, J., Ferreira, M.T., Elias, C., Serra, S.R.Q.,

Buffagni, A., Cambra, J., Chauvin, C., Delmas, F., D€

orﬂinger, G., Erba, S., Flor, N.,

Ferr

eol, M., Germ, M., Mancini, L., Manolaki, P., Marcheggiani, S., Minciardi, M.R.,

Munn

e, A., Papastergiadou, E., Prat, N., Puccinelli, C., Rosebery, J., Sabater, S.,

Ciadamidaro, S., Torn

es, E., Tziortzis, I., Urbani

c, G., Vieira, C., 2014. Least disturbed

condition for European mediterranean rivers. Sci. Total Environ. 476–477, 745–756.

https://doi.org/10.1016/j.scitotenv.2013.05.056.

Feio, M.J., Aguiar, F.C., Almeida, S.F.P., Ferreira, M.T., 2012. AQUAFLORA: a predictive

model based on diatoms and macrophytes for streams water quality assessment. Ecol.

Indic. 18, 586–598. https://doi.org/10.1016/j.ecolind.2012.01.016.

Feio, M.J., Ferreira, W.R., Macedo, D.R., Eller, A.P., Alves, C.B.M., França, J.S.,

Callisto, M., 2013. Deﬁning and testing targets for the recovery of tropical streams

based on macroinvertebrate communities and abiotic conditions. River Res. Appl. 31,

70–84. https://doi.org/10.1002/rra.2716.

Feio, M.J., Hughes, R.M., Callisto, M., Nichols, S.J., Odume, O.N., Quintella, B.R.,

Kuemmerlen, M., Aguiar, F.C., Almeida, S.F.P., Alonso-EguíaLis, P., Arimoro, F.O.,

Dyer, F.J., Harding, J.S., Jang, S., Kaufmann, P.R., Lee, S., Li, J., Macedo, D.R.,

Mendes, A., Mercado-Silva, N., Monk, W., Nakamura, K., Ndiritu, G.G., Ogden, R.,

Peat, M., Reynoldson, T.B., Rios-Touma, B., Segurado, P., Yates, A.G., 2021. The

biological assessment and rehabilitation of the world's rivers: an overview. Water 13,

1–46. https://doi.org/10.3390/w13030371.

Feio, M.J., Hughes, R.M., Serra, S.R.Q., Nichols, S.J., Kefford, B.J., Lintermans, M.,

Robinson, W., Odume, O.N., Callisto, M., Macedo, D.R., Harding, J.S., Yates, A.G.,

Monk, W., Nakamura, K., Mori, T., Sueyoshi, M., Mercado-Silva, N., Chen, K.,

Baek, M.J., Bae, Y.J., Tachamo-Shah, R.D., Shah, D.N., Campbell, I., Moya, N.,

Arimoro, F.O., Keke, U.N., Martins, R.T., Alves, C.B.M., Pompeu, P.S., Sharma, S.,

2022. Fish and macroinvertebrate assemblages reveal extensive degradation of the

world’s rivers. Glob. Change Biol. 1–20. https://doi.org/10.1111/gcb.16439.

Feio, M.J., Norris, R.H., Graça, M.A.S., Nichols, S., 2009. Water quality assessment of

Portuguese streams: regional or national predictive models? Ecol. Indic. 9, 791–806.

https://doi.org/10.1016/j.ecolind.2008.09.012.

Feio, M.J., Poquet, J.M., 2011. Predictive models for freshwater biological assessment:

statistical approaches, biological elements and the iberian peninsula experience: a

review. Int. Rev. Hydrobiol. 96, 321–346. https://doi.org/10.1002/iroh.201111376.

Feio, M.J., Reynoldson, T.B., Ferreira, V., Graça, M.A.S., 2007. A predictive model for

freshwater bioassessment (Mondego River, Portugal). Hydrobiologia 589, 55–68.

https://doi.org/10.1007/s10750-006-0720-0.

Feio, M.J., Reynoldson, T.B., Graça, M.A., 2006. Effect of seasonal changes on predictive

model assessments of streams water quality with macroinvertebrates. Int. Rev.

Hydrobiol. 91, 509–520. https://doi.org/10.1002/iroh.200610877.

Feio, M.J., Viana-Ferreira, C., Costa, C., 2014. Testing a multiple machine learning tool

(HYDRA) for the bioassessment of fresh waters. Freshw. Sci. 33, 1286–1296. https://

doi.org/10.1086/678768.

Ferreira, H.L.M., Melo, M.C., Freitas, M. d,



Avila, F.P., de, P., Meyer, S.T., Campos, M. de

C.S., Felipe, G.A., 2017. Ambientes Aqu

aticos Em Minas Gerais, 1a ediç~

ao. Belo

Horizonte, Prodemge.

Firmiano, K.R., Ligeiro, R., Macedo, D.R., Juen, L., Hughes, R.M., Callisto, M., 2017.

Mayﬂy bioindicator thresholds for several anthropogenic disturbances in neotropical

savanna streams. Ecol. Indic. 74, 276–284. https://doi.org/10.1016/

J.ECOLIND.2016.11.033.

Gao, Y., Qian, H., Huo, C., Chen, J., Wang, H., 2020. Assessing natural background levels

in shallow groundwater in a large semiarid drainage Basin. J. Hydrol. 584, 124638.

https://doi.org/10.1016/j.jhydrol.2020.124638.

Garuana, L., Macedo, D.R., Matta-Machado, A.T.G., Callisto, M., 2020. Integraç~

ao de

indicadores ecol

ogicos, ambientais e de saúde humana em microbacias urbanas. Rev.

Espinhaço 9, 1–16. https://doi.org/10.5281/zenodo.3937470.

Godoy, B.S., Faria, A.P.J., Juen, L., Sara, L., Oliveira, L.G., 2019. Taxonomic sufﬁciency

and effects of environmental and spatial drivers on aquatic insect community. Ecol.

Indic. 107, 105624. https://doi.org/10.1016/j.ecolind.2019.105624.

Grecchi, R.C., Gwyn, Q.H.J., B

eni

e, G.B., Formaggio, A.R., Fahl, F.C., 2014. Land use and

land cover changes in the Brazilian Cerrado: a multidisciplinary approach to assess

the impacts of agricultural expansion. Appl. Geogr. 55, 300–312. https://doi.org/

10.1016/j.apgeog.2014.09.014.

Groffman, P.M., Baron, J.S., Blett, T., Gold, A.J., Goodman, I., Gunderson, L.H.,

Levinson, B.M., Palmer, M.A., Paerl, H.W., Peterson, G.D., Poff, N.L.R., Rejeski, D.W.,

Reynolds, J.F., Turner, M.G., Weathers, K.C., Wiens, J., 2006. Ecological thresholds:

the key to successful environmental management or an important concept with no

practical application? Ecosystems 9, 1–13. https://doi.org/10.1007/s10021-003-

0142-z.

Hamada, N., Nessimian, J., Querino, R., 2014. Insetos aqu

aticos na Amaz^

onia brasileira:

Taxonomia, biologia e ecologia. INPA, Manaus, Brazil.

Hargett, E.G., ZumBerge, J.R., Hawkins, C.P., Olson, J.R., 2007. Development of a

RIVPACS-type predictive model for bioassessment of wadeable streams in Wyoming.

Ecol. Indic. 7, 807–826. https://doi.org/10.1016/j.ecolind.2006.10.001.

Hawkins, C.P., Norris, R.H., Hogue, J.N., Feminella, J.W., 2000. Development and

evaluation of predictive models for measuring the biological integrity of streams.

Ecol. Appl. 10, 1456–1477.

Hering, D., Borja, A., Carstensen, J., Carvalho, L., Elliott, M., Feld, C.K., Heiskanen, A.S.,

Johnson, R.K., Moe, J., Pont, D., Solheim, A.L., de Bund, W. van, 2010. The European

Water Framework Directive at the age of 10: a critical review of the achievements

with recommendations for the future. Sci. Total Environ. 408, 4007–4019. https://

doi.org/10.1016/j.scitotenv.2010.05.031.

Herlihy, A.T., Paulsen, S.G., Van Sickle, J., Stoddard, J.L., Hawkins, C.P., Yuan, L.L., 2008.

Striving for consistency in a national assessment: the challenges of applying a

reference-condition approach at a continental scale. J. North Am. Benthol. Soc. 27,

860–877. https://doi.org/10.1899/08-081.1.

Herlihy, Alan T., Sifneos, J.C., Hughes, R.M., Peck, D.V., Mitchell, R.M., 2020. The

relation of lotic ﬁsh and benthic macroinvertebrate condition indices to

environmental factors across the conterminous USA. Ecol. Indic. 112, 105958.

https://doi.org/10.1016/j.ecolind.2019.105958.

Hughes, R.M., Dunham, S., Maas-Hebner, K.G., Yeakley, J.A., Schreck, C., Harte, M.,

Molina, N., Shock, C.C., Kaczynski, V.W., Schaeffer, J., 2014. A review of urban water

body challenges and approaches: (1) rehabilitation and remediation. Fisheries 39,

18–29. https://doi.org/10.1080/03632415.2013.836500.

Hughes, R.M., Larsen, D.P., Omernik, J.M., 1986. Regional reference sites: a method for

assessing stream potentials. Environ. Manag. 10, 629–635. https://doi.org/10.1007/

BF01866767.

Hughes, R.M., Zeigler, M., Stringer, S., Linam, G.W., Flotemersch, J., Jessup, B.,

Joseph, S., Jacobi, G., Guevara, L., Cook, R., Bradley, P., Barrios, K., 2022. Biological

assessment of western USA sandy bottom rivers based on modeling historical and

current ﬁsh and macroinvertebrate data. River Res. Appl. 38, 639–656. https://

doi.org/10.1002/rra.3929.

Hunke, P., Mueller, E.N., Schr€

oder, B., Zeilhofer, P., 2015. The Brazilian Cerrado:

assessment of water and soil degradation in catchments under intensive agricultural

use. Ecohydrology 8, 1154–1180. https://doi.org/10.1002/eco.1573.

Johnson, R., Sandin, L., 2001. Development of a Prediction and Classiﬁcation System for

Lake (Littoral, SWEPAC) and Stream (Rifﬂe SWEPAC) Macroinvertebrate

Communities. Department of Environmental Assessment, Swedish University of

Agricultural Sciences, Uppsala, Sweden.

Jovem-Azev^

edo, D., Bezerra-Neto, J.F., Molozzi, J., Feio, M.J., 2020. Rehabilitation

scenarios for reservoirs: predicting their effect on invertebrate communities through

machine learning. River Res. Appl. 1–15. https://doi.org/10.1002/rra.3641.

Junqueira, M.V., Alves, K.C., Paprocki, H., Campos, M. de S., Carvalho, M.D. de,

Mota, H.R., Rolla, M.E., 2018.



Indices bi

oticos para avaliaç~

ao de qualidade de 

agua

de rios tropicais –síntese do conhecimento e estudo de caso: Bacia do alto rio doce.

Rev. Bras. Ci^

encia Avícola 15–33. https://doi.org/10.5327/z2176-947820180322.

Karr, J.R., 1999. Deﬁning and measuring river health. Freshw. Biol. 41, 221–234. https://

doi.org/10.1046/j.1365-2427.1999.00427.x.

Koke

s, J., Zahr

adkov

a, S., N

emejcov

a, D., Hodovský, J., Jarkovský, J., Sold

an, T., 2006.

The PERLA system in the Czech Republic: a multivariate approach for assessing the

ecological status of running waters. Hydrobiologia 566, 343–354. https://doi.org/

10.1007/s10750-006-0085-4.

Lake, P.S., 2000. Disturbance, patchiness, and diversity in streams. J. North Am. Benthol.

Soc. 19, 573–592. https://doi.org/10.2307/1468118.

Lenat, D.R., Resh, V.H., 2001. Taxonomy and stream ecology - the beneﬁts of genus- and

species-level identiﬁcations. J. North Am. Benthol. Soc. 20, 287–298. https://

doi.org/10.2307/1468323.

Li, J., Herlihy, A., Gerth, W., Kaufmann, P., Gregory, S., Urquhart, S., Larsen, D.P., 2001.

Variability in stream macroinvertebrates at multiple spatial scales. Freshw. Biol. 46,

87–97. https://doi.org/10.1111/j.1365-2427.2001.00628.x.

Li, L., Liu, L., Hughes, R.M., Cao, Y., Wang, X., 2014. Towards a protocol for stream

macroinvertebrate sampling in China. Environ. Monit. Assess. 186, 469–479. https://

doi.org/10.1007/s10661-013-3391-0.

Ligeiro, R., Hughes, R.M., Kaufmann, P.R., Heino, J., Melo, A.S., Callisto, M., 2020.

Choice of ﬁeld and laboratory methods affects the detection of anthropogenic

disturbances using stream macroinvertebrate assemblages. Ecol. Indic. 115, 106382.

https://doi.org/10.1016/j.ecolind.2020.106382.

Ligeiro, R., Hughes, R.M., Kaufmann, P.R., Macedo, D.R., Firmiano, K.R., Ferreira, W.R.,

Oliveira, D., Melo, A.S., Callisto, M., 2013. Deﬁning quantitative stream disturbance

gradients and the additive role of habitat variation to explain macroinvertebrate taxa

richness. Ecol. Indic. 25, 45–57. https://doi.org/10.1016/j.ecolind.2012.09.004.

Linares, M.S., Callisto, M., Macedo, D.R., Hughes, R.M., 2021. Chronic urbanization

decreases macroinvertebrate resilience to natural disturbances in neotropical

streams. Curr. Res. Environ. Sustain. 3, 100095. https://doi.org/10.1016/

j.crsust.2021.100095.

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

Linke, S., Norris, R.H., Faith, D.P., Stockwell, D., 2005. ANNA: a new prediction method

for bioassessment programs. Freshw. Biol. 50, 147–158. https://doi.org/10.1111/

j.1365-2427.2004.01286.x.

Lorenz, A., Feld, C.K., Hering, D., 2004. Typology of streams in Germany based on benthic

invertebrates: Ecoregions, zonation, geology and substrate. Limnologica 34,

379–389. https://doi.org/10.1016/S0075-9511(04)80007-0.

Macedo, D.R., Callisto, M., Linares, M.S., Hughes, R.M., Romano, B.M.L., Rothe-

Neves, M., Silveira, J.S., 2022. Urban stream rehabilitation in a densely populated

Brazilian metropolis. Front. Environ. Sci. 10, 921934. https://doi.org/10.3389/

fenvs.2022.921934.

Macedo, D.R., Hughes, R.M., Ligeiro, R., Ferreira, W.R., Castro, M.A., Junqueira, N.T.,

Oliveira, D.R., Firmiano, K.R., Kaufmann, P.R., Pompeu, P.S., Callisto, M., 2014. The

relative inﬂuence of catchment and site variables on ﬁsh and macroinvertebrate

richness in cerrado biome streams. Landsc. Ecol. 29, 1001–1016. https://doi.org/

10.1007/s10980-014-0036-9.

Martins, I., Ligeiro, R., Hughes, R.M., Macedo, D.R., Callisto, M., 2018. Regionalization is

key to establish reference conditions for neotropical savanna streams. Mar. Freshw.

Res. 69, 82–94. https://doi.org/10.1071/MF16381.

Martins, I., Macedo, D.R., Hughes, R.M., Callisto, M., 2021. Major risks to aquatic biotic

condition in a Neotropical Savanna River basin. River Res. Appl. 37, 858–868.

https://doi.org/10.1002/rra.3801.

Martins, I., Macedo, D.R., Hughes, R.M., Callisto, M., 2020. Are multiple multimetric

indices effective for assessing ecological condition in tropical basins? Ecol. Indic. 110,

105953. https://doi.org/10.1016/j.ecolind.2019.105953.

Mendes, T., Calapez, A.R., Elias, C.L., Almeida, S.F.P., Feio, M.J., 2014. Comparing

alternatives for combining invertebrate and diatom assessment in stream quality

classiﬁcation. Mar. Freshw. Res. 65, 612–623. https://doi.org/10.1071/MF13135.

Merritt, R., Cummins, K., 1996. An Introduction to the Aquatic Insects of North America.

Iowa, USA.

Mitchell, R.M., Herlihy, A.T., Hughes, R.M., 2025. Striving for consistency in a national

lake assessment: deﬁning reference status and littoral macroinvertebrate condition in

lakes across the conterminous United States. Ecol. Indic. 170, 112992. https://

doi.org/10.1016/j.ecolind.2024.112992.

Molozzi, J., Feio, M.J., Salas, F., Marques, J.C., Callisto, M., 2012. Development and test

of a statistical model for the ecological assessment of tropical reservoirs based on

benthic macroinvertebrates. Ecol. Indic. 23, 155–165. https://doi.org/10.1016/

j.ecolind.2012.03.023.

Moreno, P., França, J.S., Ferreira, W.R., Paz, A.D., Monteiro, I.M., Callisto, M., 2009. Use

of the BEAST model for biomonitoring water quality in a neotropical basin.

Hydrobiologia. https://doi.org/10.1007/s10750-009-9796-7.

Moya, N., Hughes, R.M., Domínguez, E., Goitia, E., 2011. Macroinvertebrate-based

multimetric predictive models for evaluating the human impact on biotic condition of

Bolivian streams. Ecol. Indic. 11, 840–847. https://doi.org/10.1016/

j.ecolind.2010.10.012.

Mugnai, R., Nessimian, J., Baptista, D., 2009. Guide for the Identiﬁcation of Aquatic

Macroinvertebrates of Rio de Janeiro State. Technical Books Editora, Rio de Janeiro.

Oliveira, R.B.S., Baptista, D.F., Mugnai, R., Castro, C.M., Hughes, R.M., 2011. Towards

rapid bioassessment of wadeable streams in Brazil: development of the Guapiau-

Macau Multimetric Index (GMMI) based on benthic macroinvertebrates. Ecol. Indic.

11, 1584–1593. https://doi.org/10.1016/j.ecolind.2011.04.001.

Oliveira, U., Paglia, A.P., Brescovit, A.D., Carvalho, C.J.B., Silva, D.P., Rezende, D.T.,

Leite, F.S.F., Batista, J.A.N., Barbosa, J.P.P.P., Stehmann, J.R., Ascher, J.S.,

Vasconcelos, M.F., De Marco, P., L€

owenberg-Neto, P., Dias, P.G., Ferro, V.G.,

Santos, A.J., 2016. The strong inﬂuence of collection bias on biodiversity knowledge

shortfalls of Brazilian terrestrial biodiversity. Divers. Distrib 22, 1232–1244. https://

doi.org/10.1111/ddi.12489.

Pardo, I., G

omez-Rodríguez, C., Abraín, R., García-Rosell

o, E., Reynoldson, T.B., 2014. An

invertebrate predictive model (NORTI) for streams and rivers: sensitivity of the

model in detecting stress gradients. Ecol. Indic. 45, 51–62. https://doi.org/10.1016/

j.ecolind.2014.03.019.

Paulsen, S.G., Peck, D.V., Kaufmann, P.R., Herlihy, A.T., 2020. Rivers and streams:

upgrading monitoring of the nation's freshwater resources - meeting the spirit of the

clean water act. In: Summers, K. (Ed.), Water Quality - Science, Assessments and

Policy. IntechOpen, p. 182. https://doi.org/10.5772/intechopen.77531.

Pereira, L.E.T., Herling, V.R., Tech, A.R.B., 2022. Current scenario and perspectives for

nitrogen fertilization strategies on tropical perennial grass pastures: a review.

Agronomy 12, 1–19. https://doi.org/10.3390/agronomy12092079.

P

erez, G.R., 1988. Guía para el estudio de los macroinvertebrados acu

aticos del

Departamento de Antioquia. Editorial Presencia Ltda, Bogot

a, Colombia, p. 217p.

P

es, A.M., Hamada, N., Nessimian, J.L., 2005. Identiﬁcation keys of larvae for families

and genera of Trichoptera (Insecta) of Central Amazonia, Brazil. Rev. Bras. Entomol.

49, 181–204.

Poikane, S., Johnson, R.K., Sandin, L., Schartau, A.K., Solimini, A.G., Urbani

c, G.,

Arba

ciauskas, K. stutis, Aroviita, J., Gabriels, W., Miler, O., Pusch, M.T., Timm, H.,

B€

ohmer, J., 2016. Benthic macroinvertebrates in lake ecological assessment: A review

of methods, intercalibration and practical recommendations. Sci. Total Environ. 543,

123–134. doi:10.1016/j.scitotenv.2015.11.021.

Poikane, S., Salas Herrero, F., Kelly, M.G., Borja, A., Birk, S., van de Bund, W., 2020.

European aquatic ecological assessment methods: a critical review of their sensitivity

to key pressures. Sci. Total Environ. 740, 140075. https://doi.org/10.1016/

j.scitotenv.2020.140075.

Pont, D., Hughes, R.M., Whittier, T.R., Stefan, S., 2009. A predictive index of biotic

integrity model for aquatic-vertebrate assemblages of western U . S . Streams. Trans.

Am. Fish. Soc. 138 (1), 292–305. https://doi.org/10.1577/T07-277.

Pont, D., Hugueny, B., Beier, U., Goffaux, D., Melcher, A., Noble, R., Rogers, C., Roset, N.,

Schmutz, S., 2006. Assessing river biotic condition at a continental scale : a European

approach using functional metrics and ﬁsh. J. Appl. Ecol. 43, 70–80. https://doi.org/

10.1111/j.1365-2664.2005.01126.x.

Reid, A.J., Carlson, A.K., Creed, I.F., Eliason, E.J., Gell, P.A., Johnson, P.T.J., Kidd, K.A.,

MacCormack, T.J., Olden, J.D., Ormerod, S.J., Smol, J.P., Taylor, W.W., Tockner, K.,

Vermaire, J.C., Dudgeon, D., Cooke, S.J., 2019. Emerging threats and persistent

conservation challenges for freshwater biodiversity. Biol. Rev. 94, 849–873. https://

doi.org/10.1111/brv.12480.

Reynoldson, T.B., Norris, R.H., Resh, V.H., Day, K.E., Rosenberg, D.M., 1997. The

reference condition: a comparison of multimetric and multivariate approaches to

assess water-quality impairment using benthic macroinvertebrates. J. North Am.

Benthol. Soc. 16, 833–852. https://doi.org/10.2307/1468175.

Ruaro, R., Gubiani,



E.A., 2013. A scientometric assessment of 30 years of the Index of

Biotic Integrity in aquatic ecosystems: applications and main ﬂaws. Ecol. Indic. 29,

105–110. https://doi.org/10.1016/j.ecolind.2012.12.016.

Ruaro, R., Gubiani,



E.A., Hughes, R.M., Mormul, R.P., 2020. Global trends and challenges

in multimetric indices of biological condition. Ecol. Indic. 110, 105862. https://

doi.org/10.1016/j.ecolind.2019.105862.

Sano, E.E., Rodrigues, A.A., Martins, E.S., Bettiol, G.M., Bustamante, M.M.C.,

Bezerra, A.S., Couto, A.F., Vasconcelos, V., Schüler, J., Bolfe, E.L., 2019. Cerrado

ecoregions: a spatial framework to assess and prioritize Brazilian savanna

environmental diversity for conservation. J. Environ. Manag. 232, 818–828. https://

doi.org/10.1016/j.jenvman.2018.11.108.

Seaber, P.R., Kapinos, F.P., Knapp, G.L., 1987. Hydrologic Unit Maps. US Geological

Survey, Denver.

Signor, D., Cerri, C.E.P., Conant, R., 2013. N2O emissions due to nitrogen fertilizer

applications in two regions of sugarcane cultivation in Brazil. Environ. Res. Lett. 8.

https://doi.org/10.1088/1748-9326/8/1/015013.

Silva, D.R.O., Herlihy, A.T., Hughes, R.M., Callisto, M., 2017. An improved

macroinvertebrate multimetric index for the assessment of wadeable streams in the

neotropical savanna. Ecol. Indic. 81, 514–525. https://doi.org/10.1016/

j.ecolind.2017.06.017.

Silva, D.R.O., Herlihy, A.T., Hughes, R.M., Macedo, D.R., Callisto, M., 2018. Assessing the

extent and relative risk of aquatic stressors on stream macroinvertebrate assemblages

in the neotropical savanna. Sci. Total Environ. 633, 179–188. https://doi.org/

10.1016/j.scitotenv.2018.03.127.

Silva, D.R.O., Ligeiro, R., Hughes, R.M., Callisto, M., 2016. The role of physical habitat

and sampling effort on estimates of benthic macroinvertebrate taxonomic richness at

basin and site scales. Environ. Monit. Assess. 188, 1–12. https://doi.org/10.1007/

s10661-016-5326-z.

Smith, M.J., Kay, W.R., Edward, D.H.D., Papas, P.J., Richardson, K.S.J., Simpson, J.C.,

Pinder, A.M., Cale, D.J., Horwitz, P.H.J., Davis, J.A., Yung, F.H., Norris, R.H.,

Halse, S.A., 1999. AusRivAS: using macroinvertebrates to assess ecological condition

of rivers in Western Australia. Freshw. Biol. 41, 269–282. https://doi.org/10.1046/

j.1365-2427.1999.00430.x.

Souza, C.M., Shimbo, J.Z., Rosa, M.R., Parente, L.L., Alencar, A.A., Rudorff, B.F.T.,

Hasenack, H., Matsumoto, M.G., Ferreira, L., Souza-Filho, P.W.M., de Oliveira, S.W.,

Rocha, W.F., Fonseca, A.V., Marques, C.B., Diniz, C.G., Costa, D., Monteiro, D.,

Rosa, E.R., V

elez-Martin, E., Weber, E.J., Lenti, F.E.B., Paternost, F.F., Pareyn, F.G.C.,

Siqueira, J.V., Viera, J.L., Neto, L.C.F., Saraiva, M.M., Sales, M.H., Salgado, M.P.G.,

Vasconcelos, R., Galano, S., Mesquita, V.V., Azevedo, T., 2020. Reconstructing three

decades of land use and land cover changes in Brazilian biomes with Landsat archive

and Earth engine. Remote Sens. 12, 2735. https://doi.org/10.3390/rs12172735.

Stoddard, J.L., Herlihy, A.T., Peck, D.V., Hughes, R.M., Whittier, T.R., Tarquinio, E., 2008.

A process for creating multimetric indices for large-scale aquatic surveys. J. North

Am. Benthol. Soc. 27 (1), 878–891. https://doi.org/10.1899/08-053.

Stoddard, J.L., Larsen, D.P., Hawkins, C.P., Johnson, R.K., Norris, R.H., 2006. Setting

expectations for the ecological condition of streams: the concept of reference

condition. Ecol. Appl. 16, 1267–1276. https://doi.org/10.1890/1051-0761(2006)

016[1267:SEFTEC]2.0.CO;2.

Sudaryanti, S., Trihadiningrum, Y., Hart, B.T., Davies, P.E., Humphrey, C., Norris, R.,

Simpson, J., Thurtell, L., 2001. Assessment of the biological health of the brantas

river, east java, Indonesia using the Australian river assessment system (AUSRIVAS)

methodology. Aquat. Ecol. 35, 135–146. https://doi.org/10.1023/A:

1011458520966.

Urbanetz, C., Shimizu, G.H., Lima, M.I.S., 2013. An illustrated angiosperm ﬂora of

cerrado and riparian forest, S~

ao carlos, Brazil. Check List. 9, 275–293.

USEPA, 2016. National Rivers and Streams Assessment 2008-2009: A Collaborative

Survey (EPA/841/R-16/007). U.S. Environmental Protection Agency, Washington,

DC. https://doi.org/10.5754/hge11869.

Vadas, R.L., Hughes, R.M., Bae, Y.J., Baek, M.J., Gonz

ales, O.C.B., Callisto, M.,

Carvalho, D.R. de, Chen, K., Ferreira, M.T., Fierro, P., Harding, J.S., Infante, D.M.,

Kleynhans, C.J., Macedo, D.R., Martins, I., Silva, N.M., Moya, N., Nichols, S.J.,

Pompeu, P.S., Ruaro, R., Silva, D.R.O., Stevenson, R.J., Terra, B. de F., Thirion, C.,

Ticiani, D., Wang, L., Yoder, C.O., 2022. Assemblage-based biomonitoring of

freshwater ecosystem health via multimetric indices: a critical review and

suggestions for improving their applicability. Water Biol. Secur. 1. https://doi.org/

10.1016/j.watbs.2022.100054.

Valente-Neto, F., Martínez, B.T., Hughes, R.M., Ferreira, A., Severo-Neto, F., Souza, F.L.

de, Souza, R.P., Escarpinati, S.C., Roque, F. de O., 2021. Incorporating costs,

thresholds and spatial extents for selecting stream bioindicators in an ecotone

between two Brazilian biodiversity hotspots. Ecol. Indic. 127. https://doi.org/

10.1016/j.ecolind.2021.107761.

Van Sickle, J., Hawkins, C.P., Larsen, D.P., Herlihy, A.T., 2005. A null model for the

expected macroinvertebrate assemblage in streams. J. North Am. Benthol. Soc. 24,

178–191. https://doi.org/10.1899/0887-3593(2005)024<0178:ANMFTE>2.0.CO;

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

Van Sickle, J., Huff, D.D., Hawkins, C.P., 2006. Selecting discriminant function models for

predicting the expected richness of aquatic macroinvertebrates. Freshw. Biol. 51,

359–372. https://doi.org/10.1111/j.1365-2427.2005.01487.x.

Van Sickle, J., Paulsen, S.G., 2008. Assessing the attributable risks, relative risks, and

regional extents of aquatic stressors. J. North Am. Benthol. Soc. 27, 920–931. https://

doi.org/10.1899/07-152.1.

Vitecek, S., Johnson, R.K., Poikane, S., 2021. Assessing the ecological status of european

rivers and lakes using benthic invertebrate communities: a practical catalogue of

metrics and methods. Water (Switzerland) 13. https://doi.org/10.3390/w13030346.

Whittier, T.R., Stoddard, J.L., Larsen, D.P., Herlihy, A.T., 2007. Selecting reference sites

for stream biological assessments: best professional judgment or objective criteria. J.

North Am. Benthol. Soc 26, 349–360. https://doi.org/10.1899/0887-3593(2007)26

[349:SRSFSB]2.0.CO;2.

Wiggins, G., 1996. Larvae of the North American Caddisﬂy Genera (Trichoptera), second

ed. University of Toronto Press, Toronto, Canada.

Wright, J.F., 1995. Development and use of a system for predicting the macroinvertebrate

fauna in ﬂowing waters. Aust. J. Ecol. 20, 181–197. https://doi.org/10.1111/j.1442-

9993.1995.tb00531.x.

Yoder, C.O., Rankin, E.T., Smith, M.A., Alsdorf, B.C., Altfater, D.J., Boucher, C.E.,

Miltner, R.J., Mishne, D.E., Sanders, R.E., Thoma, R.F., 2005. Changes in ﬁsh

assemblage status in Ohio's nonwadeable rivers and streams over two decades. Am.

Fish. Soc. Symp. 2005, 399–429.

P.F. Cordeiro et al. Water Biology and Security xxx (xxxx) xxx

ResearchGate has not been able to resolve any citations for this publication.

Striving for consistency in a national lake assessment: Defining reference status and littoral macroinvertebrate condition in lakes across the conterminous United States

Article

Full-text available

Jan 2025
ECOL INDIC

Biomonitoring for Watershed Protection from a Multiscale Land-Use Perspective

Article

Full-text available

May 2023
Diversity

The types and intensification of land use in the watershed affect the living organisms in aquatic ecosystems differently; this impact will also vary according to temporal and spatial scales. Understanding these interactions is crucial in the design of biomonitoring programs to detect the effect of different pollutants in freshwater ecosystems and improve watershed management and conservation strategies. Therefore, this paper qualitatively reviews biomonitoring studies in freshwater ecosystems to evaluate the impact of different land use types on multiple scales in watersheds. The paper is organized into four sections. The first section presents biomonitoring in different freshwater systems (streams, rivers, lakes, and reservoirs). In the second section, we describe the biomonitoring characteristics of the main land use types. In the third section, we explain how spatial and temporal scales affect biomonitoring. Finally, in the fourth section, we focus on biomonitoring planning and future prediction and discuss how to design biomonitoring programs and how to use models and eDNA in biomonitoring. Our review will assist in decision-making regarding biomonitoring programs in watersheds and will guide future studies on the different bioindicators for various land use types in diverse ecosystems worldwide.

Fish and macroinvertebrate assemblages reveal extensive degradation of the world's rivers

Article

Full-text available

Oct 2022

Fish and macroinvertebrate assemblages reveal extensive degradation of the world's rivers

Article

Full-text available

Sep 2022

Rivers suffer from multiple stressors acting simultaneously on their biota, but the consequences are poorly quantified at the global scale. We evaluated the biological condition of rivers globally, including the largest proportion of countries from the Global South published to date. We gathered macroinvertebrate‐ and fish‐based assessments from 72,275 and 37,676 sites, respectively, from 64 study regions across six continents and 45 nations. Because assessments were based on differing methods, different systems were consolidated into a 3‐class system: Good, Impaired, or Severely Impaired, following common guidelines. The proportion of sites in each class by study area was calculated and each region was assigned a Köppen‐Geiger climate type, Human Footprint score (addressing landscape alterations), Human Development score (addressing social welfare), % rivers with good ambient water quality, % protected freshwater key biodiversity areas; and % of forest area net change rate. We found that 50% of macroinvertebrate sites and 42% of fish sites were in Good condition, whereas 21% and 29% were Severely Impaired, respectively. Poorest biological conditions occurred in Arid and Equatorial climates and the best conditions occurred in Snow climates. Severely Impaired conditions were associated (Pearson correlation coefficient) with higher Human Development Index scores, poorer physico‐chemical water quality, and lower proportions of protected freshwater areas. Good biological conditions were associated with good water quality and increased forested areas. It is essential to implement statutory bioassessment programs in Asian, African and South American countries, and continue them in Oceania, Europe and North America. There is a need to invest in assessments based on fish, as there is less information globally and fish were strong indicators of degradation. Our study highlights a need to increase the extent and number of protected river catchments, preserve and restore natural forested areas in the catchments, treat wastewater discharges, and improve river connectivity.

Urban stream rehabilitation in a densely populated Brazilian metropolis

Article

Full-text available

Sep 2022

Environmental rehabilitation of urban streams has been widely applied in Global North countries, at least since the 1970s, but it is a recent approach in Global South countries. The objective of this paper is to evaluate whether the rehabilitation experience carried out since 2006 in three urban stream sites in the third-largest Brazilian metropolis (c. 5.5 million inhabitants) was and continues to be effective in terms of socio-environmental improvement after 10 years of interventions. These interventions included the rehabilitation of watercourses (e.g., improvement of water quality through the management of sewage and garbage, stabilization of riverbanks, revegetation of riparian zones, riverbed naturalization, removal of riverbank housing). We evaluated water quality, physical habitat structure, and benthic macroinvertebrate assemblages in three test sites in three sampling periods: pre-intervention (2004–2005), early post-intervention (2008–2011) and late post-intervention (2018–2019). Additionally, three reference-stream sites (2018–2019) were assessed to compare the conditions of the three tested sites versus the reference sites. We also assessed citizen perceptions concerning the interventions through questionnaires given to urban stream residents at the three tested sites in early and late post-rehabilitation (215 in 2008, 180 in 2019). The results of water quality monitoring showed a significant improvement in most parameters used to calculate the Water Quality Index (WQI) in the early intervention phase, and WQI scores have improved since. The physical habitat and macroinvertebrate indicators indicated moderate improvements. The residents indicated increased appreciation of the environmental improvements over 10 years. Given the results in Belo Horizonte, we believe that implementation and evaluation of similar projects and programs aimed at rehabilitating urban streams are technically viable using our approaches throughout the Global South.

Current Scenario and Perspectives for Nitrogen Fertilization Strategies on Tropical Perennial Grass Pastures: A Review

Article

Full-text available

Aug 2022

Traditional Nitrogen (N) fertilization practices for tropical perennial grass are still based on annual amounts, following flat N rates instalments. This strategy does not consider variations in climatic conditions along the growing season, their impacts on the plant’s demand and the soil N availability. At regrowth cycles where the amount of soil N released from mineralization or through animal excreta surpass the plant’s demand, most of the N may be lost (as ammonia, nitrate, and nitrous oxide), increasing production costs and environmental pollution. This paper examines current N fertilization and discusses possible gaps in knowledge for the definition of more precise fertilization guidelines in pasture-based livestock systems based on tropical perennial grasses. More precise fertilization practices, based upon site and seasonal-specific recommendations, will substantially contribute to the establishment of best fertilization guidelines. Sustainable approaches can be defined by combining the identification of regrowth cycles where high N rates are required, with enhanced efficiency fertilizers, and/or using grass species with the potential for biological nitrification inhibition. The lack of information on tropical grasses requirements, and soil and climatic factors driving the N fate into the mineralization and immobilization processes and how these factors affect plant’s N demand, still prevents opportunities for tactical applications and the establishment of best management guidelines.

Diatom and Macroinvertebrate assemblages to inform management of Brazilian savanna’s watersheds

Article

Full-text available

Sep 2021
ECOL INDIC

Human activities are increasingly affecting freshwater ecosystems and biodiversity, especially in neotropical regions like the Brazilian savanna. Limited research and data availability have inhibited the development and implementation of systematic bioassessment programs and management guidelines. Identifying drivers of biological assemblages’ composition and ecological thresholds along human disturbance gradients is an important step to protect and recover freshwater ecosystems while avoiding threats to biodiversity, goods and services of value to humans. The objectives of this study were to: 1) assess changes in the composition and density of periphytic diatom and macroinvertebrate assemblages relative to natural and human disturbance gradients in Brazilian savanna’s streams, and 2) identify ecological thresholds for direct and indirect human disturbances and potential indicator taxa that could inform the development of biomonitoring programs in Brazil and other neotropical countries. Samplings were carried out in 52 stretches of streams located in central Brazil during two campaigns throughout 2018. Ordination analyses (NMDS) were applied to identify the main drivers of biological assemblages' composition and analyses of taxa distribution in disturbance gradients (TITAN) were carried out to detect possible thresholds and potential bioindicator taxa. Our results pointed out that the scale of land use in the catchment, treated sewage input, and water quality variables (nitrate, phosphate, and conductivity) were the main drivers of diatom and macroinvertebrate assemblages. Taxa such as Eunotia (diatoms) and some families of the orders Ephemeroptera, Plecoptera, and Trichoptera (EPT, macroinvertebrates) were associated with natural conditions, while Nitzschia palea and Gomphonema (diatoms) and Oligochaeta and Hirudinea (macroinvertebrates) were related to environments with a higher degree of disturbance. Although ecological thresholds along disturbance gradients varied among taxa and biotic groups, our results revealed that relatively minor increases in land use in the riparian zone (>0%) and in the upstream catchment (0–33%) were sufficient to trigger significant changes in macroinvertebrate and diatom assemblages. The limits of tolerance for conductivity, nitrate and phosphate were also low, varying between 7-163 µS.cm⁻¹, 0.3–1.0 mg.L⁻¹ and 0.03–1.5 mg.L⁻¹, respectively. In general, the values we found were more restrictive than those provided by Brazilian government guidelines, suggesting that the latter would be insufficient to maintain the integrity of biological assemblages in Brazilian savanna’s watersheds. We provided valuable knowledge about the sensitivities and tolerances of diatom and macroinvertebrate’s taxa/assemblages that can be especially useful for the proper freshwater and watersheds management.

Incorporating costs, thresholds and spatial extents for selecting stream bioindicators in an ecotone between two Brazilian biodiversity hotspots

Article

Full-text available

Aug 2021
ECOL INDIC

Biomonitoring is critical for characterizing and monitoring status, spatial patterns, and long-term trends in the ecological condition of freshwater ecosystems. The selection of cost-effective bioindicators is a critical step in establishing such monitoring programs. Key indicator considerations are a reliable response to anthropogenic disturbances, a high benefit-cost-ratio and sensitivity at multiple spatial extents. We evaluated non-linear responses of Ephemeroptera, Plecoptera and Trichoptera (EPT) and fish to the effects of native vegetation loss within buffers of 100 m and 1000 m and assessed the sampling and processing costs involved for each assemblage. We sampled 37 neotropical stream sites in the Formoso River network, a karstic region of the Bodoquena Plateau, midwest Brazil, lying in the ecotone between the Cerrado and Atlantic Forest biological hotspots. We used TITAN (threshold indicator taxa analysis) to identify six indicator taxa, four EPT genera and two fish species. The four EPT genera had low negative thresholds to native vegetation loss, whereas the two fish species had positive thresholds. Thresholds were lower for the 100 m buffers than the 1000 m buffers for EPT. The most sensitive taxon (Macronema, Trichoptera) had a threshold of 0% native vegetation loss in the 100 m buffers and nearly 40% in the 1000 m buffers. For taxa richness, we found no non-linear response to the effects of native vegetation loss for buffer extent nor assemblage. The total cost for EPT biomonitoring was US

3,616; whereas for fish, the total was US

1,901. Although fish were less expensive than EPT, they did not respond negatively to native vegetation loss and their positive threshold started at 48%, a level of vegetation loss that was highly disruptive of EPT. Therefore, we do not recommend using fish to monitor the effects of native vegetation loss on headwater streams in the Bodoquena Plateau. Although EPT monitoring costs 52% more than fish assemblage monitoring, it detected earlier impacts of the effects of native vegetation loss on stream biota, especially in the 100 m buffer. Therefore, EPT are more cost-effective early warning indicators for monitoring the effects of native vegetation loss in Bodoquena Plateau headwaters.

Biological assessment of western USA sandy bottom rivers based on modeling historical and current fish and macroinvertebrate data

Article

Jan 2022

Biological monitoring is important for assessing the ecological condition of surface waters. However, there are challenges in determining what constitutes reference conditions, what assemblages should be used as indicators, and how assemblage data should be converted into quantitative indicator scores. In this study, we developed and applied biological condition gradient (BCG) modeling to fish and macroinvertebrate data previously collected from large, sandy bottom southwestern USA rivers. Such rivers are particularly vulnerable to altered flow regimes resulting from dams, water withdrawals and climate change. We found that sensitive ubiquitous taxa for both fish and macroinvertebrates had been replaced by more tolerant taxa, but that the condition assessment ratings based on fish and macroinvertebrate assemblages differed. We conclude that the BCG models based on both macroinvertebrate and fish assemblage condition were useful for classifying the condition of southwestern USA sandy bottom rivers. However, our fish BCG model was slightly more sensitive than the macroinvertebrate model to anthropogenic disturbance, presumably because we had historical fish data, and because fish may be more sensitive to dams and altered flow regimes than are macroinvertebrates.

Chronic urbanization decreases macroinvertebrate resilience to natural disturbances in neotropical streams

Article

Jan 2021

Natural disturbances play important roles in the functioning and structure of lotic ecosystems, especially in small streams. Adaptation to natural disturbances, in the form of resilience, can be affected by anthropogenic disturbances such as urbanization and industrial zones, which in turn can limit stream biodiversity. The aim of this study was to assess the effects of runoff from urban and industrial zones on the resilience of benthic macro-invertebrate assemblages in small streams. For that, we tested the hypothesis that benthic macroinvertebrate assemblages in streams affected by urbanization and industrialization have lower resilience to natural disturbances than those in reference areas. We calculated the recovery proportions of Taxa Richness, Taxa Abundance, Resistant Taxa Richness, Resistant Taxa Abundance, Sensitive Taxa Richness and Sensitive Taxa Abundance. Recovery proportions of freshwater biodiversity were calculated as the target variable values during the dry season divided by the same variable in the previous rainy season. Taxa Richness recovery proportion and Sensitive Taxa Richness recovery proportion were significantly higher (p < 0.01) in the reference sites. Resistant Taxa Richness and Sensitive Taxa Abundance followed the same pattern but were less significant (p < 0.1). These results indicate that streams draining urban and industrial areas have significantly lower resilience to natural disturbances than their counterparts in reference areas. Our results also suggest that both landscape and local environmental conditions play important roles in maintaining naturally resilient lotic ecosystems and biodiversity in the neotropics.

A new predictive model (MINASPACS) for spatially extensive biological assessments in southeastern Brazilian streams

Abstract

Recommended publications

Major risks to aquatic biotic condition in a Neotropical Savanna River basin

Human pressures degrade the ecological condition of the Upper Graipu River

Benthic macroinvertebrate assemblages detect the consequences of a sewage spill: a case study of a S...

Are multiple multimetric indices effective for assessing ecological condition in tropical basins?