Anda di halaman 1dari 6

Ecological Informatics 6 (2011) 303308

Contents lists available at ScienceDirect

Ecological Informatics
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e c o l i n f

Using multi-target clustering trees as a tool to predict biological water quality indices
based on benthic macroinvertebrates and environmental parameters in the
Chaguana watershed (Ecuador)
Luis Dominguez-Granda a, b, Koen Lock a,, Peter L.M. Goethals a
a
b

Ghent University, Laboratory of Environmental Toxicology and Aquatic Ecology, Belgium


Escuela Superior Politcnica del Litoral (ESPOL), Instituto de Ciencias Qumicas y Ambientales (ICQA), Guayaquil, Ecuador

a r t i c l e

i n f o

Article history:
Received 22 December 2009
Received in revised form 23 May 2011
Accepted 24 May 2011
Available online 1 June 2011
Keywords:
Aquatic insects
Biological indices
Multitarget clustering trees
Diversity
Macroinvertebrates
Richness

a b s t r a c t
Macroinvertebrates were sampled in the Chaguana river basin in SW Ecuador in the wet season (March) and
the dry season (September) of 2005 and 2006. To assess the robustness of several biological indicators,
correlations were calculated between both years and between the wet and the dry season. In addition, it was
tested if the indices gave signicantly different results for sites with a bad, poor, moderate and good ecological
water quality. Composition measures performed poorly in most cases, however, abundance, diversity and
richness measures often performed better and tolerance measures, the so-called biotic indices, performed
very well, even indices developed for temperate regions. By using pruned multitarget clustering trees, it was
possible to predict several well-performing ecological water quality indices simultaneously on the basis of the
occurring key macroinvertebrate taxa or, alternatively, on the basis of key environmental variables. In
contrast to unpruned trees, which resulted in complex trees that were difcult to interpret and performed
inferiorly, pruning resulted in transparent trees. Water quality indices scored high when Hydropsychidae
were present and even higher when in addition also Megapodagrionidae were present. When no
Hydropsychidae nor Libellulidae were present, the indices reached the lowest scores. However, this model
based on key taxa occurrences did not perform well during validation. Water quality indices scored higher
with increasing dissolved oxygen concentrations and a strong current velocity. The latter model based on
environmental variables also performed well during validation. In the presented study, the ecological water
quality could thus be accurately predicted solely on the basis of dissolved oxygen concentration and current
velocity. It can therefore be concluded that multitarget clustering trees can be easily used as a practical tool for
cost-effective decision support by water quality managers.
2011 Elsevier B.V. All rights reserved.

1. Introduction
Assessment of river health using biological methods is currently
commonplace in most temperate countries. Several of these methods
have been standardized and included in national and regional
monitoring programs (De Pauw et al., 2006; Hering et al., 2003),
serving as a basis for policy decisions concerning surface water
management. However, this is not the case in most tropical countries,
where physicalchemical methods, some of which require expensive
laboratory analysis, are predominantly used to assess running water
quality. Since most tropical regions consist of developing countries,
their limited technical and nancial resources for environmental
issues constrain the establishment of national monitoring programs
and therefore, cost-effective monitoring programs are needed. After a
Corresponding author at: Ghent University, Laboratory of Environmental Toxicology and Aquatic Ecology, J. Plateaustraat 22, 9000 Ghent, Belgium. Tel.: + 32 9 2643996;
fax: + 32 9 2643766.
E-mail address: Koen.Lock@UGent.be (K. Lock).
1574-9541/$ see front matter 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ecoinf.2011.05.004

process of adaptation, testing and standardization, biotic indices for


macroinvertebrates can be reliable systems for application in river
management of tropical regions.
Several studies have already evaluated the applicability of water
quality assessment methods from temperate regions in neotropical
rivers with satisfactory results (Baptista et al., 2007; Fenoglio et al., 2002;
Jacobsen, 1998; Marques and Barbosa, 2001; Silveira et al., 2005;
Umana-Villalobos and Springer, 2006). In addition, several bioassessment methods have been developed for tropical regions, generally based
on a biotic approach and predominantly adaptations of the English
BMWP (Biological Monitoring Working Party) (Astorga et al., 1997;
Baptista et al., 2007; Chessman, 1995, 2003; Chutter, 1972; Dickens and
Graham, 2002; Junqueira and Campos, 1998; Mustow, 2002; Roldn,
2003; Sharma and Moog, 1998). These methods are usually adapted for
tropical rivers through (1) the inclusion of local taxa and the exclusion of
absent ones and (2) the modication of tolerance value.
A range of techniques has already been applied to model the water
quality of running waters. Articial neural networks (Dedecker et al.,
2007), fuzzy logic (Adriaenssens et al., 2004a; Mouton et al., 2009),

304

L. Dominguez-Granda et al. / Ecological Informatics 6 (2011) 303308

et al., 2009). The latter technique was already used to predict the
presence of several alien macroinvertebrates based on the measured
environmental variables (Everaert et al., in press) and is potentially also
a suitable tool in water quality assessments.
The general objectives of the present study were (1) to evaluate the
suitability of several biological indices for water quality assessment of
the Chaguana watershed in Ecuador and (2) using multi-target
clustering trees to predict several reliable biological indices simultaneously on the basis of the presence of key macroinvertebrate taxa or,
alternatively, based on key environmental variables. In this way, it was
assessed whether multi-target clustering trees could be used as a tool for
cost-effective water quality assessment.

classication trees (Dakou et al., 2007; Dzeroski et al., 2000), Bayesian


belief networks (Adriaenssens et al., 2004b) and support vector
machines (Ambelu et al., 2010; Hoang et al., 2010) have proven to
have a high potential for water quality assessment. However, in the
present study, multi-target clustering trees were used, which have
rarely been used for water quality assessment (Everaert et al., in
press). This technique is applicable for relatively small datasets and
has the advantage that several parameters can be predicted simultaneously. In addition, their transparency makes them suitable as a
practical tool for decision support by water quality managers.
Walley and Dzeroski (1995) already used data mining techniques for
biological river quality assessment, however, Dzeroski et al. (1997) were
among the rst to apply classication trees for river community
analysis: based on biological data, British rivers were classied and
Slovenian rivers were classied based on the inuence of physical and
chemical parameters on selected bioindicator organisms. Simultaneous
predictions of multiple physicalchemical properties were made from
its biological properties using a single decision tree by Blockeel et al.
(1999) and they also predicted past physicalchemical properties of the
river water from its current biological properties. On the basis of
biological data, Dzeroski et al. (2000) predicted physicalchemical
variables: taxa that occurred in many trees were considered as useful
indicator taxa. Habitat suitability for six macroinvertebrate taxa in the
river Axios (Northern Greece) were predicted by Dakou et al. (2007) by
inducing decision trees. Also D'heygere et al. (2003, 2006) predicted the
occurrence of several macroinvertebrate taxa in Flanders based on a
selection of environmental variables. River types for abiotic features,
rife zones, sh assemblages and macroinvertebrate assemblages from
river edges were dened by Turak and Koop (2008) on the basis of
reference sites in New South Wales (Australia). In contrast to the
classical single-target approach, which learns a model for each target
attribute separately, a multi-target approach was recently developed,
which builds one model for all target attributes simultaneously (Kocev

2. Materials and methods


2.1. Study area
The Chaguana river basin is situated in El Oro Province in SW
Ecuador (Fig. 1). The basin drains an area of approximately 32,000 ha,
owing from the occidental slope of the W Andes to a larger watershed
called the Pagua river basin. The Chaguana river basin drains a
mountainous area of difcult access with headwaters located at about
2900 m above sea level. Near natural conditions (humid forests and
brushes, mangroves and uncultivated land) cover 37% of the land, while
activities performed in the basin are shrimp farming, human settlements and in recent years gold mining (Matamoros, 2004).
2.2. Sampling
Within the Chaguana river basin, 29 sampling sites, distributed
among different land use categories, were selected to be surveyed in the
wet season (March) and the dry season (September) of 2005 and 2006.
However, in 2006, access to some stations was denied due to land use

11
10

1
17

12
22

14

13

28
16
20
15

26

24
29 23

27

21

25

19
18
6

9
2
km

Fig. 1. Location of the sampling stations in the Chaguana watershed, with indication of the sites with a high (black), an intermediate (grey) and a low human impact (white).

L. Dominguez-Granda et al. / Ecological Informatics 6 (2011) 303308

conicts. In total, 104 samples were taken: 29 during the dry season of
2005, 29 during the wet season of 2005, 24 during dry season of 2006
and 22 during the wet season of 2006. Macroinvertebrate samples were
always collected by the same operator by means of a standard hand net
consisting of a metal frame holding a conical net (20 30 cm, 300 m
mesh size). Sampling duration was 3 min active sampling in 2005, in
2006 sampling duration was increased to 10 min active sampling.
Organisms were collected from the different habitats present at the
sampling site. Rife habitats were sampled by holding the net
downstream while the operator disturbed the substratum by kicking
directly in front of the net opening. Stream edge habitats were sampled
by vigorously sweeping along the stream margins disturbing bottom
and bank substratum. The objective of the sampling was to collect the
most representative taxa of macroinvertebrates at the site examined.
After separation, macroinvertebrates were identied under a stereomicroscope. The taxonomical knowledge of stream fauna in Ecuadorian
streams is still scarce; therefore aquatic insects were identied at family
level with the available literature containing identication keys and
descriptions of the riverine fauna of the region (Domnguez et al., 1994;
Fernndez and Dominguez, 2001; Roldn, 1988). Non insects were
mostly identied at higher taxonomic levels.
2.3. Indices
This study deals with the performance of biological methods
developed in Western Europe, North, South and Central America,

305

Africa, Asia and Australia for the assessment of the Chaguana river
basin. The Biological Monitoring Working Party (BMWP) (Armitage et
al., 1983), which is the water quality index used in the UK, has its
origin in the Trent Biotic Index, the rst biotic index developed for the
assessment of running water. The BMWP was improved by Walley
and Hawkes (1996, 1997). If the BMWP index is divided by the
number of scoring families present in the taxa list, the result is known
as the Average Score Per Taxon (ASPT) index. The BMWP was adapted
for Colombia (BMWP/Col) by Roldn (2003) and for Costa Rica
(BMWP/CR) by Astorga et al. (1997). The Stream Invertebrate Grade
Number Average Level index (SIGNAL) was developed by Chessman
(1995) for the assessment of organic pollution in running waters of SE
Australia and later adjusted for application in the whole country
(Chessman, 2003). This index can be calculated with and without the
abundance weighing. Also the Nepalese Biotic Score (NEPBIOS)
(Sharma and Moog, 1998) is an adaptation of the BMWP and Mustow
(2002) developed the BMWP for Thailand (BMWP THAI). The South
African Scoring System (SASS) was originally developed by Chutter
(1972) for river quality assessment in South Africa. Several improvements have been made and here, the fth version (SASS5) was
applied (Dickens and Graham, 2002). The Family Biotic Index (FBI)
was developed by Hilsenhoff (1988) for application in the US. The
Iberian BMWP (IBMWP) was adapted for the Iberian Peninsula (AlbaTercedor and Sanchez-Ortega, 1988). When available, latest versions
of these indices were applied aiming to take into account their recent
improvements (e.g. updated tolerance values, inclusion of new taxa).

Table 1
Spearman rank correlations between the years 2005 and 2006 and between the dry and the wet season. The discriminative power (MannWhitney U test) is given for the samples
with a bad, poor, moderate and good ecological quality according to the BMWP-Colombia.
Year

Abundance measures
# Individuals
Diversity measures
Margalef
Shannon
Simpson
Evenness
Richness measures
# Taxa
# EPT taxa
# Ephemeroptera taxa
# Plecoptera taxa
# Trichoptera taxa
# Diptera taxa
Tolerance measures
BMWP
BMWP-ASPT
IBMWP
IBMWP-IASPT
Family Biotic index
BMWP/Col
BMWP-ASPT/Col
BMWP (CR)
BMWP-ASPT (CR)
Signal 2 score (ab.)
Signal 2 score (not ab.)
SASS5
SASS5-ASPT
NEPBIOS
NEPBIOS-ASPT
BMWPTHAI
BMWP-ASPTTHAI
Composition measures
% Hydropsychidae of Trichoptera
% EPT
% Ephemeroptera
% Trichoptera
% Diptera
% Chironomidae

Season

Discriminative power
Badgood

Badmoderate

Poorgood

Badpoor

Poor_moderate

Moderategood

0.19

nc

0.33

***

***

***

**

**

0.55
0.64
0.56
0.15

***
***
***
nc

0.49
0.49
0.31
0.26

***
***
*
nc

***
***
***

***
***
**

***
***
***

***
**

***
**

***
***
**

0.47
0.72
0.60
0.80
0.60
0.55

**
***
***
***
***
***

0.42
0.59
0.53
0.29
0.44
0.56

**
***
***
*
**
***

***
***
***
**
***
***

***
***
***

***
***
**

***
***
***

***
***
*

***
***

***
***
***
*
***
***

***
**

***
***

0.59
0.66
0.56
0.40
0.052
0.64
0.53
0.66
0.74
0.69
0.68
0.68
0.63
0.68
0.66
0.64
0.52

***
***
***
**
nc
***
***
***
***
***
***
***
***
***
***
***
***

0.47
0.53
0.45
0.32
0.00054
0.61
0.39
0.64
0.62
0.64
0.59
0.60
0.50
0.60
0.68
0.57
0.46

***
***
***
*
nc
***
**
***
***
***
***
***
***
***
***
***
***

***
***
***
***

***
***
***
***

***
***
***
***

***
***
***
***

***

***

***
*

***

***
***
***
***
***
***
***
***
***
***
***
***

***
***
***
***
***
***
***
***
***
***
***
***

***
***
***
***
***
**
***

***
***
***
***
**
***
***
***
***
***
***
***

***

0.44
0.59
0.48
0.75
0.046
0.012

*
***
**
***
nc
nc

***
**

***
***
*
***

0.056
0.16
0.23
0.58
0.080
0.096

nc
nc
nc
***
nc
nc

***

p b 0.05 *; p b 0.01 **, p b 0.001 ***, nc = no signicant correlation, ab. = abundance.

***
***
***
***

***

**
***
*
***
***

*
***

***

***
*
***
*

***
**
***
*

**

*
*

***

***

306

L. Dominguez-Granda et al. / Ecological Informatics 6 (2011) 303308

present

vector of class values, instead of storing a single class value like singletarget classication trees do. This means that each component of the
vector is a prediction for one of the target attributes. Multi-target
clustering trees were constructed by top-down induction and the trees
were pruned by only generating clusters with at least 10 instances in
each subset. The stability of the trees was maximised using a 10-fold
cross-validation procedure. Model performance was evaluated based on
Pearson correlation.

Number of EPT-taxa = 5.6


BMWP Colombia = 115
BMWP Costa Rica = 87

Megapodagrionidae
present

absent

Number of EPT-taxa = 5.0


BMWP Colombia = 78
BMWP Costa Rica = 59

present

Number of EPT-taxa = 1.7


BMWP Colombia = 46
BMWP Costa Rica = 30

Hydropsychidae
absent

3. Results

Libellulidae
absent

The total number of individuals was not constant over the years
because the sampling effort was increased during the second year of
sampling, however, the number of individuals already gave a
relatively good idea of the ecological quality (Table 1). The diversity
numbers of Margalef and Shannon performed very well, while the
Simpson index performed less good and the evenness performed very
badly: results varied depending on the year and the season and it had
a very low discriminative power. Most richness measures, especially
the number of EPT taxa, were good indicators for the water quality:
constant results were obtained over the years and the seasons and
discriminative power was high. Only the number of Plecoptera taxa
did not perform that good. With one exception, the biotic measures
performed very good: results were hardly affected by the year or the
season of sampling and the discriminative power was usually high.
However, the Family Biotic Index varied between the years and the
seasons and only the sites with a high and a low human impact could
be separated. The discriminative power of the indices was usually
higher than the average score per taxon (ASPT) variant of the
respective indices. The composition measures were poor indicators,
only the fraction Trichoptera performed relatively well.
Using a multi-target clustering tree, three ecological water quality
indices with a good performance (number of EPT-taxa, BMWPColombia and BMWP-Costa-Rica) were predicted based on the
occurring macroinvertebrate taxa (Pearson R = 0.79) (Fig. 2). If
Hydropsychidae were present, water quality was better. When in
addition also Megapodagrionidae were present, the highest water
quality scores were obtained. However, when no Hydropsychidae nor
Libellulidae were present, the lowest scores were predicted. However,
when the model developed for the samples of 2005 was validated
with the samples of 2006, the ecological water quality could not be
predicted accurately (Pearson R = 0.21).
The ecological water quality indices could also be predicted on the
basis of the environmental variables (Pearson R = 0.73) (Fig. 3). When
the dissolved oxygen concentration was lower than 6.83 mg l 1, the
lowest ecological water quality scores were found. When the dissolved

Number of EPT-taxa = 1.6


BMWP Colombia = 25
BMWP Costa Rica = 19

Fig. 2. Multi-target clustering tree which predicted the number of EPT-taxa, the BMWPColombia and the BMWP-Costa Rica based on the occurrence of macroinvertebrate
taxa.

Apart from biotic indices, also several diversity indices were calculated:
the Margalef (1951) index, the Simpson (1949) index, the Shannon
Wiener index (Shannon and Weaver, 1963) and the Shannon evenness
index, which is calculated by dividing the ShannonWiener index by the
natural logarithm of the number of taxa.
2.4. Statistics
To check the robustness of the biological indicators, Spearman
rank correlations were applied between the years 2005 and 2006 and
between the dry and the wet season. As the BMWP-Colombia
included the highest number of taxa that were present in the
Chaguana watershed and because Colombia is the country which is
closest to Ecuador of all countries which developed biological indices,
the performance of all indices was evaluated based on the outcome of
the BMWP-Colombia. In addition, the BMWP-Colombia best reected
the three classes of human impact, which were separated based on the
macroinvertebrate community composition using multivariate analysis (Fig. 1) (Dominguez-Granda et al., in press). Four water quality
classes were recognised based on outcome of the BMWP-Colombia:
bad (040), poor (4170), moderate (71100) and good (N100). The
MannWhitney U Test was used to identify signicant differences
between sites with different water quality classes.
Based on a training set consisting of the 58 samples taken in 2005,
multi-target clustering trees were built using CLUS (Blockeel and Struyf,
2002). A test set consisting of the 46 samples taken in 2006 was used for
model validation. The leaves of a multi-target classication tree store a

Dissolved oxygen
6.83 mg.l-1
Number of EPT-taxa = 2.0
BMWP Colombia = 35
BMWP Costa Rica = 24

>6.83 mg.l-1
Current velocity
32 cm.s-1

Number of EPT-taxa = 2.9


BMWP Colombia = 68
BMWP Costa Rica = 49

>32 cm.s-1
Dissolved oxygen
7.62 mg.l-1

Number of EPT-taxa = 5.1


BMWP Colombia = 81
BMWP Costa Rica = 61

>7.62 mg.l-1
Number of EPT-taxa = 6.3
BMWP Colombia = 122
BMWP Costa Rica = 93

Fig. 3. Multi-target clustering tree which predicted the number of EPT-taxa, the BMWP-Colombia and the BMWP-Costa Rica based on environmental variables.

L. Dominguez-Granda et al. / Ecological Informatics 6 (2011) 303308

oxygen concentration was higher, water quality further improved with


a current velocity higher than 32 cm s 1 and when the dissolved
oxygen concentration was higher than 7.62 mg l 1, the highest scores
were obtained. In contrast to trees based on taxa composition, the trees
based on environmental variables performed well during validation
(Pearson R = 0.66).
4. Discussion
With the exception of the Family Biotic Index (Hilsenhoff, 1988),
all biotic measures performed quite well. This could be expected
because these measures were especially developed for this purpose.
Not only the methods developed for Colombia (Roldn, 2003) and
Costa Rica (Astorga et al., 1997), but even the methods that were
developed for temperate regions (Alba-Tercedor and Sanchez-Ortega,
1988; Armitage et al., 1983; Walley and Hawkes, 1996, 1997) and
Thailand (Mustow, 2002), South Africa (Chutter, 1972; Dickens and
Graham, 2002), Australia (Chessman, 1995, 2003) and Nepal (Sharma
and Moog, 1998), gave good results in the Chaguana watershed.
Notwithstanding the fact that most biotic measures performed quite
well, it is preferable to use an index of an area with a similar
macrobenthic fauna. Since Colombia and Costa Rica had the macrobenthic fauna that most resembled that of the Chaguana river basin in
Ecuador, these indices are preferable until Ecuador develops its own
index. The biotic indices performed usually better than the average
score per taxon (ASPT) variant of the respective indices: not taking
the number of occurring taxa into account thus seemed to decrease
the performance of the biotic indices.
The reduction of the dimensionality of a clustering tree contributes
to a more easy interpretation of the revealed trends in the data,
focusing the attention on the important variables (Dzeroski et al.,
1997). Optimal pruning is an important mechanism as it improves the
transparency of the induced trees by reducing their size and enhances
the accuracy by eliminating errors that are present due to noise in the
data (Dakou et al., 2007). In addition, the generalisation capacity of
complex trees is usually corrupted because these trees can be
overtted to the used dataset. Therefore, the tree size in the present
study was reduced by using a large minimum group size for division.
In this way, small trees were obtained which are easy to interpret and
which can be used for decision support by water quality managers.
Based on the occurring taxa and the environmental variables, multitarget clustering trees could be induced that were able to predict several
ecological water quality indices simultaneously. The performance for
trees based on the occurring taxa and the environmental variables was
similar for the models developed based on the samples of 2005.
However, validation with the samples of 2006 indicated that trees based
on taxa composition did not result in accurate predictions. Although
sampling effort was higher in 2006 (10 min active sampling) than in
2005 (3 min active sampling), it is unlikely that sampling effort caused
the poor performance of the clustering tree based on taxa composition
since the indicator taxa were usually present at high densities if they
were present. It is more probable that taxa composition of macroinvertebrate communities is to variable to make decisions based on the
occurrence of a single taxon. However, multi-target clustering trees
based on environmental variables performed much better during
validation and their predictive performance would probably further
improve when additional variables such as nutrient concentrations
could be incorporated as well.
5. Conclusions
The presented results indicate that indices developed for other, even
remote, countries can already give a good idea of the ecological water
quality. However, for application in routine water quality assessments, it
is advised that region-specic methods are developed. With some minor
modications to adapt the existing indices to the local fauna, it should

307

therefore be possible to develop a biotic index for macroinvertebrates


for water quality assessment in Ecuador. Using multi-target clustering
trees, it was possible to predict the ecological water quality on the basis
of the occurring taxa as well as on the basis of the environmental
parameters. However, validation of these models indicated that models
based on occurrences did not perform well, whereas models based on
environmental variables were more reliable and the inclusion of
additional parameters could even improve their predictive power.
Multi-target clustering trees can therefore be a valuable tool for costeffective decision support in water quality assessment.
Acknowledgments
The rst author is grateful for the nancial support of the VLIRESPOL IUC programme in Ecuador and SENACYT. In particular, we
would like to thank Magda Vincx, coordinator of the VLIR-ESPOL IUC
programme, Nancy Fockedey, pioneer of this programme in Ecuador
and Pilar Cornejo, director of this program in Ecuador. We would also
like to thank Galo, Erick, Christian L., Christian R., Christian V., Felix
and Santiago for the help during the eld work. Koen Lock is currently
supported by a post-doctoral fellowship from the Fund for Scientic
Research (FWO-Vlaanderen, Belgium).
References
Adriaenssens, V., De Baets, B., Goethals, P.L.M., De Pauw, N., 2004a. Fuzzy rule-based
models for decision support in ecosystem management. Sci. Total. Environ. 319,
112.
Adriaenssens, V., Goethals, P.L.M., Charles, J., De Pauw, N., 2004b. Application of
Bayesian Belief Networks for the prediction of macroinvertebrate taxa in rivers.
Ann. Limnol.- Int. J. Limnol. 40, 181191.
Alba-Tercedor, J., Sanchez-Ortega, A., 1988. Un mtodo rpido y simple para evaluar la
calidad biolgica de las aguas corrientes basado en el de Helawell (1978). Limnetica
4, 5156.
Ambelu, A., Lock, K., Goethals, P.L.M., 2010. Comparison of modeling techniques to
predict macroinvertebrate community composition in rivers of Ethiopia. Ecol.
Inform. 5, 147152.
Armitage, P.D., Moss, D., Wright, J.F., Furse, M.T., 1983. The performance of a new
biological water quality score system based on macroinvertebrates over a wide
range of unpolluted running-water sites. Wat. Res. 17, 333347.
Astorga, Y., De Pauw, N., Persoone, G., 1997. Development and application of cost-effective
methods for biological monitoring of rivers in Costa Rica. European Community. Final
report, Joint research European Union Project No NCI1* CT-92-0094.
Baptista, D.F., Buss, D.F., Egler, M., Giovanelli, A., Silveira, M.P., Nessimian, J.L., 2007. A
multimetric index based on benthic macroinvertebrates for evaluation of Atlantic
forest streams at Rio de Janeiro State, Brazil. Hydrobiologia 575, 8394.
Blockeel, H., Struyf, J., 2002. Efcient algorithms for decision tree cross-validation.
J. Mach. Learn. Res. 3, 621650.
Blockeel, H., Dzeroski, S., Grbovic, J., 1999. Simultaneous prediction of multiple
chemical parameters of river water quality with TILDE. Lect. Notes Artif. Intell.
1704, 3240.
Chessman, B.C., 1995. Rapid assessment of rivers using macroinvertebrates: a
procedure based on habitat-specic sampling, family level identication and a
biotic index. Aust. J. Ecol. 20, 122129.
Chessman, B.C., 2003. SIGNAL 2 a scoring system for macroinvertebrate (water
bugs) in Australian rivers. Monitoring River Heath Initiative Technical Report no
31Commonwealth of Australia, Canberra. 32 pp.
Chutter, F.M., 1972. An empirical biotic index of the quality of water in South African
streams and rivers. Wat. Res. 6, 1930.
D'heygere, T., Goethals, P.L.M., De Pauw, N., 2003. Use of genetic algorithms to select
input variables in decision tree models for the prediction of benthic macroinvertebrates. Ecol. Model. 160, 291300.
D'heygere, T., Goethals, P.L.M., De Pauw, N., 2006. Genetic algorithms for optimisation
of predictive ecosystems models based on decision trees and neural networks. Ecol.
Model. 195, 2029.
Dakou, E., D'heygere, T., Dedecker, A.P., Goethals, P.L.M., Lazaridou-Dimitriadou, M., De
Pauw, N., 2007. Decision tree models for prediction of macroinvertebrate taxa in
the river Axios (Northern Greece). Aquat. Ecol. 41, 399411.
De Pauw, N., Gabriels, W., Goethals, P.L.M., 2006. River monitoring and assessment
methods based on macroinvertebrates. In: Ziglio, G., Siligardi, M., Flaim, G. (Eds.),
Biological Monitoring of Rivers: Applications and Perspectives. John Wiley & Sons,
Chichester, pp. 113134.
Dedecker, A., Van Melckebeke, K., Goethals, P.L.M., De Pauw, N., 2007. Development of
migration models for macroinvertebrates in the Zwalm river basin (Flanders,
Belgium) as tools for restoration management. Ecol. Model. 203, 7286.
Dickens, C.W.S., Graham, P.M., 2002. The South African Scoring System (SASS) version 5
rapid bioassessment method for rivers. Afr. J. Aquat. Sci. 27, 110.
Domnguez, E., Hubbard, M.D., Pescador, M.L., 1994. Los Ephemeroptera en Argentina.
Fauna de Agua Dulce de la Republica Argentina 33, 1142.

308

L. Dominguez-Granda et al. / Ecological Informatics 6 (2011) 303308

Dominguez-Granda, L., Lock, K., Goethals, P.L.M., in press. Application of classication


trees to determine biological and chemical indicators for river assessment: casestudy in the Chaguana watershed (Ecuador). J. Hydroinform.
Dzeroski, S., Grbovic, J., Walley, W.J., 1997. Machine learning applications in biological
classication of river water quality. In: Michalski, R.S., Bratko, I., Kubat, M. (Eds.),
Machine Learning and Data Mining: Methods and Applications. John Wiley and
Sons Ltd., New York, pp. 429448.
Dzeroski, S., Demsar, D., Grbovic, J., 2000. Predicting chemical parameters of river water
quality from bioindicator data. Appl. Intell. 13, 717.
Everaert, G., Boets, P., Lock, K., Goethals, P.L.M., in press. Application of decision trees to
analyze the ecological impact of invasive species in polder lakes in Flanders,
Belgium. Ecol. Model. doi:10.1016/j.ecolmodel.2010.08.013.
Fenoglio, S., Badino, G., Bona, F., 2002. Benthic macroinvertebrate communities as
indicators of river environment quality: an experience in Nicaragua. Rev. Biol. Trop.
50, 11251131.
Fernndez, H.R., Dominguez, E., 2001. Gua para la determinacin de los artrpodos
bentnicos sudamericanos. Universidad Nacional de Tucumn, Tucumn. 282 pp.
Hering, D., Buffagni, A., Moog, O., Sandin, L., Sommerhuser, M., Stubauer, I., Feld, C., Johnson,
R., Pinto, P., Skoulikidis, N., Verdonschot, P., Zahrdkov, S., 2003. The development of a
system to assess the ecological quality of streams based on macroinvertebrates design
of the sampling programme within the AQEM project. Int. Rev. Hydrobiol. 88, 345361.
Hilsenhoff, W.L., 1988. Rapid eld assessment of organic pollution with a family-level
biotic index. J. N. Amer. Benth. Soc. 7, 6568.
Hoang, T.H., Lock, K., Mouton, A., Goethals, P.L.M., 2010. Application of decision trees
and support vector machines to model the presence of macroinvertebrates in rivers
in Vietnam. Ecol. Inform. 5, 140146.
Jacobsen, D., 1998. The effect of organic pollution on the macroinvertebrate fauna of
Ecuadorian highland streams. Arch. Hydrobiol. 143, 179195.
Junqueira, V.M., Campos, S.C.M., 1998. Adaptation of the BMWP method for water
quality evaluation to Rio das Velhas watershed (Minas Gerais, Brazil). Acta Limnol.
Bras. 10, 123135.
Kocev, D., Dzeroski, S., White, M.D., Newell, G.R., Griffoens, P., 2009. Using single and
multi-target regression trees and ensembles to model a compound index of
vegetation condition. Ecol. Model. 220, 11591168.
Margalef, R., 1951. Diversidad de especies en las comunidades naturales. Publ. Inst. Biol.
Appl. Barc. 6, 5972.
Marques, M., Barbosa, F., 2001. Biological quality of waters from an impacted tropical
watershed (middle Rio Doce basin, southeast Brazil), using benthic macroinvertebrate communities as an indicator. Hydrobiologia 457, 6976.

Matamoros, D., 2004. Predicting river concentrations of pesticides from banana


plantations under data-poor condition. Ghent University, Ghent. 204 pp.
Mouton, A.M., De Baets, B., Goethals, P.L.M., 2009. Knowledge-based versus data-driven
fuzzy habitat suitability models for river management. Environ. Modell. Softw. 24,
982993.
Mustow, S.E., 2002. Biological monitoring of rivers in Thailand: use and adaptation of
the BMWP score. Hydrobiologia 479, 191229.
Roldn, G., 1988. Gua para el estudio de los macroinvertebrados acuticos del
Departamento de Antioquia. Fondo FEN Colombia, Conciencias-Universidad de
Antioquia, Santaf de Bogota (Colombia).
Roldn, G., 2003. Bioindicacin de la calidad del agua en Colombia. Uso del mtodo
BMWP/Col. Universidad de Antioquia, Medelln, Medelln (Colombia). 182 pp.
Shannon, C.E., Weaver, W., 1963. The Mathematical Theory of Communication.
University of Illinois Press, Urbana, Urbana (Illinois). 117 pp.
Sharma, S., Moog, O., 1998. The applicability of biotic indices and scores in water quality
assessment of Nepalese rivers. In: Chalise, S.R., Herrmann, A., Khanal, N.R., Lang, H.,
Molnar, L., Pokhrel, A.P. (Eds.), Ecohydrology of High Mountain Areas, Proceedings
of the International Conference on Ecohydrology of High Mountain Areas. ICIMOD,
UNESCO, Kathmandu. 680 pp.
Silveira, M.P., Baptista, D.F., Buss, D.F., Nessimian, J.L., Egler, M., 2005. Application of
biological measures for stream integrity assessment in South-East Brazil. Environ.
Monit. Assess. 101, 117128.
Simpson, E.H., 1949. Measurement of diversity. Nature 163 688688.
Turak, E., Koop, K., 2008. Multi-attribute ecological river typology for assessing
ecological condition and conservation planning. Hydrobiologia 603, 83104.
Umana-Villalobos, G., Springer, M., 2006. Environmental variation in the Grande de
Terraba river and some of its tributaries, south Pacic of Costa Rica. Rev. Biol. Trop.
54, 265272.
Walley, W.J., Dzeroski, S., 1995. Biological monitoring: a comparison between Bayesian,
neural and machine learning methods of water quality classication. In: Denzer, R.,
Shimak, G., Russell, D. (Eds.), Environmental Software Systems. Chapman & Hall,
London, pp. 229240.
Walley, W.J., Hawkes, H.A., 1996. A computer-based reappraisal of Biological
Monitoring Working Party scores using data from the 1990 River Quality Survey
of England and Wales. Wat. Res. 30, 20862094.
Walley, W.J., Hawkes, H.A., 1997. A computer-based development of the Biological
Monitoring Working Party score system incorporating abundance rating, biotope
type and indicator value. Wat. Res. 31, 201210.

Anda mungkin juga menyukai