Use of the water quality index and dissolved oxygen deficit as simple indicators of watersheds pollution

This article is also available online at:www.elsevier.com/locate/ecolind

Ecological Indicators 7 (2007) 315–328

Use of the water quality index and dissolved oxygen

deficit as simple indicators of watersheds pollution

Enrique Sanchez a, Manuel F. Colmenarejo a, Juan Vicente b, Angel Rubio b,Marıa G. Garcıa a, Lissette Travieso c, Rafael Borja c,*

a Centro de Ciencias Medioambientales (CSIC), C/Serrano, 115-Duplicado, 28006 Madrid, Spainb Ayuntamiento of Las Rozas, Madrid, Spain

c Instituto de la Grasa (CSIC), Avda Padre Garcıa Tejero 4, E-41012 Sevilla, Spain

Received 21 November 2005; received in revised form 16 February 2006; accepted 21 February 2006

Abstract

The use of the water quality index (WQI) and the dissolved oxygen deficit (D) as simple indicators of the watersheds

pollution was investigated and compared in the Municipality of Las Rozas (north-west of Madrid, Spain). The quality of the

water in Guadarrama and Manzanares rivers and Paris Park ponds, the main watersheds of this area was investigated during 2

years (from September 2001 to September 2003). It was found that the WQI was very useful for the classification of the waters

monitored. The WQI was 70, which corresponds to ‘‘good’’ quality water at the sampling point 1 (entrance of Las Rozas) and

decreased to around 64 (medium quality) at the sampling point 6 (outlet of Las Rozas) in the case of Guadarrama River. The

WQI was around 65 in the influents of Manzanares River. Finally, in Paris Park the WQI ranged from around 72–55, which

corresponded to a classification from ‘‘good’’ to ‘‘medium’’ quality, respectively. A high linear relationship between the WQI

and the dissolved oxygen deficit (D) was found. Therefore, a fast determination of WQI may be carried out knowing the values of

D, which are easily obtainable by field measurements. It was found an influence of the climate conditions on the values of WQI

and D.

# 2006 Elsevier Ltd. All rights reserved.

Keywords: Water quality index (WQI); Dissolved oxygen deficit (D); Watersheds pollution

1. Introduction

Different regions of the world are faced with diffe-

rent types of problems associated with the occurrence,

* Corresponding author. Tel.: +34 95 4689654;

fax: +34 95 4691262.

E-mail address: [email protected] (R. Borja).

1470-160X/$ – see front matter # 2006 Elsevier Ltd. All rights reserved

doi:10.1016/j.ecolind.2006.02.005

use and control of water resources, which may endanger

the sustainable development of these resources. The

quality of surface waters is a very sensitive issue.

Anthropogenic influences as well as natural processes

degrade surface waters and impair their use for

drinking, industry, agriculture, recreation and other

purposes (Carpenter et al., 1998; Jarvie et al., 1998;

Simeonov et al., 2003). Due to the spatial and temporal

.

mailto:[email protected]

http://dx.doi.org/10.1016/j.ecolind.2006.02.005

E. Sanchez et al. / Ecological Indicators 7 (2007) 315–328316

variations in water chemistry, a monitoring programme

and a representative and reliable estimation of the

quality of surface waters are necessary (Bollinger et al.,

1999). The water quality index (WQI) has been

considered to give a criteria for surface water

classification based on the use of standard parameters

for water characterization (Couillard and Lefebvre,

1985; House and Newsome, 1989; Smith, 1989;

Melloul and Collin, 1998; Nives, 1999; Pesce and

Wunderlin, 2000; Swamee and Tyagi, 2000; Bordalo

et al., 2001; Cude, 2001; Nagel, 2001; Jonnalagadda

and Mhere, 2001; Liou et al., 2003; Hernandez-Romero

et al., 2004). This index is a mathematical instrument

used to transform large quantities of water character-

ization data into a single number, which represents the

water quality level. The use of WQI is a simple

practice, which allows adequate classification of

water quality. The determination of WQI requires a

normalization step where each parameter is trans-

formed into a 0–100 scale, where 100 represents the

maximum quality. The next step is to apply a

weighting factor in accordance with the importance

of the parameter as an indicator of water quality

(Nives, 1999; Pesce and Wunderlin, 2000; Jonnala-

gadda and Mhere, 2001).

Dissolved oxygen (DO) and dissolved oxygen

deficit (D) are parameters frequently used to evaluate

the water quality on different reservoirs and water-

sheds. These parameters are strongly influenced by a

combination of physical, chemical, and biological

characteristics of streams of oxygen demanding

substances, including algal biomass, dissolved organic

matter, ammonia, volatile suspended solids, and

sediment oxygen demand (Spanou and Chen, 1999;

Cox, 2003; Mullholand et al., 2005; Quinn et al.,

2005). Williams et al. (2000) studied the water quality

variation in three rivers of United Kingdom. The

authors established an empirical equation between the

oxygen deficit variation, the average photosynthesis

rate and the average respiration rate. The use of

dissolved oxygen content as an index of water quality

was also used to estimate the effect of industrial and

municipal effluents on the waters of San Vicente Bay,

Chile (Rudolf et al., 2002). The results suggested that

the oxygen depletion was a representative parameter

for establishing a relative scale of water quality in

these waters. A method based on the maximum and

minimum dissolved oxygen (DO) deficits was derived

to estimate metabolism rates (photosynthesis and

respiration) in streams. This method was applied to

DO concentrations that were measured in two creeks

located in urbanized and agricultural watersheds,

respectively (Wang et al., 2003). An oxygen equiva-

lent model for water quality dynamics was applied

in a macrophyte dominated river (Park et al., 2003).

The model simulated seven coupled state variables:

BOD5, DO, organic nitrogen, ammonia nitrogen,

nitrite/nitrate nitrogen, total organic phosphorus and

dissolved inorganic phosphorus.

The subject of the present work was the use of

the water quality index (WQI) and the dissolved

oxygen deficit (D) as indicators of the environmental

quality of watersheds. As a particular case the main

surface watersheds located in Las Rozas, Madrid

(Spain) were monitored for a period of 2 years

(September 2001 to September 2003). For the deter-

mination of the WQI, European Standards (EU,

1975) for clean water were used as reference in each

case. Finally, the study of the influence of the climate

condition on the water quality was other objective

of the present paper.

2. Materials and methods

2.1. Description of the watersheds investigated

Las Rozas, Madrid (Spain) is a town located in

the north-west of Madrid with a total surface area

of around 59 km2, the urban area corresponds to 44%

of the total area. The total population is 56,000, 77%

of which lives in the urban area. The main watersheds

are distributed in the rivers of Guadarrama and

Manzanares and their influents and in the ponds

of Paris Park. These resources are also of great

importance for the city of Madrid.

Fig. 1 shows the three watersheds monitored:

Guadarrama River, located on the west border and

made up of the Guadarrama River and the influent

creeks (La Torre, La Virgen and Fuentecillas),

Manzanares River on the east border and made up

of the La Trofa and Barrancohondo creeks and the

Paris Park located in the south, a recreational area with

two connected lakes called Superior and Inferior. The

water is recycled from the Inferior Lake to the

Superior Lake by pumping.

E. Sanchez et al. / Ecological Indicators 7 (2007) 315–328 317

Fig. 1. Map of the area monitored at scale 1:100,000 (MWWTP: location of municipal wastewater treatment plants).


2.2. Procedure for watershed sampling

Sampling of the watersheds was carried during

2 years (from September 2001 to September 2003),

Table 1

Description and location of the sampling points monitored

Watershed Sampling point

1 2 3

Guadarrama

Guadarrama River

Name of the

sample point

Guadarrama I Guadarrama II Guadar

Distance (km) 0 1.75 2.43

La Torre creek

Sampling point 1 2 3

Name of the

sample point

La Torre I La Torre II La Tor


Motilona creek


Name of the

sample point

Motilona I Motilona II Motilo


La Virgen creek


Name of the

sample point

La Virgen I Bridge Channe


Fuentecillas creek


Name of the

sample point

Chopera channel Fuentecillas II Fuente


Manzanares

La Trofa creek

Sampling point 1

Name of the

sample point

La Trofa

Distance (km) 1.50

Barrancohondo creek

Sampling point 1

Name of the

sample point

Barrancohondo

Distance (km) 0.70

Paris Park

Paris Park


Name of the

sample point

Wharf Lake superior Estuary

covering all seasons. The samples were taken every

2 weeks and after the determination of field para-

meters they were transported to the laboratory. A

detailed description of the sampling points is

4 5 6

rama III Retamar Guadarrama IV Guadarrama V

3.36 4.45 6.00

4

re III La Tore IV

4.61

4

na III Motilona IV

2.0

l

cillas III

4

Lake inferior


Table 2

Temperatures and rainfalls during the period evaluated (September 2001–September 2003)

Season Temperature variationa (8C) Rainfalla (mm)

TMin TMedium TMax Monthly Total

Autumn 4.5 � 1.1 9.6 � 4.2 14.4 � 4.9 65.7 � 8.5 197.0 � 25.5

Winter 1.8 � 0.5 5.1 � 3.2 9.6 � 3.9 66.3 � 9.4 199.0 � 28.2

Spring 7.2 � 3.5 13.4 � 3.3 19.8 � 5.2 56.0 � 7.3 168.0 � 21.9

Summer 14.8 � 4.3 22.8 � 2.8 28.4 � 5.2 15.0 � 6.3 45.0 � 18.9

a Mean values � S.E. of 56 samples.

summarized in Table 1. Sample point 1 of Guadarrama

River corresponded to the inlet to ‘‘Las Rozas’’

municipality.

2.3. Climate conditions

The sampling started at the beginning of September

2001 and finished at the end of the September 2003.

Table 2 shows the average values of temperature and

rainfall for each season during this period.

2.4. Field determinations and laboratory analyses

Field determinations of pH, conductivity (K),

temperature (T 8C) and dissolved oxygen (DO) were

carried out using portable equipments according to

the Standard Methods for the Examination of Water

and Wastewaters (APHA, 1999). The conductivity, pH

and DO were measured using ‘‘Hanna’’, ‘‘Crison’’ and

‘‘Inolab WTW’’ portable equipments, respectively.

Table 3

Values of Cia and Pi for different parameters of water quality

Parameter Pi Ci

100 90 80 70 60

Range of analytical value

pH 1 7 7–8 7–8.5 7–9 6.5–7

Kb 2 <0.75 <1.00 <1.25 <1.50 <2.00

TSS 4 <20 <40 <60 <80 <100

Amm. 3 <0.01 <0.05 <0.10 <0.20 <0.30

NO2� 2 <0.005 <0.01 <0.03 <0.05 <0.10

NO3� 2 <0.5 <2.0 <4.0 <6.0 <8.0

PT 1 <0.2 <1.6 <3.2 <6.4 <9.6

COD 3 <5 <10 <20 <30 <40

BOD5 3 <0.5 <2.0 <3 <4 <5

DO 4 �7.5 >7.0 >6.5 >6.0 >5.0

T 1 21/16 22/15 24/14 26/12 28/10

a All values, except pH, in mg/l.b Conductivity in mS/cm.

Laboratory analyses were carried out for the

determination of total suspended solids (TSS),

ammonia, nitrite, nitrate, total phosphorus, chemical

oxygen demand (COD) and biochemical oxygen

demand (BOD5). These analyses were also performed

using the methodology recommended by the Standard

Methods (1999). The oxygen deficit (D) was

determined from the difference between the dissolved

oxygen concentrations measured with the portable

dissolved oxygen meter in the corresponding sampling

point (C) and the saturation concentration of pure

water at a similar temperature and pressure (CS). All

these determinations were carried out in triplicate

samples, and the results expressed as averages.

For the determination of the water quality index of

the different watersheds studied, the following empiri-

cal equation was used (Pesce and Wunderlin, 2000):

WQI ¼ k

Pi CiPiP

i Pi(1)

50 40 30 20 10 0

6–9.5 5–10 4–11 3–12 2–13 1–14

<2.50 <3.00 <5.00 <8.00 <12.00 >12.00

<120 <160 <240 <320 <400 >400

<0.40 <0.50 <0.75 <1.00 <1.25 >1.25

<0.15 <0.20 <0.25 <0.50 <1.00 >1.00

<10.0 <15.0 <20.0 <50.0 <100.0 >100.0

<16.0 <32.0 <64.0 <96.0 <160.0 >160.0

<50 <60 <80 <100 <150 >150

<6 <8 <10 <12 <15 >15

>4.0 >3.5 >3.0 >2.0 >1.0 <1.0

30/5 32/0 36/–2 40/�4 45/�6 >45/<�6


Table 4

Water characteristics of Guadarrama Rivera

Sampling point

1 2 3 4 5 6

pH 7.07 � 0.07 7.00 � 0.08 7.01 � 0.08 7.0 � 0.03 7.01 � 0.07 6.88 � 0.09

K (mS/cm) 0.40 � 0.01 0.40 � 0.01 0.39 � 0.01 0.39 � 0.01 0.41 � 0.01 0.41 � 0.01

TSS (mg/l) 5.7 � 0.5 35.1 � 13.9 9.5 � 1.3 12.2 � 1.5 13.2 � 0.1 15.2 � 1.8

Amm. (mg/l) 1.9 � 0.2 1.2 � 0.2 1.0 � 0.1 1.1 � 0.1 1.2 � 0.2 1.2 � 0.2

NO2� (mg/l) 0.21 � 0.02 0.17 � 0.01 0.16 � 0.01 0.19 � 0.01 0.17 � 0.01 0.17 � 0.01

NO3� (mg/l) 15.9 � 1.5 15.1 � 1.2 15.3 � 1.3 11.2 � 0.6 15.2 � 1.3 14.8 � 1.2

PT (mg/l) 2.6 � 0.2 2.4 � 0.1 2.1 � 0.1 2.2 � 0.1 1.9 � 0.1 2.2 � 0.1

COD (mg/l) 14.6 � 1.3 16.3 � 1.6 15.7 � 1.8 12.5 � 0.9 16.1 � 1.5 14.3 � 1.5

BOD5 (mg/l) 6.0 � 0.8 5.8 � 0.9 6.1 � 1.0 3.8 � 0.5 4.8 � 0.7 4.7 � 1.0

DO (mg/l) 9.9 � 0.2 6.2 � 0.3 5.6 � 0.2 5.7 � 0.3 5.7 � 0.2 5.8 � 0.2

D (mg/l) 3.9 � 0.1 4.3 � 0.3 4.9 � 0.2 4.7 � 0.2 4.8 � 0.2 4.9 � 0.2

T (8C) 13.0 � 0.6 13.6 � 0.6 13.7 � 0.7 13.8 � 0.7 13.1 � 0.7 13.0 � 0.7


where k is a subjective constant with a maximum

value of 1 for apparently good quality water and

0.25 for apparently highly polluted water, Ci is the

normalized value of the parameter and Pi is the

relative weight assigned to each parameter. In this

work, such as in other studies reported in literature,

the constant k was not considered in order not to

introduce a subjective evaluation (Nives, 1999;

Hernandez-Romero et al., 2004). In relation to

the parameter Pi, the maximum value of 4 was

assigned to parameters of relevant importance for

aquatic life as for example DO and TSS, while the

minimum value (unity) was assigned to parameters

Table 5

Water characteristics of the ‘‘La Torre and ‘‘Motilona’’ creeks (Guadarra

‘‘La Torre’’

Sampling

point 1

Sampling

point 2

Sampling

point 3

Sampling

point 4

pH 7.13 � 0.05 7.31 � 0.04 7.23 � 0.05 7.30 � 0.0

K (mS/cm) 0.69 � 0.02 0.69 � 0.01 0.56 � 0.02 0.58 � 0.0

TSS (mg/l) 37.2 � 4.9 12.1 � 1.2 11.0 � 1.6 19.8 � 4.6

Amm. (mg/l) 6.4 � 0.8 7.5 � 1.2 5.2 � 0.9 4.0 � 0.6

NO2� (mg/l) 0.06 � 0.01 0.11 � 0.01 0.12 � 0.01 0.17 � 0.0

NO3� (mg/l) 2.3 � 0.4 5.3 � 0.9 3.7 � 0.3 11.2 � 1.1

PT (mg/l) 2.5 � 0.3 2.0 � 0.2 3.2 � 0.3 2.2 � 0.3

COD (mg/l) 94.8 � 9.2 60.7 � 9.9 44.9 � 7.6 32.5 � 6.8

BOD5 (mg/l) 45.4 � 5.2 28.5 � 3.8 14.5 � 2.2 11.2 � 1.6

DO (mg/l) 2.6 � 0.2 4.6 � 0.2 2.3 � 0.3 5.3 � 0.2

D (mg/l) 8.0 � 0.2 6.2 � 0.2 5.6 � 0.2 5.1 � 0.1

T (8C) 12.8 � 0.5 12.1 � 0.5 11.8 � 0.6 12.3 � 0.5


with minor relevance such as for example tempera-

ture and pH.

Table 3 shows the values suggested for the

parameters Ci and Pi, used in the calculation of

WQI, which were based on European Standards (EU,

1975). When the values of WQI are in the range of 0–

25, the water must be classified as ‘‘very bad’’; for a

WQI value in the range of 25–50 the water is classified

as ‘‘bad’’; for WQI values in the range of 51–70 the

water classification is ‘‘medium’’; finally, when the

WQI values are within the range of 71–90 and 91–100

the water is classified as ‘‘good’’ and as ‘‘excellent’’,

respectively (Jonnalagadda and Mhere, 2001).

ma watershed)a

‘‘Motilona’’

Sampling

point 1

Sampling

point 2

Sampling

point 3

Sampling

point 4

2 7.22 � 0.04 7.23 � 0.02 7.56 � 0.03 7.39 � 0.08

2 0.63 � 0.01 0.77 � 0.01 0.72 � 0.02 0.73 � 0.01

41.0 � 0.8 102.0 � 1.6 28.9 � 3.7 37.7 � 6.8

4.6 � 0.2 12.2 � 1.6 6.1 � 0.6 10.1 � 1.4

2 0.06 � 0.01 0.04 � 0.01 0.14 � 0.02 0.03 � 0.01

2.8 � 0.2 3.4 � 0.6 8.1 � 1.4 0.9 � 0.2

2.3 � 0.2 3.3 � 0.4 4.4 � 0.4 5.5 � 0.7

159 � 4 359 � 32 163.3 � 18.4 161.1 � 21.5

51.0 � 2.5 109.0 � 11.6 60.7 � 8.7 66.6 � 11.7

2.4 � 0.2 2.7 � 0.2 4.7 � 0.2 4.1 � 0.2

7.8 � 0.2 8.1 � 0.2 5.9 � 0.2 6.9 � 0.2

14.1 � 0.3 12.4 � 0.6 12.6 � 0.6 11.2 � 0.5


Table 6

Water characteristics of the ‘‘La Virgen’’ and ‘‘Fuentecillas’’ creeks (Guadarrama watershed)a

‘‘La Virgen’’ ‘‘Fuentecillas’’

Sampling point 1 Sampling point 2 Sampling point 3 Sampling point 1 Sampling point 2 Sampling point 3

pH 7.09 � 0.07 7.34 � 0.02 7.33 � 0.07 7.19 � 0.06 7.21 � 0.10 7.01 � 0.10

K (mS/cm) 0.69 � 0.02 0.75 � 0.01 0.74 � 0.01 0.84 � 0.02 0.74 � 0.03 0.77 � 0.04

TSS (mg/l) 26.4 � 4.3 11.5 � 1.6 22.9 � 4.3 1 414 � 493 74.8 � 9.2 1919 � 508

Amm. (mg/l) 6.8 � 0.9 5.5 � 0.9 4.3 � 0.3 5.5 � 0.9 2.9 � 0.2 0.6 � 0.1

NO2� (mg/l) 0.08 � 0.02 0.02 � 0.03 0.09 � 0.02 0.09 � 0.01 0.14 � 0.02 0.03 � 0.0

NO3� (mg/l) 1.2 � 0.2 1.0 � 0.1 5.2 � 1.0 3.6 � 0.5 7.0 � 0.3 7.6 � 0.5

PT (mg/l) 7.5 � 0.1 7.6 � 0.2 8.3 � 0.1 8.2 � 0.1 2.5 � 0.1 0.9 � 0.1

COD (mg/l) 75.7 � 5.8 54.6 � 5.4 49.5 � 5.6 593 � 154 78.0 � 10.7 37.2 � 5.7

BOD5 (mg/l) 27.7 � 3.2 14.0 � 1.8 15.6 � 1.8 209 � 39 9.5 � 0.9 3.0 � 0.5

DO (mg/l) 3.7 � 0.2 3.6 � 0.3 4.3 � 0.3 2.7 � 0.2 5.1 � 0.1 7.3 � 0.1

D (mg/l) 6.6 � 0.1 7.1 � 0.2 6.6 � 0.3 7.6 � 0.2 5.9 � 0.1 4.1 � 0.1

T (8C) 14.3 � 0.5 12.3 � 0.4 11.5 � 0.3 14.4 � 0.4 10.7 � 0.3 9.8 � 0.4


3. Results and discussion

3.1. Characteristics of the waters of the different

watersheds

Tables 4–7 show the mean values and standard

errors of the data obtained during the study of the three

watersheds. Table 4 summarizes the results obtained

for the six sampling points tested in the Guadarrama

River. The mean values of pH remained practically

constant between the sampling points 1–5, but at the

sampling point 6 the mean value of pH decreased to

Table 7

Water characteristics of ‘‘Manzanares’’ and ‘‘Paris Park’’ watershedsa

‘‘Manzanares’’ ‘‘Paris Park’’

Sampling point 1b Sampling point 2c Sampling po

pH 6.96 � 0.04 6.60 � 0.05 8.04 � 0.16

K (mS/cm) 0.32 � 0.01 0.34 � 0.02 0.23 � 0.02

TSS (mg/l) 6.6 � 0.1 40.3 � 5.9 7.8 � 0.8

Amm. (mg/l) 1.6 � 0.2 0.8 � 0.1 0.9 � 0.1

NO2� (mg/l) 0.07 � 0.01 0.03 � 0.01 0.01 � 0.01

NO3� (mg/l) 6.9 � 0.5 8.9 � 0.9 0.7 � 0.1

PT (mg/l) 1.4 � 0.1 0.6 � 0.1 0.2 � 0.1

COD (mg/l) 32.7 � 3.5 33.0 � 2.0 52.5 � 6.8

BOD5 (mg/l) 10.3 � 1.0 8.0 � 0.9 13.4 � 1.6

DO (mg/l) 5.2 � 0.3 5.3 � 0.2 6.9 � 0.2

D (mg/l) 5.5 � 0.2 6.4 � 0.3 2.4 � 0.2

T (8C) 12.4 � 0.4 8.6 � 0.3 18.6 � 0.3

a Mean values � S.E. of 56 samples.b ‘‘La Trofa creek’’.c ‘‘Barrancohondo creek’’.

6.88, probably due to the increase in organic acid

concentration caused by the organic matter decom-

position introduced by La Torre and La Virgen creeks.

Similar behaviour was observed by other authors

(Bollinger et al., 1999; Jonnalagadda and Mhere,

2001; Simeonov et al., 2003) in other streams. The

conductivity, indirect measurement of dissolved solids

concentration, remained practically constant between

the sampling points 1–6. On the other hand, the

concentration of TSS augmented significantly from

point 1 to point 2 and decreased in point 3, increasing

again in points 4–6. The increase of suspended solids

int 1 Sampling point 2 Sampling point 3 Sampling point 4

8.17 � 0.15 7.63 � 0.12 7.87 � 0.14

0.26 � 0.01 0.27 � 0.01 0.25 � 0.01

30.8 � 4.5 41.0 � 4.2 28.5 � 4.6

0.9 � 0.1 1.1 � 0.2 0.9 � 0.1

0.01 � 0.01 0.01 � 0.01 0.01 � 0.01

5.2 � 1.8 3.5 � 1.2 3.7 � 1.0

0.4 � 0.1 0.4 � 0.1 0.5 � 0.1

59.1 � 8.7 64.0 � 8.1 45.3 � 8.0

24.8 � 3.8 20.9 � 2.0 16.7 � 2.9

5.5 � 0.3 3.6 � 0.3 5.4 � 0.3

3.9 � 0.3 5.9 � 0.2 4.0 � 0.3

18.6 � 0.4 18.2 � 0.4 18.2 � 0.4


may affect the metabolism of photosynthetic organ-

isms and the production of oxygen (Hernandez-

Romero et al., 2004; Jarvie et al., 1998; Jonnalagadda

and Mhere, 2001; Simeonov et al., 2003; Williams

et al., 2000). Ammonia, nitrite and nitrate concentra-

tions decreased from point 1 to point 2 and had slight

variation in points 3–6. Total phosphorus concentra-

tion was slightly higher in point 1 with respect to the

other sampling points. The concentration of COD and

BOD5 increased slightly from points 1–3, decreased

in point 4, increasing again in points 5 and 6.

However, significant differences of the mean values

could not be observed. The dissolved oxygen

concentration decreased from point 1 to point 3 and

remained practically constant in points 3–6. There-

fore, the oxygen deficit increased slightly as far as

point 3 and remained practically constant down-

stream. Similar results have been obtained by other

authors (Bollinger et al., 1999; Bordalo et al., 2001;

Carpenter et al., 1998; Jonnalagadda and Mhere,

2001; Liou et al., 2003; Nives, 1999; Smith, 1989).

Among the six sampling points evaluated, the water

quality appears to be affected after the point of mixing

between Guadarrama River and La Torre creek

(sampling point 3) and in point 5 due to the mixing

with La Virgen creek as may be observed in Fig. 1.

Table 5 shows the profiles of the parameters

monitored in La Torre creek. The values of pH

remained in the range of 7.1–7.3. The values of

conductivity decreased compared to the value

observed at the initial sampling point. The concen-

tration of TSS decreased down-stream between points

1 and 3 but increased at point 4 due to the discharge of

the final effluent of municipal wastewater treatment

plants (MWWTPs) (Fig. 1). The mean value of

ammonia concentration decreased while nitrites and

nitrate concentrations increased throughout the creek,

which may be caused by the nitrification process and

the incorporation of nitrified effluents from the

MWWTPs. The mean concentration of total phos-

phorus suffered a minimum variation in the points

evaluated although a slight tendency to increase was

appreciated from the point 1 to 4. The mean COD and

BOD5 concentrations decreased down-stream, the

minimum being observed at the point of discharge in

the Guadarrama River probably due to the oxidation

of the organic matter causing the reduction of

dissolved oxygen concentration and the increase in

oxygen deficit (D). Similar results were reported by

other authors (Bollinger et al., 1999; Bordalo et al.,

2001; Carpenter et al., 1998; Cox, 2003; Jonnala-

gadda and Mhere, 2001; Liou et al., 2003; Mullholand

et al., 2005; Nives, 1999; Quinn et al., 2005; Smith,

1989).

Table 5 also shows the characteristics of Motilona

creek water, influent of La Torre creek. The mean

value of pH and conductivity increased between

points 1 and 3, but decreased at point 4. The

concentration of TSS increased from point 1 to point

2, decreased at point 3 and increased again at point 4.

Ammonia and phosphorus concentrations increased

down-stream while the concentration of nitrate

increased between points 1 and 3, but decreased at

point 4. Values of COD and BOD5 were high in all

sampling points throughout the creek with mean

values of concentrations in the range of 159–359 mg/l

and from 51 to 109 mg/l, respectively, both equiva-

lents to a low-strength domestic wastewater. In

consequence, the DO concentration was lower than

5 mg/l, while the oxygen deficit values were higher

than 5.5 mg/l, respectively. The low quality of the

Motilona creek water may be caused by the discharge

of MWWTP effluents in points 1–3.

Table 6 summarizes the characteristics of La

Virgen creek. As can be seen the mean values of pH

and conductivity increased throughout the creek.

The concentration of TSS decreased from point 1 to

point 2 but increased again at point 3. The ammonia

concentration decreased, while the concentration of

nitrate and phosphorus increased throughout the

creek. Moreover, the values of COD and BOD5

decreased. These values were lower than those

observed in Motilona creek but may still be considered

high taking into account that the mean COD and

BOD5 values ranged from 49 to 76 mg/l and from 14

to 28 mg/l, respectively. Values of DO and oxygen

deficit were in the range of 3.6–4.4 mg/l and 6.5–

7.1 mg/l, respectively.

Table 6 also shows the characteristics of Fuente-

cillas creek. The mean value of pH slightly increased,

while the conductivity decreased throughout the

creek. The concentration of TSS decreased at point

2 but increased again at point 3. Ammonia concentra-

tion decreased while the nitrate concentration

increased throughout the creek, showing that the

nitrification process took place. Moreover, the COD


Fig. 2. Variation of the water quality index (WQI) and dissolved

oxygen deficit (D) for the different sampling points assessed in

‘‘Guadarrama’’ River.


oxygen deficit (D) for the different sampling points assessed in ‘‘La

Torre’’ creek.

and BOD5 values also decreased considerably from

593 to 37 mg/l and from 209 to 3 mg/l, respectively,

showing that a high rate of organic matter oxidation

occurred throughout the creek. This fact was remarked

because an increase of the DO concentration and a

decrease of the oxygen deficit were observed at the

same time.

Table 7 summarizes the characteristics of La Trofa

and Barrancohondo creeks (Manzanares watershed).

The mean value of pH was around 7 for La Trofa and

6.6 for Barrancohondo although significant differ-

ences could not be established. TSS concentration was

lower in La Trofa water than in Barrancohondo.

Nutrients concentration was very similar when

compared samples of La Trofa and Barrancohondo

creeks and Guadarrama River, but they were

significantly lower than those observed in the samples

of Guadarrama watershed. Mean values of COD and

BOD5 were around 30 and 10 mg/l, respectively, for

both points analysed. These values were significantly

lower than those obtained in the Guadarrama

watershed influents but comparable to the values

observed for Guadarrama River.

Finally, Table 7 also shows the characteristics of the

water in Paris Park. The mean values of pH were

higher than 7.60 in all points sampled with a

maximum of 8.17 for point 2, corresponding to

Superior Lake. Therefore, the water pH was appro-

priate for the presence and metabolism of photosyn-

thetic organisms (Bollinger et al., 1999; Bordalo et al.,

2001; Carpenter et al., 1998; Jarvie et al., 1998;

Melloul and Collin, 1998; Nagel, 2001). The mean

values of conductivity were lower than those observed

in waters from other watersheds. The concentration of

TSS was in the range of 7–41 mg/l, with a maximum

for point 2. The concentrations of ammonia were

lower than those observed in the waters from other

watersheds evaluated, while the concentrations of

nitrite were higher. The concentration of nitrate was at

a maximum at point 2. However, the values of this

parameter were lower than those obtained in the other

surface waters. The values of COD were in the range

of 45–65 mg/l, while the BOD5 values were in the

range of 13–25 mg/l, showing a maximum at point 2.

Finally, the values of DO were higher than 3 mg/l, the

minimum being at point 3, while the oxygen deficit

was lower than 6 mg/l at all points, the maximum

value being located at point 3.

3.2. Evaluation of the water quality with the use

of the WQI and D

In order to evaluate the feasibility of the WQI

and D as indicators of the level of pollution of the

water samples analysed, the values of these para-

meters were determined in the different sampling

points. Figs. 2–6 show plots of the variation of values

of WQI and the oxygen deficit (D) for the different




‘‘Motilona’’ creek.



‘‘Fuentecillas’’ creek.
sampling points taken from the Guadarrama river,
La Torre, La Motilona, La Virgen and Fuentecillas

creeks, respectively, during the monitoring of the

Guadarrama watersheds. Fig. 2 shows the values of

WQI and D along the Guadarrama River at the

different points sampled. The value of WQI was

70.4 at point 1; the value decreased to 64.8 at point

2 and increased to 70.0 at point 3, just before the

mixing point with waters coming from La Torre creek.

After this point a slight decrease was appreciated


oxygen deficit (D) for the different sampling points assessed in ‘‘La

Virgen’’ creek.

down-stream, achieving values of 68.6, 67.8 and

64.6 at points 4, 5 and 6, respectively. Fig. 2 also

shows the variation of oxygen deficit throughout the

river, this parameter being at a minimum at point 1 and

increasing at points 2 and 3, followed by a decrease

at point 4 and finally another increase at points 5 and

6. Therefore, a tendency of oxygen deficit to increase

with the decrease in WQI was appreciated. The

values of WQI obtained indicated that the quality of

the water in the Guadarrama River decreased in the

Las Rozas area. According to the values of WQI, the

water in the Guadarrama River may be classified as of

‘‘medium’’ quality. Fig. 3 shows the variation of the

WQI and D in La Torre creek. The value of WQI was

51.9 at point 1 and it increased to 56.0 at point 2 and

to 62.2 at point 3 just before the mixing with the water

in the Motilona, with a WQI of 52.0. This fact

determined a reduction in the WQI to 61.1 at point 4.

The water quality increased through sampling points

1–3, showing a certain self-purification capacity of

La Torre creek.

The variation of D appears to be inversely propo-

rtional to the WQI, the maximum value being at point

1 and the minimum at point 3. According to the values

of WQI obtained, the water quality in this case was

lower than that observed in the Guadarrama River,

although it may still be classified as a ‘‘medium’’




‘‘Paris’’ Park.

quality. Fig. 4 shows the variation of WQI and D for

the Motilona creek. The values of WQI progressively

increased from a value of 40 at the point 1 due to the

discharge of final effluent of a MWWTP to a value of

52 through sampling points 1–4 and at the same time

the value of D decreased from about 8 mg/l to around

6 mg/l. From sampling points 3–4, final effluents of a

two WWTPs were discharged into this creek but the

value of WQI was in the range of 49–52, which

corresponded to ‘‘medium’’ quality.

Fig. 5 shows the variation of WQI and D through

the sampling points of La Virgen creek. Three

sampling points were selected. The first sampling

point received treated wastewater from a municipal

wastewater treatment plant. At this point the WQI

was 53.3 and the value of D was 6.58 mg/l. The WQI

increased through this creek up to a value of 62.1 at

the point of discharge in the Guadarrama River

(point 3). At this point, the value of D decreased to

6.09 mg/l. The value of WQI shows that the quality

of the water in La Virgen creek was similar to the

water in La Torre creek and better than the water in

Motilona creek.

Fig. 6 shows the values of WQI and D for the

Fuentecillas creek. The values of WQI increased

throughout the creek, while the value of D decreased.

The first point coincided with the effluent from a

municipal wastewater treatment plant delivering

water with a WQI equal to 37.7 and D equal to

7.6 mg/l. The water quality improved considerably

down-stream, the values of WQI and D being equal to

52.2 mg/l and to 5.99 mg/l, respectively at point 2 and

70.5 and 4.11 mg/l, respectively at point 3 (discharge

in the Guadarrama River). Hence, the water in the

Fuentecillas creek was of better quality than that

observed in the Guadarrama River and in the other

creeks evaluated. Finally, taking into account all the

points sampled the water from the Fuentecillas creek

may be classified as ‘‘bad’’ at point 1 and as ‘‘good’’ at

the point of discharge in the Guadarrama River.

The evaluation of the data corresponding to the

Manzanares watershed (La Trofa and Barrancohondo

creeks) shows that the value of WQI was 65.2 and

62.2, while the value of D was 5.51 and 6.42 for the

first and the second point, respectively. These values

indicate a water quality classified as ‘‘medium’’ and

with the same quality as that obtained in the

Guadarrama River.

Fig. 7 shows the values of WQI and D for the

different samples corresponding to the Paris Park.

Point 1 (Wharf) showed the highest value of WQI and

the lowest value of D while point 3 (Estuary) showed

the lowest value of WQI and the highest value of D.

According to the average value of WQI of 72.2, the

water from the Wharf may be classified as ‘‘good’’,

while the others may be classified as of ‘‘medium’’

quality, very near the classification of ‘‘good’’, except

in the case of the water from the Estuary, with a WQI

of 55.5 and oxygen deficit of 5.66 mg/l. The values of

water quality in this basin are acceptable for

recreational use according to E.U. standards (EU,

1975).

3.3. Determination of the empirical relationship

between WQI and D

The results obtained show that the values of the

water quality index (WQI) appear to be related to the

values of oxygen deficit (D). It was found that when

the value of D increased, the value of WQI decreased.

Fig. 8 shows a plot of the average values of WQI

against the average values of D in all watersheds

evaluated. A straight line was obtained, with a linear

equation as follows:

WQI ¼ �6:39Dþ 93:61 (2)


Fig. 8. Regression line between the WQI and dissolved oxygen

deficit (D).

The regression coefficient (R2) was found to be

equal to 0.91 with p � 0.1. In order to demonstrate the

feasibility of Eq. (2) to determine the water quality,

based on the data of dissolved oxygen deficit (D),

Table 8

Comparison of the results obtained for WQI by using Eqs. (1) and (2)

Sample WQI (Eq. (1)) WQI (Eq. (2))

P1. Guadarrama 70.4 68.6






P1. La Torre 51.9 42.4

P2. La Torre 56.0 53.9

P3. La Torre 62.2 52.6

P4. La Torre 61.1 57.5

P1. Motilona 40.0 41.5




P1. La Virgen 53.3 51.6

P2. La Virgen 60.7 54.5

P3. La Virgen 62.1 54.7

P1. Fuentecillas 37.7 44.8



P1. La Trofa 65.2 58.4

P2. Barrancohondo 62.2 60.8

P1. Wharf 72.7 78.1

P2. Superior 69.1 68.7

P3. Estuary 55.5 57.4

P4. Inferior 65.6 68.7

the classification of the water was obtained by using

Eq. (2) and the data obtained compared to those

obtained in Eq. (1). Table 8 shows the results obtained

in both equations, which coincided in 93% of the

samples analysed with a probability level of 95% on

the basis of the 26 samples studied.

3.4. Influence of the climate conditions on the

WQI

The influence of the climate conditions on the WQI

of the Guadarrama River was evaluated. Fig. 9 shows

the variation of the WQI for the six sampling points of

Guadarrama River as a function of the season. Values

of WQI were in the range of 80–90 during the winter,

and these values were significantly higher than those

observed in other seasons. In the rest of the seasons the

values of WQI ranged from 50 to 70 and significant

differences could not be clearly observed among them.

Three factors may influence the WQI: the precipita-

tion level, the temperature of the surroundings and

solar radiation. The high level of rainfall helped to

Water classification (Eq. (1)) Water classification (Eq. (2))

Good Medium

Medium Medium

Medium Medium

Medium Medium

Medium Medium

Medium Medium

Medium Bad

Medium Medium

Medium Medium

Medium Medium

Bad Bad

Bad Bad

Medium Bad

Medium Medium

Medium Medium

Medium Medium

Medium Medium

Bad Bad

Medium Medium

Medium Medium

Medium Medium

Medium Medium

Good Good

Medium Medium

Medium Medium

Medium Medium


Fig. 9. Variation of the WQI for the six sampling points of Gua-

darrama River as a function of the season.

decrease the WQI because it increased the amount of

water in the river, thus increasing the runoff and

uncontrolled pollution. An increase in temperature

contributes to an increase in the biological activity,

decreasing the dissolved oxygen concentration and

increasing the value of D. The increase in solar

radiation is favourable for photosynthesis, increasing

the concentration of oxygen producing organisms

(microalgae and aquatic plants), which increase the

dissolved oxygen concentration during the day but

cause the reduction of oxygen concentration at night.

Moreover, the increase of the concentration of

photosynthetic organisms produced an additional

oxygen demand during the decomposition of dead

biomass. For all these reasons the water quality

decreased in autumn, spring and summer, when

compared with winter. Other authors observed the

same phenomenon (Bordalo et al., 2001; Couillard

and Lefebvre, 1985; Hernandez-Romero et al., 2004;

Jonnalagadda and Mhere, 2001; Pesce and Wunderlin,

2000; Rudolf et al., 2002).

4. Conclusions

The monitoring of the Las Rozas watersheds

demonstrated that water quality of Guadarrama

watershed was slightly affected in the section of

the river within this town. The characteristics of the

waters corresponding to the Manzanares River and

the Paris Park watersheds indicate an acceptable level

of quality for the assigned uses. The determination of

WQI to the waters monitored demonstrated the

importance of this index in order to classify the

waters studied. It was found that the WQI was around

70 at the entrance to Las Rozas section and was 64.6

at the outlet of the section. This water was classified

as of ‘‘medium’’ quality. The water quality was influ-

enced by the quality of the creek influents of La Torre

and La Virgen with a value of WQI of 61.1 and 62.1,

respectively. The values of WQI for the Manzanares

River show a quality classified as ‘‘medium’’. The

results obtained in the monitoring of the Paris Park

show a water quality between ‘‘medium’’ and good.

The best quality was found in the Wharf and the worst

in the Estuary. A high linear relationship between the

WQI and the oxygen deficit (D) of the samples was

found. The classifications of water based on the two

methods coincided in 93% of the samples studied.

This allowed the determination of WQI based on the

values of the oxygen deficit. The estimation of the

WQI by the calculation of the oxygen deficit is an

advantageous way for a simple, rapid and economical

determination of a water quality. It was found that

water quality was influenced by the climatic condi-

tions, the highest qualities being observed during

the winter.

Acknowledgements

We thank the municipal Government of ‘‘Las

Rozas’’ (Madrid, Spain), the ‘‘Consejerıa de Medio

Ambiente’’ for providing financial support for the

present work through the programme of ‘‘Optimiza-

cion de Recursos’’ and to the ‘‘Laboratorio Municipal

de Salud Publica’’.

References

American Public Health Association (APHA), 1999. Standard

Methods for the Examination of Waters and Wastewaters,

20th ed. APHA, Washington, DC, USA.

Bollinger, J.E., Steinberg, L.J., Harrison, M.J., Crews, J.P., Eng-

lande, A.J., Velasco-Gonzalez, C., White, L.E., George, W.J.,

1999. Comparative analysis of nutrient data in the lower Mis-

sissippi River. Water Res. 33, 2627–2632.

Bordalo, A.A., Nilsumranchi, W., Chalermwat, K., 2001. Water

quality and uses of the Bangpakong River (Eastern Thailand).

Water Res. 35, 3535–3642.


Carpenter, S.R., Caraco, N.F., Correll, D.L., Howarth, R.W.,

Sharpley, A.N., Smith, V.H., 1998. Nonpoint pollution of

surface waters with phosphorus and nitrogen. Ecol. Appl. 8,

559–568.

Couillard, D., Lefebvre, Y., 1985. Analysis of water quality indexes.

J. Env. Manage. 21, 161–179.

Cox, B.A., 2003. A review of dissolved oxygen modelling tech-

niques for lowland rivers. Sci. Total Environ. 314–316, 303–

334.

Cude, C., 2001. Oregon water quality index: a tool for evaluating

water quality management effectiveness. J. Am. Water Res.

Assoc. 37, 125–137.

European Union (EU), 1975. Council Directive 75/440/EEC of 16

June 1975 concerning the quality required of surface water

intended for the abstraction of drinking water in the Member

States. Official Journal L 194, 25/07/1975, 0026–0031.

Hernandez-Romero, A.H., Tovilla-Hernandez, C., Malo, E.A.,

Bello-Mendoza, R., 2004. Water quality and presence of pes-

ticides in a tropical coastal wetland in southern Mexico. Marine

Poll. Bull. 48, 1130–1141.

House, M.A., Newsome, D.H., 1989. Water quality indices for the

management of surface water quality. Water Sci. Technol. 21,

1137–1148.

Jarvie, H.P., Whitton, B.A., Neal, C., 1998. Nitrogen and phosphorus

in east coast British rivers: speciation, sources and biological

significance. Sci. Total Environ. 210/211, 79–109.

Jonnalagadda, S.B., Mhere, G., 2001. Water quality of the Odzi river

in the eastern highlands of Zimbabwe. Water Res. 35, 2371–

2376.

Liou, S.M., Lo, S.L., Hu, C.Y., 2003. Application of two-stage fuzzy

set theory to river quality evaluation in Taiwan. Water Res. 37,

1406–1416.

Melloul, A.J., Collin, M., 1998. A proposed index for aquifer water-

quality assessment: the case of Israel’s Sharon region. J. Env.

Manage. 54, 131–142.

Mullholand, P.J., Houser, J.N., Maloney, O.K., 2005. Stream diurnal

dissolved oxygen profiles as indicators of in-stream metabolism

and disturbance effects: fort Benning as a case study. Ecol. Ind. 5,

243–252.

Nagel, J.W., 2001. A water quality index for contact recreation.

Water Sci. Technol. 43, 285–292.

Nives, S.G., 1999. Water quality evaluation by index in Dalamatia.

Water Res. 33, 3423–3440.

Park, S.S., Na, Y., Uchrin, C.G., 2003. An oxygen equivalent model

for water quality dynamics in a macrophyte dominated river.

Ecol. Model. 168, 1–12.

Pesce, S.F., Wunderlin, D.A., 2000. Use of water quality indices to

verify the impact of Cordoba City (Argentina) on Suquia River.

Water Res. 34, 2915–2926.

Quinn, N.T.W., Jacobs, K., Chen, K.W., Stringfellow, W.T., 2005.

Elements of decision support system for real-time management

of dissolved oxygen in the San Joaquin River Deep Water Ship

Cannel. Env. Mod. Software 20, 1495–1504.

Rudolf, A., Ahumada, R., Perez, C., 2002. Dissolved oxygen content

as an index of water quality in San Vicente Bay, Chile (36

degrees, 450S). Env. Monit. Assess. 78, 89–100.

Simeonov, V., Stratis, J.A., Samara, C., Zachariadis, G., Voutsa, D.,

Anthemidis, A., Sofoniou, M., Kouimtzis, Th., 2003. Assess-

ment of the surface water quality in Northern Greece. Water Res.

37, 4119–4124.

Smith, D.G., 1989. A new form of water quality index for rivers and

streams. Water Sci. Technol. 21, 123–127.

Spanou, M., Chen, D., 1999. An object-oriented tool for the control

of point-source pollution in river systems. Env. Mod. Software

15, 35–54.

Swamee, P.K., Tyagi, A., 2000. Describing water quality with

aggregate index. J. Environ. Eng. 126, 451–455.

Wang, H., Hondzo, M., Xu, C., Poole, V., Spacie, A., 2003. Dis-

solved oxygen dynamics of streams draining an urbanized and

an agricultural catchment. Ecol. Model. 160, 145–161.

Williams, R.J., White, C., Harrow, M.L., Neal, C., 2000. Temporal

and small-scale spatial variations of dissolved oxygen in the

Rivers Thames, Pang and Kennet, UK. Sci. Total Environ. 251/

252, 497–510.

Use of the water quality index and dissolved oxygen deficit as simple indicators of watersheds pollution

Documents