Waste Management 25 (2005) 768779 www.elsevier.

com/locate/wasman

Environmental diagnosis methodology for municipal waste landlls

F. Calvo a, B. Moreno
a

a,b

, M. Zamorano

a,*

, M. Szanto

Department of Civil Engineering, E.T.S.I.C.C.P., University of Granada (Spain), Campus de Fuentenueva s/n, 18071 Granada, Spain b Institute of Water Research, University of Granada, 18071 Granada, Spain c Building Engineering, Catholic University of Valparaiso, Avda, Brasil, 2147 Valparaso, Chile Accepted 9 February 2005 Available online 17 May 2005

Abstract A large number of countries are involved in a process of transformation with regard to the management of municipal solid waste. This process is a consequence of environmental requirements that occasionally materialise in legislation, such as the European Council Directive 31/99/EC on waste release in the European Union. In some cases, the remediation of old landlls can be carried out in compliance with environmental requirements; in other cases, it is necessary to proceed with the closure of the landll and to assimilate it into its own environment. In both cases, it is necessary to undertake a diagnosis and characterisation of the impacted areas in order to develop an adequate action plan. This study presents a new methodology by which environmental diagnosis of landll sites may be carried out. The methodology involves the formulation of a series of environmental indeces which provide information concerning the potential environmental problems of the landlls and the particular impact on dierent environmental elements, as well as information related to location, design and operation. On the basis of these results, it would be possible to draw up action plans for the remediation or closure of the landll site. By applying the methodology to several landlls in a specic area, it would be possible to prioritize the order of actions required. 2005 Elsevier Ltd. All rights reserved.

1. Introduction The environmental impact produced by municipal solid wastes has received special social and environmental attention in recent decades. The great number of studies involving the degradation of wastes in release points provides evidence of leachate and biogas formation. These emissions are generated during long periods of time, even after waste disposal ceases (Hirata et al., 1995; Christensen et al., 1996; Bramryd, 1997; Fatta et al., 1997; Meadows et al., 1997). However, leachate and biogas are not the only emissions with a potential negative impact, other impacts include: settlements in the landll mass (Siege, 1990; Sing and Murphy, 1990; Tieman, 1990), odours, as well as dioxin, furans and

Corresponding author. Tel.: +34 958 249458; fax: +34 958 246138. E-mail address: zamorano@ugr.es (M. Zamorano).

other emissions of toxic gases (Deipser and Stegmann, 1997), risk of explosion and re (Espinace, 1992), and noise and risks on public health (Tieman, 1990 and Stegmann, 2000). Today, there is specic legislation in a number of countries aimed at environmental control of municipal solid waste landlls, from start-up to closure and assimilation into the environment. In Europe, a Council Directive passed in 1999 provided for the regulation of waste disposal in landlls and the adaptation of old currently operational release points which are compatible with the environment. This new situation forced the different environmental organisations to carry out an inventory of release points located within their territory, with the aim of developing a Conditioning Plan or a Closure Plan according to the new legislation and depending on the type and degree of environmental problems found in each case (Directive 31/99/EC). The deadline for the Conditioning Plans was the end of

0956-053X/$ - see front matter 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2005.02.019

F. Calvo et al. / Waste Management 25 (2005) 768779

769

2002; however, in the specic case of Spain, it has been reported that about 52% of the landlls do not as yet have a plan (Uriarte et al., 2003). An environmental diagnosis of landlls is necessary in order to identify the environmental problems generated by each site. A number of authors have worked on dierent methods for evaluating the environmental impacts of new landlls: cause-eect matrix, revision lists, interconnection networks, ad-hoc methods, etc., (Gomez, 1999; Hernandez, 1995; Carter et al., 1997; Conesa, 1997). However, these studies were not applicable in the present case, since we are concerned with operating facilities. Nevertheless, on the basis of these methods, other methods have been developed to carry out environmental diagnosis in operating facilities, with the aim of resolving particular problems in certain provinces or groups of municipalities, as has been done in Huesca (Barrenchea et al., 1997), the province of Granada (Moya et al., 2001) and the Valparaiso Region in Chile (Szanto et al., 1984). In most cases, the methods involved the performance of an inventory of local natural phenomena in order to compile lists of impacts in the landlls where monitoring was undertaken, but the sphere of application was limited. Such methods made it possible to compare landlls on an environmental basis, but not to take decisions about their control, closure, capping, or recovery. Finally, it should be emphasised that in these studies there were practically no studies of the physical medium. All of the assessments were related to the release point, without taking into account the characteristics of their environment (Calvo, 2003). For these reasons, we aimed to develop a new methodology whose main objective was to characterise municipal solid waste landlls and to carry out an environmental diagnosis. In addition, the methodology was carried out at dierent release points in a region, providing environmental information on present and future conditions and allowing future action plans to be established. In collaboration with the Catholic University of Valparaiso, the University of Granada has developed the new method and has tested it in over 30 landlls in the province of Granada and the Fifth Region of Valparaiso in Chile. As described below, the method is based on the assessment of environmental indeces which identify the impacts generated by each release point in their environment. Taking those indeces as a starting point, appropriate measures may be considered.

 Environmental interaction between the release point and certain potentially aected environmental elements or parameters.  Environmental values of the parameters taking into consideration: surface water, groundwater, atmosphere, soil and health.  Operational conditions of the landll from the environmental point of view. The following initial hypotheses are proposed for carrying out the methodology (Calvo, 2003):  The diagnosis is only valid at the moment of assessment and its validity decreases with time if the landll is not monitored.  The methodology can only be performed for municipal solid waste landlls, regardless of the scale of operation.  Landll waste composition may be obtained from reported historical data, waste characterisation data at the population centre or from in situ characterisation. The methodology is based on the formulation of a general index called Global EnvironmentLandll Interaction Index or Impact Index (ELI). The purpose of this index is to assess the environmentlandll interaction. By assessing dierent impacts as a whole for each parameter, the ELI aims to represent the overall environmental state in the landll. Its rate expression is as follows: X X ELI ELIi ERIi EWCi ERIgroundwater EWCgroundwater ERIsurfacewater EWCsurfacewater ERIatmosphere EWCatmosphere ERIsoil EWCsoil ERIhealth EWChealth ; 1 where ELI is the Global EnvironmentalLandll Interaction Index or Impact Index for the global environmental impact; ELIi is the Environmental Landll Interaction Index for parameter i; i, the parameters: groundwater, surface water, atmosphere, soil, and health; ERIi is the Environmental Risk Index. Risk Index for the Environmental Aectation of parameter i; EWCi is the Environmental Weighting Coecient. Environmental Coecient to weigh parameter i. The parameters involved in the calculation of the EnvironmentalLandll Index are dened below. By means of this methodology, ranges of scores are obtained for ELI. These scores are used to classify the overall environmental impact of landlls as low, average and high, as shown in Table 1.

2. Denition of the methodology The methodology is based on the use of environmental indeces in order provide a quantitative assessment of the following aspects in each landll analysed:

770

F. Calvo et al. / Waste Management 25 (2005) 768779

Table 1 Classication of release-point impact (Calvo, 2003) Value of ELI 035 3671 72105 Impact qualication Low Average High

2.1. Environmental risk index (ERI) The ERI aims to gauge the potential for environmental impact for each observed parameter, reecting whether or not interaction exists between the processes in the release point and the characteristics of the environment. The ERI has the following rate expression: X ERIi Pbci eVi ; 2 where (i) is the parameter; Pbci is the probability of parameter i contamination; eVi is the environmental value of parameter i. 2.1.1. Probability of parameter contamination The probability of contamination for each parameter will depend on the scale of operation, as well as waste characteristics and on the spread of impacts in the landll environment. In order to assess the probability of contamination, a number of landll variables are selected. These are phenomena which show high sensitivity to the biochemical and physical processes directly or indirectly inuencing the environmental impact of a given parameter. For example, the variables chosen to assess the probability of contamination of surface water are: degree of compaction, types of waste and of organic matter, age of landll, type and amount of cover material, slope to surface beds, permeability of surrounding strata, surface water in the surroundings, surface drainage systems, rain fall, landll lining system, nal cover, control of leachate, location of release-point in surface runo, location of release-point in the storage of ood water, and operationality of the release point. Evaluation of these variables makes it possible to assess contamination risks in the landlls by means of the Pbcj, or Probability of Contamination for each contamination variable. This probability has the following expression (3): Pbcj Cj W j ; 8 3

tween disposal processes and the environmental characteristics related to the variable. Values obtained range from 0 to 4. Wj is the weighting of variable j. Since not all the variables aect the deposit point environment in the same way, the concept of weighting (Wj) is dened for each variable. Each variable weighting is related to the concept of structural elements at the deposit point, or in other words elements which intervene in the impact presented by each parameter. The structural elements considered are: presence of organic matter, moisture content and density of the wastes. These three concepts participate in the main biochemical and physical processes produced in the release point and lead to the production of gas (Szanto et al., 1997; Tchobanouglus et al., 1998) and leachate (Costa, 1982; Tchobanouglus et al., 1998), aecting all parameters and providing greater weighting to the dierent landll variables. In the calculation of probability of impact, all variables which are directly related to these structural elements have a maximum weighting. Similarly, those variables which are not directly related to the structural elements but are a direct eect of impact risk for the parameter considered are also weighted at their maximum values. The data in Tables 2 and 3 show contamination probability values for each variable based on variable classication and weighting. A justication of the variables intervening in the contamination probability of each parameter and an account of variable classication and weighting may be found in Calvo et al. (2004). By way of illustration, we might consider the variable aquifer characteristics (hydrogeological importance), which is involved in the evaluation of probability of contamination of groundwater. This variable may acquire

Table 2 Contamination probability values for variables with no relation to structural elements and no direct inuence on the parameter evaluated Weighting 1 Classication Very high High Mean Low Very low Classication value 4 3 2 1 0 Probabilistic value Pbcj = Cj Wj/8 0.500 0.375 0.250 0.125 0.000

where j is a variable for parameter i. All the variables taken into account for each parameter have a theoretical justication of their state, which is closely related to the processes of emissions taking place in the landll and is based on the guidelines established in European Council Directive 1999/31/EC. Cj is the classication of variable j. This depends on the state of the variable and provides information on the situation of the deposit point or the interaction be-

Table 3 Contamination probability values for variables related to structural elements or with direct inuence on the evaluated parameter Weighting 2 Classication Very high High Mean Low Very low Classication value 4 3 2 1 0 Probabilistic value Pbcj = Cj Wj/8 1.00 0.75 0.50 0.25 0.00

F. Calvo et al. / Waste Management 25 (2005) 768779 Table 4 Classication and weighting of the variable aquifer characteristics for the parameter groundwater (Calvo, 2003) Variable Drastic evaluator Condition Very high High Mean Low Very low Weighting 2 Classication value 4 3 2 1 0

771

the classication and weighting shown in Table 4. Overall evaluation of aquifer characteristics is performed by means of a cartographic technique known as Drastic (Larry, 1997). This evaluation provides a scale of vulnerability for the aquifer obtaining values of very high, high, mean, low and very low. The variable of aquifer characteristics is assigned maximum weighting (2), since it is directly related to the contamination of groundwater. The evaluation of variables facilitates the assessment of contamination risks in the landlls. Such assessment is carried out through the concept of Probability of Parameter Contamination (Pbci) which has the following mathematical expression (4): P Pbcj Pbci ; 4 N where N is the number of variables of each parameter. Pbcj and Pbci obtain a nal value between 0 and 1 with classications of maximum, high, average, low, and zero, depending upon the results obtained from the appropriate mathematical formulation. 2.1.2. Environmental value As stated in Eq. (2), the Environmental Risk Index depends not only on probability but also on the environmental value of the parameter considered. The aim of

this concept is to identify and quantify the environmental aspect of each parameter in the landll environment, taking as a starting point the relationship between the landlls environmental and/or social and political characteristics and the emissions in the release point. In order to quantify the environmental value, characteristics related to the landll processes for each parameter are assigned a numeric value. Subsequently, a rate expression is established for each parameter to connect the values assigned. The data in Table 5 show water state characteristics related to landll processes, together with their justication and quantication in the case of surface water. The expression used to quantify these ve characteristics is as follows: Environmental Value Surface Water A1 A2 2 A3 A4 A5. 5

The environmental assessment is considered null for parameters related to surface and groundwater when water is more than 1000 m downstream from the edge of the release mass (Barcelona et al., 1990; Stegmann, 2000). If the landll is within a protected area (natural park, nature reserve or protected landscape) the environmental values will be considered as maximum for all parameters, so that these locations receive priority of action as long as the above exception does not occur. The nal environmental value for each parameter is classied as high, average, low, and null, depending upon the results obtained from the appropriate mathematical expression. These classications correspond to nal values of 3, 2, 1 and 0, respectively. 2.2. Environmental weighting coecient The EnvironmentLandll Interaction Index, as expressed in Eq. (1), results from the addition of the

Table 5 Characteristics considered to quantify the environmental value of surface water (Calvo, 2003) Characteristic Type of surface water ow (A1) Condition Seasonal (brooks less than 5 m in width), supply channels, irrigation ditches Permanent (rivers, lakes, springs or sources) Primary branch line Secondary branch line Tertiary branch line Human consumption Irrigation/Agricultural Without use Non-existence of species Existence of some species Existence of many species In a high runo area In a low runo area Outside runo area Environmental value 1 2 3 2 1 3 2 1 1 2 3 1 2 3

Type of surface ow branch line (A2)

Use of water (A3)

Existence of animal or vegetable species (A4)

Release point in runo areas (A5)

772

F. Calvo et al. / Waste Management 25 (2005) 768779

Table 6 Indicators used in evaluating environmental weighting coecients Indicator Intensity (I) Indicator value Of no importance = 0 Low = 1 Moderate = 2 High = 3 Reversibility (R) Reversible = 1 Irreversible = 3 Temporary = 1 Permanent = 3 Localised = 0 Partial = 1 Extensive = 2 Total = 3 Impact on relevant environmental element is reversible To reverse environmental degradation would be highly dicult, whether for technical or nancial reasons Alteration of environment is temporary Alteration of environment is indenite Limited to landll site area Extends over microbasin of landll site Reaches areas outside microbasin including neighbouring inhabited areas Involves macrobasin of landll site Observations Due to characteristics of relevant environmental element, impact is not important or not considered

Duration of eect (D) Extension (E)

Environmental Risk Indexes (ERI) of each parameter multiplied by their Environmental Weighting Coecients (EWC). This coecient takes into account the fact that some parameters are environmentally more important than others when quantifying the total or overall impact. Quantication is carried out on the basis of indicators that connect the impact with spatial, temporal and legislative characteristics of the parameter (Leat and Rodrguez, 1998; Gomez, 1999). The EWC represents a technical/political/social decision regarding the relative importance of, e.g., human health, landll settlement, or groundwater protection so that priorities may be established within a common framework. Assessment of environmental weighting coecients is carried out through the following expression: EWC I R D E ; 3 6

we shall describe the results obtained in three landlls from the province of Granada, in order to illustrate how the methodology works and its potential results. 3.1. Denition of the release points selected The landlls selected, denominated A, B and C, are located in three villages in the province of Granada. The initial approach and denition of these release points was obtained from studies carried out by the Instituto Tecnologico Geominero de Espana (Geologi cal Survey of Spain) and Granadas Provincial Government (Moya et al., 2001). This information was complemented by two visits to the release points per season. The data presented in Table 7 show the relevant information obtained from each landll. 3.2. Methodology implementation The methodology was implemented once the minimum necessary information had been compiled following several visits to the release points. The results were as follows: Fig. 1 shows the values of the EnvironmentalLandll Interaction Index (ELIi) for each parameter (surface water, groundwater, atmosphere, soil and health) for all three release points, as well as the values of the Global EnvironmentalLandll Interaction Indexes or Impact Indexes (ELIA, ELIB and ELIC). On the basis of the impact scale established with values from the EnvironmentalLandll Interaction Index (Table 1), the overall impact of Landll A is considered Average, while B and C are considered Low. This index allows us to establish priorities and plan the order of actions on the landlls. Landll A shows the highest environmental impact and there is consequently a higher priority to undertake environmental action on this landll.

where, for the impact on each parameter considered: I is the Intensity; R is the reversibility; D the duration of eect; E is the extension. Values for I, R, D and E are calculated as shown in Table 6. The hierarchy obtained for EWC by means of Eq. (6) is subsequently contrasted with current legislation on environmental protection. In accordance with Gomez (1999), dierent values are introduced for each indicator involved in the environmental weighting coecient for each parameter. The EWC for each parameter obtains values between 0 and 9, with classications of null, low, average, high or maximum.

3. Implementation example: results Implementation of the methodology for 30 landlls located in the Fifth Region of Valparaso (Chile) and the province of Granada (Spain) formed the basis on which our research was developed and tested. Here,

F. Calvo et al. / Waste Management 25 (2005) 768779 Table 7 Landll information Name Location Population served Composition of disposed waste Geology A Communal area of Huetor Santillan 1200 habitants Domestic waste with organic content higher than 40%. Debris and some hazardous waste Detrital sediments from marly rocks: limestones and dolomites with karstic characteristics B Communal area of Cenes de la Vega 1300 habitants Domestic waste together with other waste such as tyres, hazardous and agricultural waste Sedimentary rock C

773

Geomorphology

Surface hydrology Ground hydrology

Climate

Soil uses Biota

Natural site known as Sierra de Huetor, in a trough middle mountain. Inclinations are between 10% and 15%. High risk of undermining the release material Conuence of many courses from surface runo The aquifer is in a substrate composed mainly by marbles and dolomites from the Alpujarra mountains between the subbetic and penibetic mountains. The substrate may be considered permeable with groundwater at approximately 5-m depth Mild Continental-Mediterranean climate. Between 1000 and 1500-m altitude. Eective rain: 100300 mm. Low exposure to wind Sierra de Huetor, Natural Park Pines and bushes. Fauna without special characteristics Access road not asphalted. Main road network at 600 m from release point Population centres further than 2.5 km

Trough. Inclination between 15% and 20%. High risk of undermining the release material as a consequence of high inclinations Courses from surface runo with torrential character The substrate is impermeable or semipermeable. The hydro-geological importance is high-average, with aquifers at approximately 5-m depth

Communal area of Las Gabias 10,000 habitants Mainly organic. Presence of waste from construction and demolition, as well as bulky and hazardous waste Located on the aquifer Vega de Granada, constituted by materials from the lling of depressed areas between mountains and alluvial formations Low inclination (0.5%)

No water bodies near the release point Substrate permeability is variable, but low in general. Hydro-geological importance is not high as the aquifer area is 80 m under the release point

Mild Continental-Mediterranean climate. Between 500 and 1000-m altitude. Eective rain: 100300 mm. Low exposure to wind Diversity of crops Bushes, olive groves, fruit trees and conifers. Fauna without special characteristics Access road to the release point not asphalted. Main road network over 500 m from the release point Population centres further than 200 m

Mild Continental-Mediterranean climate. Moderate wind intensity around population centres. Eective rain: 0100 mm Agricultural soil. Dry-land olive cultivation Fauna without special characteristics

Infrastructure

Population centres

Access road to the release point not asphalted. Main road network at 500 m from release point Population centres further than 500 m

Analysis of each release point using the Environmental Risk Indexes for each parameter obtained the following results: 3.2.1. Landll A Fig. 1 shows that for Landll A, groundwater was the parameter with the highest environmental impact, followed by soil, surface water and atmosphere. The least aected parameter was health. Tables 8 and 9 show values obtained by variables aecting probability of contamination for each parameter. On the basis of this information, it was possible to determine the Probability of Contamination (Pbci) for each parameter, the state of the variables in this release point and the question of whether they need any type of control. Fig. 2 shows the probability of contamination for all three release points according to each parameter. For Landll A, the parameters

health, soil and atmosphere presented the same value and had the highest probability of contamination; surface water was the parameter with lowest probability of contamination. All parameters were greater than 0.6, so they all presented a high probability of contamination. Fig. 3 shows the Environmental Risk Index (ERIi) for all the release points with respect to each parameter. The gure shows that health, soil and atmosphere were the parameters with the highest environmental risk in Landll A, followed by groundwater and surface water. All parameters presented average or high Environmental Risk Indeces when compared with the maximum value of the ERI. Fig. 4 shows the Probability of Contamination and the Environmental Risk Index for Landll A. In accordance with the expression for Environmental Risk Index [Eq. (2)], environmental values for each parameter

774

F. Calvo et al. / Waste Management 25 (2005) 768779

Surface water

E-L I = (ERIiEWCi) E-L IA= 39 E-L IB= 29 E-L IC= 21.6

Global Environmental-Landll Interaction Index or Impact Index for Landll A was 39. 3.2.2. Landll B Fig. 1 shows that groundwater was the parameter with the highest environmental impact for Landll B, followed by soil. The least aected parameters were health, atmosphere and nally surface water, all of which presented similar EnvironmentalLandll Interaction Indexes. As stated above, Tables 8 and 9 show the values obtained by variables aecting probability of contamination for each parameter, while Fig. 2 shows the probability of contamination for all the release points according to each parameter. For Landll B, health was the parameter with the highest probability of contamination, followed by soil and atmosphere with the same value, and nally surface water and groundwater. All parameters except groundwater were higher than 0.6, so they all had a high probability of contamination, while groundwater showed average probability. Fig. 3 shows that for Landll B, the parameter with the highest environmental risk index was health, followed by groundwater, soil and atmosphere. Surface water presented the lowest Environmental Risk Index.

7.2 4 7.2 4.8 0 3.6 1.8 4.2 9 7 6.3

Health

4.2

Groundwater
10.8

Soil 10.5
E-Li Indexes for landfill A E-Li Indexes for landfill B E-Li Indexes for landfill C

Atmosphere

Fig. 1. EnvironmentalLandll Interaction Index and Global EnvironmentalLandll Interaction Index for each Landll.

achieved a maximum value of 3, due to the fact that the landll was located in a natural park. Finally, the weighting coecients for each parameter obtained values of 4, 6, 3, 5 and 2, respectively, while the

Table 8 Variables aecting surface water and groundwater contamination probability Parameters Variables Landlls A Surface water Compaction Waste and organic matter types Age Cover material Inclination to the surface beds Permeability of surrounding strata Surface water in the surroundings Surface drainage systems Rainfall Landll lining system Control of leachate Final cover Release point location in surface runo Release point location in oodwater storage volume Operationality Compaction Waste and organic matter types Age Cover material Aquifer characteristics Surface drainage system Rainfall Landll lining system Control of leachate Final cover Fault Release-point location in surface runo Release-point location in oodwater storage volume Operationality 4 6 2 6 2 1 3 1 1 6 6 6 0 0 2 4 6 2 6 0 3 1 6 3 6 1 0 0 2 B 4 6 2 6 2 1 3 1 1 6 6 6 0 0 2 4 6 2 6 0 3 1 6 3 6 1 0 0 2 C 4 6 2 4 1 2 3 1 0 6 6 4 0 0 2 4 6 2 4 1 3 0 6 3 4 1 0 0 2

Groundwater

F. Calvo et al. / Waste Management 25 (2005) 768779 Table 9 Variables aecting atmosphere, soil and health contamination probability Parameters Variables Landlls A Atmosphere Compaction Waste and organic matter types Cover material Rainfall Gas control Control of leachate Paths Final cover Age Operationality Compaction Waste slope Cover material Gas control Control of leachate Final cover Age and organic matter percentage Landll lining system Release-point location in surface runo Release-point location in oodwater storage volume Operationality Compaction Waste and organic matter types Age Cover material Gas control Control of leachate Final cover Environmental control Distance to population Distance to infrastructure Available equipment Operationality 4 6 6 1 6 3 3 6 2 2 1 1 0 1 1 1 4 1 6 6 2 6 4 2 6 3 3 6 3 0 0 6 3 B 4 6 6 1 6 3 3 6 2 2 1 1 0 1 1 1 4 1 6 6 2 4 6 2 6 3 3 6 3 6 0 6 2

775

C 4 6 4 0 6 3 2 4 2 2 1 1 4 1 1 4 1 1 6 6 2 4 6 2 4 3 3 4 3 0 0 6 2

Soil

Health

All the parameters showed average environmental risk except health, which had high risk. Probability of Contamination and the Environmental Risk Index for Landll B are shown in Fig. 5. This indicates that environmental values for health and groundwater were 3, while the remaining parameters obtained a value of 2. Finally, the weighting coecients for each parameter obtained values of 4, 6, 3, 5 and 2, respectively, and the Global EnvironmentalLandll Interaction Index or Impact Index for Landll B was 29. 3.2.3. Landll C Fig. 1 shows that the parameter with the highest environmental impact for Landll C was soil, followed by groundwater, health and atmosphere. Surface water did not present any environmental impact. For values of variables aecting probability of contamination for each parameter of Landll C, see Tables 8 and 9. The information in Fig. 2 shows that health,

soil and atmosphere were the parameters with the highest probability of contamination, followed by surface water and groundwater. Health, soil and atmosphere had high probability of contamination while surface and groundwater had average values. With regard to the Environmental Risk Index (ERI), the parameters with highest risk for Landll C were health and soil, followed by groundwater (Fig. 3). Atmosphere presented a low environmental risk and surface water presented no risk. Health and soil showed high risk of environmental impact, groundwater had average risk, atmosphere had a low risk and surface water presented null risk. Fig. 6 shows Probability of Contamination and the Environmental Risk Index for Landll C. Environmental values were 3 for health and soil, 2 for groundwater, 1 for atmosphere and 0 for surface water. The Environmental Risk and EnvironmentalLandll Indexes for surface water obtained a score of 0 because this parameter presented an environmental value of 0.

776

F. Calvo et al. / Waste Management 25 (2005) 768779

Ppci =0 Nonexistent 0<Ppci0.3 Low 0.3<Ppci0.6 Mean 0.6<Ppci1 High

Surface water
1

Surface water
1.8

ERIi = PpciVai

0.5

0.6

Health Health
1

0.8 0.6

0.5 0.6 0.4

Groundwater
1

2.1

0.7

0.6 1.8 0.6

Groundwater

0.7

0.7 2.1

0.7 2.1

0.6 0.7

0.6 0.7
1

Soil

Atmosphere
Pbc i for landfill A ERI i for landfill A

Soil

Atmosphere
Fig. 4. Landll A: Probability of Contamination and Environmental Risk Index for each parameter.

Pbci for landfill A Pbci for landfill B Pbci for landfill C Maximum Pbci

Fig. 2. Probability of contamination for each parameter of each landll.

Surface water

ERIi = PpciVai

Surface water
1.8

ERIi =0 Nonexistent 0<ERIi1 Low 1<ERIi2 Mean 1.8

Health
2.4 0.8 0.7

0.5 0.5 0.7 1.4 1.5

Groundwater

Health
2.4

2.1 1.8 1.4 2.1 Soil 1.8

1 0 1.2 0.6 1.4

Groundwater

1.5

1.4

Soil
2.1

Atmosphere
ERI i for landfill B

Pbc i for landfill B

Atmosphere
ERI i for landfill A ERI i for landfill B ERI i for landfill C Maximum ERI i

Fig. 5. Landll B: Probability of Contamination and Environmental Risk Index for each parameter.

Fig. 3. Environmental Risk Index for each parameter of each landll.

Surface water

ERIi =PpciVai

Finally, the weighting coecients for each parameter obtained values of 4, 6, 3, 5 and 2, respectively, and the Global EnvironmentalLandll Interaction Index or Impact Index for Landll C was 21.6.

Health

1.8

0.5 0.4 0 0.6 0.6 0.6 1.2

Groundwater

4. Discussion The large number of release points with unknown environmental conditions has become a contamination problem in many countries. In view of the poor economy in many of the towns where the landlls are located, it is necessary to develop a methodological evaluation system which is both technically and eco-

Soil

1.8

Atmosphere

Pbc i for landfill C

ERI i for landfill C

Fig. 6. Landll C: Probability of Contamination and Environmental Risk Index for each parameter.

F. Calvo et al. / Waste Management 25 (2005) 768779

777

nomically feasible, in order to overcome the lack of knowledge concerning the environmental state of these landlls (Calvo et al., 2004). Implementation for 30 landlls in Chile and Spain suggests that this methodology is suitable for screening the environmental interaction between the release point and its environment. Furthermore, the methodology has proven to be exible, one of the requirements in evaluamez, 1999). tion and environmental assessment (Go The implementation of the methodology provides environmental diagnosis of several aspects:  The ELI enables us to determine the state of potential landll impact on the landlls own environment. When implemented for several landlls in one region or area, conclusions may be reached concerning the state of degradation of waste products and the most aected areas within the study area. The ELI also facilitates the development of a list of all release points with priority given to those needing urgent actions due to the scale of environmental impact.  Focusing on the study of each landll individually, the ERI enables us to determine which parameters are most aected by the landll, making it easier to prioritize suitable control actions.  The state of impact of each variable is studied by means of the concept of contamination probability, by which conditioning, closing and sealing actions may be planned and directed. Although it is not easy to act directly on certain variables, such as waste type, justication of the variables allows better understanding of landll characteristics, and thus leads to better decisions regarding the right actions to be taken. Sensitivity analysis of the dierent variables is a useful tool in the operation, management and regulation of landlls.  Analysis of index results provides information about the suitability of the release-point locations with regard to environmental values as well as for certain variables. In the case of the three landlls in Granada, the following aspects were highlighted by the results:  Landll A was the most aggressive with respect to its environment, with a Global EnvironmentalLandll Index of 39, followed by Landlls B and C with 29 and 21.6 ELI values, respectively.  Tables 8 and 9 show that variables with the most direct inuence on the environmental impact of Landll A were obtained from the analysis of the contamination probabilities: cover material, compaction, gas control, nal cover, equipment, control of leachate, landll lining system and characteristics of the aquifer. The results demonstrate that the facilities have been totally neglected. The maximum environ

mental values demonstrated the low suitability of the locations of release-points. In order for this landll to meet legislative requirements, it is necessary to draw up a closure, sealing and assimilation plan, taking into account the variables mentioned above when implementing the sealing coats and recovering materials as established in Directive 31/99/EC. Whether carrying out rehabilitation plans or plans for closure, sealing and assimilation into the environment, the necessary measures may be developed on the basis of a detailed study of the contamination probabilities for each variable in each parameter. In the case of Landll B, variables directly inuencing the environmental impact were slope of the waste in the release point, leachate and gas control, waste compaction, waste type and available equipment (Tables 8 and 9). Again, these results show a total neglect of the facility. Bearing in mind the acceptable characteristics of the location of the release-point, the option of maintaining the exploitation could be justied if a rehabilitation plan were carried out taking into account the variables mentioned above. This control is based on the justication about the state of the variables in the tables showing contamination probability. When the contamination probabilities for the dierent parameters for Landll C were analyzed (Tables 8 and 9), the variables aecting the environment directly were found to be: landll lining system, cover material and nal cover, control of leachate and gases, types of waste, equipment and distance from the population centre. The results show inadequate operation of the facilities and partial neglect. As the release point was located in a site with acceptable characteristics, operation could be maintained if a conditioning plan controlling the above mentioned variables were implemented. If we obtain a mean EnvironmentalLandll Index for each parameter for Landlls A, B and C, it becomes clear that the parameter most aected by the state of these three landlls is groundwater, followed by soil, atmosphere, health and surface water (Fig. 7). Similarly, the highest environmental values are for surface water. These two results indicate the central importance of an aquifer in the province, known as el Acufero de la Vega de Granada. By contrast, the parameter obtaining the lowest scores in environmental values for the selected landlls is the atmosphere. This is mainly due to the characteristics of the atmospheric parameter in the landlls concerned, which does not facilitate the dispersal of biogas or noises and odours generated. With regard to the soil, the environmental value is always 2 or 3, thus demonstrating the importance of recovering the impacted area as a recoverable edaphic substrate (Directive 31/99/EC). (The concept

778

F. Calvo et al. / Waste Management 25 (2005) 768779

Surface water

Acknowledgement This research has been made possible, thanks to nancial assistance from the Science and Technology Ministry, which published the research project entitled Design and implementation of methodologies for the environmental diagnosis of urban waste landlls and waste dumps.

3.7
Health Groundwater

4.2

9.0

4.1

References
Barcelona, M., Wehrmann, Keely, J., Pettyhohn, W., 1990. Contamination of Groundwater. Noyes Data Corporation, Park Ridge, NJ, USA. Barrenchea, P., Larruga, S., Varea, M., 1997. Basureros y vertederos de Aragon (Dump sites and landlls in Aragon). Fundacion Ecologa y Desarrollo, Huesca. Bramryd, T., 1997. Landlling in the perspective of the global CO2 balance, In: Sixth International Landll Symposium, Sardinia, Italy. Calvo, F., 2003. Metodologa de diagnostico y caracterizacion ambiental de vertederos de residuos solidos urbanos para su control, cierre, sellado y reinsercion (Methodology for environmental diagnosis and characterization of munipical waste landll for its control, closure, sealing and rehabilitation). Thesis Doctoral, Universidad de Granada. Calvo, F., Zamorano, M., Moreno, B., Ramos, A.F., 2004. Met odologa de diagnostico ambiental para vertederos de residuos urbanos (Environmental diagnosis methodology for municipal waste landlls). Montserrat Zamorano Toro (Ed.). Carter, L., Ward, C.H., Ginger, W., McCarty, P.L., 1997. Methods for Assessment of Groundwater Pollution Potential. Wiley, New York. Christensen, T.H., Cossu, R., Stegmann, R., 1996. Landlling of Waste: Biogas. E FN SPON Edit, London. Conesa, V., 1997. Instrumentos de la gestion ambiental en la empresa (Company tools for environmental management). Mundi Prensa Libros S.A., Madrid (Espana). Costa, H.B., 1982. Edmund Avaliaccao da producao da percolado do lixo y de la capacidad ltrante de aterras sanitarios. Cuadernos Tecnicos 21, 9099. Council Directive 1999/31/EC of 26 April 1999 on the landll of waste. Ocial Journal of the European Communities L 182 42 (16 July), 00010019. Deipser, A., Stegmann, R., 1997. Biological degradation of VCCs and CFCs under simulated anaerobic landll conditions in laboratory test digesters. Environmental Science and Pollution Research 4 (4), 209216. Espinace, R., 1992. Problemas geotecnicos de los rellenos sanitarios (Landll geotechnical problems). Revista de Ingeniera Civil. CEDEX-MOPU 77, 7783. Fatta, D., Voscos., Haralambous, K.J., Loizidou, M., 1997. An assessment of the eect of landll leachate on groundwater quality. In: Sixth International Landll Symposium, Sardinia, Italy. Gomez, D., 1999. Evaluacion de impacto ambiental (Environmental impact assessment). Ediciones Mundi-Prensa. (Editorial) Agrcola Espanola, S.A., Madrid. Hernandez, S., 1995. Ecologa Para Ingenieros (Ecology for Engineers). El Impacto Ambiental. Colegio Ingenieros de Caminos, Canales y Puertos, Madrid. Hirata, T., Hanashima, M., Matsufuji, Y., Yanase, R., Maeno, Y., 1995. Construction of facilities on closed landlls. In: Fifth International Landll Symposium, Sardinia, Italy.

Soil

8.8

Atmosphere

Fig. 7. Mean of EnvironmentalLandll Indexes for each parameter of Landlls A, B and C.

of soil in this methodology refers to the physical substrate on which wastes are deposited. The establishment of a landll modies the characteristics of the soil, causing it to lose all of its value as a physical substrate for sustaining terrestrial life and human activity).

5. Conclusions By means of the methodology described, an environmental diagnosis of dierent urban waste release points may be performed, providing sucient data to screen their environmental problems. Summarized in a series of environmental indeces, the results may be used as a basic tool to study the suitability of release points and to control their operational state. The methodology may also be used to develop projects for conditioning release points, with a view to maintaining their use with lower environmental impact and in accordance with current legislation. Finally, it may be applied to closing, sealing and recovering projects when the need to discontinue using the landll has been established. Similarly, the methodology has direct applicability as a powerful tool for planning and prioritising actions if applied to dierent release points in a particular region. It might even make it easier to plan the funding allocated to landlls. The underlying basis of the environmental diagnosis was developed in accordance with the guidelines of The European Council Directive 31/99/EC on waste release. Territorial application of this methodology may include countries in the EU, as well as every other country where similar legislation exists, or indeed where there is no legislation or the legislation is less prescriptive than this Directive. The methodology or an adapted form of the methodology might also be applied to other elds, subject to a study of the requirements in each case.

F. Calvo et al. / Waste Management 25 (2005) 768779 Larry, W., 1997. Manual De Evaluacion De Impacto Ambiental (Guide for Environmental Impact Assessment). McGraw-Hill, Madrid. Leat, J., Rodrguez, E., 1998. Guas Para La Evaluacion Del Impacto Ambiental De Proyectos De Desarrollo Local (Guides for Environmental Impact Assessment in Projects for Local Development). Instituto Latinoamericano y del Caribe de Planicacion Economica y Social (ILPES), Santiago de Chile. Meadows, C., Franklin, D., Campbell, P., 1997. Global methane emissions from solid waste disposal sites. In: Sixth International Landll Symposium, Sardinia, Italy. Moya, J., Lopez, J.A., Rubio, J.C., Beas, J., Gallardo, V., Alcan, G., 2001. Censo de Vertederos Incontrolados de la Provincia de Granada y orientaciones al sellado (Landlls census in Granada (Spain) and measures for sealing). Diputacion de Granada ITGE. Siege, R.A., 1990. Slope stability investigation at a landll in southern California. In: Arvid Landva, Davil Knowles, G. (Eds.), Geotechnics of Waste Fills. Theory and Practice. ASTM, Philadelphia, pp. 259284. Sing, S., Murphy, B., 1990. Evaluation of the stability of sanitary landlls. In: Arvid Landva, Davil Knowles, G. (Eds.), Geotechnics of Waste Fills. Theory and Practice. ASTM, Philadelphia, pp. 240 258.

779

Stegmann, R., 2000. Landll Emissions and Environmental Impact: An Introduction. Curso Superior Sobre Gestion Y Diseno De Vertederos. Club Espanol de Residuos, Madrid, pp. 413. Szanto, M., Palma, J., Espinace, R., 1984. Catastro De CONAMA (CONAMA Landlls Census). Escuela De Ingeniera En Con struccion. Universidad Catolica de Valparaso, Chile. Szanto, M., Palma, J., Espinace, R., 1997. Apuntes De Cursos Sobre Vertederos (Notes about Landlls). Universidad Catolica de Valparaso, Chile. Tchobanouglus, G., Theisen, H., Virgil, S., 1998. Gestion Integral De Residuos Solidos (Integrated Solid Waste Management). McGrawHill, Madrid. Tieman, G.E., 1990. Stability consideration of vertical landll expansions. In: Arvid Landva, Davil Knowles, G. (Eds.), Geotechnics of Waste Fills. Theory and Practice. ASTM, Philadelphia, pp. 285 302. Uriarte, J., Calvo, F., Zamorano, M., Moreno, B., 2003. Metodologa de diagnostico ambiental de vertederos como herramienta en la planicacion ambiental. Datos obtenidos en el Observatorio sobre Vertederos de residuos de 2003 (Environmental diagnosis methodology for municipal waste landlls as tool for environmental planning. Data from landlls observatory 2003). Residuos, Revista Tecnica 75, 5864.