Abstract

Rainwater harvesting is an ancient practice aiming to cover water needs for domestic, irrigation and livestock uses. In this study, the rainwater harvesting tank size was investigated to meet five water-need levels of a mixed goat–sheep farm using a daily water balance method. This method was applied using daily rainfall data for a period of 16 years from six meteorological stations in selected regions of Greece, characterized by different rainfall regimes and well-developed livestock activity, taking into account, among other parameters, the water needs of animals, the rainwater collection area and the runoff coefficient. There is a great variation in the rainwater harvesting tank size among the stations studied due to differences in the annual rainfall and the maximum dry period. Results showed that meeting full demands (100% reliability) requires tank sizes ranging from 20 m3 for short dry period stations–low demand scenario (320 L/day) to 115 m3 for long dry period stations–high demand scenario (576 L/day), assuming a maximum collection area of 450 m2. Correspondingly, reliability analysis showed that very high values of reliability (95%) can be obtained with tank sizes ranging from 10 to 85 m3, respectively.

INTRODUCTION

Rainwater harvesting tanks are a globally widespread sustainable water management practice. They are used to meet domestic, irrigation and livestock water needs (Campisano et al. 2017). In Greece, the use of rainwater harvesting tanks (cisterns) has been widespread for millennia, particularly in areas with difficult access to other freshwater sources (Angelakis 2016; Yannopoulos et al. 2016).

While there is extensive research on rainwater harvesting tank sizing for domestic use (Londra et al. 2015; Campisano et al. 2017), there is a lack of relevant research in the case of livestock, which can be a significant economic sector in several areas. Livestock farming in Greece is an important economic sector. According to the livestock research of the Hellenic Statistical Authority (HELSTAT 2015), goat–sheep farms dominate with 162,722 holdings, followed by pig and cattle farms with 18,941 and 15,899 holdings, respectively.

A critical factor for the viability of a livestock farm is the adequate amount of water in both quantity and quality. The amount of water required to feed animals depends on their age, weight and growth stage, but also on environmental factors such as temperature and relative humidity. With regard to quality (Melidis et al. 2007; Mendez et al. 2011; Gikas & Tsihrintzis 2012), the basic requirements include low salt content, lack of substances and absence of odor (Ward & McKague 2007).

The meeting of water requirements in Greek livestock farms has always been a serious problem, since they are usually located in remote and/or mountainous areas and away from water supply networks. Nowadays, these water needs are usually met either by wells or by transporting water with water tanks. Wells face problems such as: (a) legal restrictions in drilling (water-use license); (b) local water availability; and (c) high construction cost. Water transportation is often laborious, time consuming and costly. A sustainable solution to this problem is a rainwater harvesting system with a tank which should not be very large due to the high construction cost, but also should not be too small, given the risk of not meeting the water demand. Moreover, these tanks, according to Greek legislation, do not demand a water-use license.

Daily water balance models (Ghisi & Ferreira 2007; Imteaz et al. 2011; Palla et al. 2011; Campisano & Modica 2012a, 2012b; Tsihrintzis & Baltas 2014; Londra et al. 2015), or probabilistic models (Guo & Baetz 2007; Cowden et al. 2008; Basinger et al. 2010) have been used to determine the optimal size of rainwater tank. Overall, the results show that the capacity of rainwater tanks cannot be standardized, because it is strongly affected by several local variables, such as local rainfall, the collection surfaces and the water demand (Aladenola & Adeboye 2010; Ghisi 2010; Palla et al. 2012; Londra et al. 2015). Also, sizing methods depend on the standards and regulations adopted by each country (Tsihrintzis & Baltas 2014; Londra et al. 2015).

The purpose of this study is to investigate the rainwater tank capacity for livestock use in Greece. The sizing of rainwater tanks was performed using a daily water balance method in six regions of Greece, with different rainfall regimes and well-developed goat–sheep holdings. The required rainwater tank capacities were calculated meeting the full water needs (100%) of a goat–sheep mixed farm. The reduction of rainwater tank capacity by increasing the rainwater collection area was investigated. Furthermore, the required rainwater tank size was estimated using a reliability analysis procedure.

MATERIALS AND METHODS

Study area

Study areas were selected based on livestock development, particularly in goat–sheep mixed farms; 82% of livestock holdings in Greece are of this type, followed by pig and cattle farms with 10% and 8%, respectively (data processed from HELSTAT 2015).

Figure 1 depicts the goat–sheep farm distribution among the administrative regions of Greece (data processed from HELSTAT 2015). As shown, the Western Greece and Crete regions are leading with 18% and 17%, respectively, followed by Peloponessus, Thessaly, Central Greece and Epirus with 10%, 9%, 9% and 9%, respectively. The percentages for the other regions are less than 6%. Based on this distribution, study areas were selected from districts with a percentage equal to or greater than 9%.

Figure 1

Distribution of goat–sheep livestock holdings in the administrative regions of Greece (locations shown in Figure 2).

Figure 1

Distribution of goat–sheep livestock holdings in the administrative regions of Greece (locations shown in Figure 2).

Rainfall data

In order to size the rainwater harvesting tank, knowledge of the rainfall regime of the region where the tank will be installed is required. Six rainfall stations (Table 1), one from each of the abovementioned areas (Figure 2), characterized by different rainfall regimes, were selected. The daily rainfall data used were obtained from the Ministry of the Environment database (http://kyy.hydroscope.gr) regarding the time period 1980–1996.

Table 1

Rainfall station characteristics, mean annual rainfall values (P), mean, max and min values of the longest annual dry periods (Ndd, Ndd,max, Ndd,min) and the corresponding standard deviations () and () of the rainfall time series of the studied stations

Rainfall stationLocationAdministrative regionAltitude (m)Longitude (deg)Latitude (deg)P (mm)σP (mm)Ndd,max (days)Ndd,min (days)Ndd (days)σΝdd (days)Time series (years)
Armenoi Rethymnon Crete 373.3 24.45970 35.30145 977.7 223.2 155 65 118.7 29.1 16 
Kastania Korinthia Peloponnesus 987.2 22.38063 37.86678 1000.5 159.7 134 22 55.8 30.8 16 
Poros Riganiou Etolia and Akarnania Western Greece 181.8 21.74962 38.50779 1180.0 243.2 104 25 52.4 21.0 16 
Karpenisi Evritania Central Greece 962.2 21.79335 38.91478 1011.8 239.6 87 20 42.2 19.1 16 
Elati Trikala Thessaly 908.9 21.53881 39.50150 1405.2 322.0 64 21 45.4 15.3 16 
Agnanta Arta Epirus 660.0 21.08269 39.47395 1399.4 244.8 46 22 33.6 8.3 16 
Rainfall stationLocationAdministrative regionAltitude (m)Longitude (deg)Latitude (deg)P (mm)σP (mm)Ndd,max (days)Ndd,min (days)Ndd (days)σΝdd (days)Time series (years)
Armenoi Rethymnon Crete 373.3 24.45970 35.30145 977.7 223.2 155 65 118.7 29.1 16 
Kastania Korinthia Peloponnesus 987.2 22.38063 37.86678 1000.5 159.7 134 22 55.8 30.8 16 
Poros Riganiou Etolia and Akarnania Western Greece 181.8 21.74962 38.50779 1180.0 243.2 104 25 52.4 21.0 16 
Karpenisi Evritania Central Greece 962.2 21.79335 38.91478 1011.8 239.6 87 20 42.2 19.1 16 
Elati Trikala Thessaly 908.9 21.53881 39.50150 1405.2 322.0 64 21 45.4 15.3 16 
Agnanta Arta Epirus 660.0 21.08269 39.47395 1399.4 244.8 46 22 33.6 8.3 16 
Figure 2

Locations of the rainfall stations studied.

Figure 2

Locations of the rainfall stations studied.

Rainfall data were selected on an availability and completeness basis, among others. The duration of the time series (16 years) exceeds the recommendations of Mitchell et al. (2008) about the minimum length of rainfall record used for rainwater harvesting tank sizing, i.e., 10 years.

The rainfall stations were selected in areas with well-developed livestock activities and different rainfall regimes in order to reflect the effect of local conditions on rainwater tank sizing.

The mean annual rainfall and the longest annual dry period were determined for each station. The mean annual rainfall was calculated as:  
formula
(1)
where: Pt is the daily rainfall depth of the tth day; and N is the total number of recorded rainfall depths.

The longest annual dry period was defined as the maximum period in days without rain or with effective rainfall less than or equal to 1 mm (Londra et al. 2015).

Methodology

Daily water balance model

A daily water balance model (Tsihrintzis & Baltas 2014) was used to determine the rainwater harvesting tank size:  
formula
(2)
where: St is the stored volume at the end of the tth day (m3); St−1 is the stored volume at the beginning of the tth day (m3); Rt is the harvested rainwater volume during the tth day (m3); Dt is the daily water demand (m3); and Vtank is the capacity of the rainwater tank (m3).

The daily scale was used given the purpose of the study and data availability as suggested by Campisano & Modica (2015).

Taking into account that the daily harvested rainwater, Rt (m3), from a catchment area is calculated as  
formula
(3)
and the daily water demand, Dt (m3), of a goat–sheep farm is calculated as  
formula
(4)
then the daily rainwater stored volume can be calculated as  
formula
(5)
where: C is the runoff coefficient; A is the rainfall collection area (m2); Peff,t is the daily effective rainfall depth at the end of the tth day (m); Nanim is the number of served animals; and q is the daily water use per animal (m3/day).

In this study, taking into consideration that the rainwater collection areas used for livestock purposes are usually made of smooth and impermeable materials (e.g., metal, plastic), a runoff coefficient C = 0.9 was used (Kinkade-Levario 2007). The daily effective rainfall is equal to the daily rainfall minus a first flush equal to 0.33 mm (Yaziz et al. 1989) for improving the quality of harvested rainwater from concentrations of dust, leaves and bird droppings in the rainfall collection area.

Also, discrete values of rainwater collection areas of 250, 300, 350, 400 and 450 m2 were studied, because such areas are commonly found in farms, and no license is required to construct buildings and auxiliary areas for animal housing with an overall surface of 450 m2, according to the Join Ministerial Decision 281273/27-8-2004 (Official Gazette of the Hellenic Republic 2004). The value of Nanim was initially set at 80 goat–sheep, which is an average Greek farm (data processed from HELSTAT 2015) and the value of the daily water consumption per animal, q, was set at 5 L/animal/day (Ward & McKague 2007) defining a demand level of 400 L/day. However, in order to investigate more than one demand level, additional scenarios of Nanim and q were studied. Specifically, the additional values of Nanim were set at 64 and 96 goat–sheep, which represent a difference of ±20% of the average Nanim value (80), and the additional q value was set at 6 L/animal/day, which represents an increase of 20% of the basic q value (5 L/animal/day). Overall, the demand levels were defined as 320 L/day (Nanim = 64 and q = 5 L/anim/day), 384 L/day (Nanim = 64 and q = 6 L/anim/day), 400 L/day (Nanim = 80 and q = 5 L/anim/day), 480 L/day (Nanim = 80 and q = 6 L/anim/day or Nanim = 96 and q = 5 L/anim/day), and 576 L/day (Nanim = 96 and q = 6 L/anim/day).

Rainwater tank sizing

The calculation of daily stored water volume is iterative and starts from an initial stored water volume St−1 = S0 at time t = 0. The most conservative value of S0 is S0 = 0 for an initially empty rainwater tank, and the less conservative is the maximum value S0 = Vtank for an initially full rainwater tank. Any other value S0 for a partially full rainwater tank can also be used (Tsihrintzis & Baltas 2014).

To take into account the capacity of the rainwater tank, Vtank, when calculating the daily stored water in the tank, the following heuristic algorithm (Tsihrintzis & Baltas 2014; Londra et al. 2015) can be used iteratively:  
formula
(6)
where: St is the stored volume at the end of the tth day (m3); St−1 is the stored volume at the beginning of the tth day (m3); and St,tank is the actual available stored water volume in the tank at the tth day.
The volume of water that overflows, Ot, from the tank when the tank is full can be computed from the following algorithm:  
formula
(7)

Overflow happens only after the demand has been met.

In the case when the stored water volume in the tank, St,tank, is inadequate to meet the demand, Dt, then the demand will be satisfied, in parts or in whole, with water delivered from other sources, Τt, which can be calculated as follows:  
formula
(8)

By applying the abovementioned procedure, using the daily rainfall record for the region where the tank will be located, we can determine the tank volume, Vtank, considering that the daily water needs of animals will be fully met by rainwater (i.e., 100% reliability of the rainwater harvesting system), taking into account that the Tt value is equal to 0 m3. In this study this procedure was realized using the optimization algorithm ‘goal seek’ of the software Microsoft Excel.

Reliability analysis of rainwater harvesting system

To estimate the rainwater tank efficiency for a demand level, the reliability coefficient (Re) was used. The coefficient Re (%) of a rainwater tank is calculated as the ratio of the number of days when the intended demand is fully met by the available stored rainwater (Nfm) to the total number of days simulated (Ntot):  
formula
(9)

Accordingly, in this study the reliability coefficient was calculated using daily rainfall data records on a continuous time series for a simulation period of 16 years (Ntot = 5,840 days).

RESULTS AND DISCUSSION

In Table 1 the mean annual rainfall values (P), the maximum (Ndd,max), minimum (Ndd,min), and mean (Ndd) values of the longest annual dry periods, as well as the corresponding standard deviations of the study areas are presented, as derived from the analysis of daily rainfall data for the stations used. Most of the stations are mountainous or semi-mountainous, and are characterized by high annual rainfall values (977 < P < 1,405) and different dry periods (33.6 < Ndd < 118.7). More specifically, among the stations studied, ‘Agnanta’ has the second highest rainfall (1,399.4 mm) and the shortest dry period (33.6 days), while ‘Armenoi’ has the lowest rainfall (977.7 mm) and the longest mean dry period (118.7 days), marking the range in which the rest of the stations lie.

In Figure 3, the 16-year time series of daily rainfall depths are presented. As shown, each station has a unique rainfall pattern, revealing its special characteristics. So, in ‘Armenoi’, it is clear that rainfall is concentrated in the winter period of each year, with high daily values (often >50 mm), leaving a distinctive summer dry period, typical of the Mediterranean climate type. In contrast, in the rest of the stations, rainfall is, more or less, spread throughout the year. This pattern is typical of the continental climate subtype, where summer storms often occur over mountains, interrupting the summer dry period. Another typical characteristic of Mediterranean climate, revealed in the figure, is the great dry-period variability through the years and the frequent occurrence of very dry years followed by wet ones. This is depicted in the relatively high values of the Ndd standard deviation, σNdd (Table 1).

Figure 3

Daily rainfall data for 16 years (1980–1996) for each selected station studied.

Figure 3

Daily rainfall data for 16 years (1980–1996) for each selected station studied.

In Figure 4, comparative results of rainwater tank sizing obtained by applying the daily water balance model in all stations are illustrated, wherein the tank volume is given as a function of the rainwater collection area (250 to 450 m2) to fully meet the demand level of 400 L/day. While the rainwater tank capacity depends on both water demand (number of animals, daily water needs) and water supply (rainfall, dry period, rainwater collection area, runoff coefficient), it is obvious that differences in rainfall regime among stations play an important role in rainwater tank sizing.

Figure 4

Rainwater tank volumes in relation to rainwater collection area to fully meet the water demand (400 L/day) of an average goat–sheep livestock holding (80 animals) in six regions of Greece with different rainfall regimes.

Figure 4

Rainwater tank volumes in relation to rainwater collection area to fully meet the water demand (400 L/day) of an average goat–sheep livestock holding (80 animals) in six regions of Greece with different rainfall regimes.

As shown in Figure 4, the smallest tank volumes (40.5 to 32 m3) required to fully meet the water demand were calculated in ‘Agnanta’ station (Arta, Epirus) where both the highest rainfall and the shortest dry period were observed among the other stations. The largest tank volumes (84.5 to 77.7 m3) were calculated in ‘Armenoi’ station (Rethymnon, Crete), with both lowest rainfall and longest dry period. Intermediate tank volumes (65.7 to 43.1 m3) were calculated in the rest of the stations, following their rank in dry period. It is also shown that further increasing the collection area has only a minor effect on reducing the required tank size. Thus, we can conclude that the minimum tank size can be obtained for a collection area of about 450 m2.

The above results are in agreement with recent studies which have shown that large capacities of rainwater tanks are required in areas with long dry periods, acting as the dominant factor against rainfall in rainwater tank sizing (Palla et al. 2012; Souza & Ghisi 2012; Londra et al. 2015). Specifically, the classification of stations with regard to the required tank size follows the size of the dry period, irrespective of the collection area or demand level, as shown in Figure 5. In this figure, results of required rainwater tank volumes to fully meet the water demands of 320, 400 and 576 L/day in relation to mean values of the longest dry period for minimum and maximum studied collection areas (250 and 450 m2) of all stations studied are presented.

Figure 5

Required rainwater tank volumes to fully meet the water demands of 320, 400 and 576 L/day in relation to the mean value of the longest dry period for min and max studied collection areas (250 and 450 m2) in six regions of Greece with different rainfall regimes.

Figure 5

Required rainwater tank volumes to fully meet the water demands of 320, 400 and 576 L/day in relation to the mean value of the longest dry period for min and max studied collection areas (250 and 450 m2) in six regions of Greece with different rainfall regimes.

However, due to the effect of high rainfall, a great reduction of tank volume by increasing the collection area is observed. As shown in Figure 4, the required tank volumes decrease as the rainwater collection area increases. In stations ‘Armenoi’, ‘Kastania’, ‘Poros Riganiou’, ‘Elati’, ‘Agnanta’ and ‘Karpenisi’ tank volumes are reduced by 6.8 m3 (8%), 7.8 m3 (13%), 8.0 m3 (13%), 10 m3 (19%), 8.5 m3 (21%) and 18.3 m3 (28%), respectively, by increasing the collection area from 250 to 450 m2 (80%). This reduction can be fitted by linear equations for ‘Agnanta’ and ‘Elati’, second-order polynomial equations for ‘Armenoi’ and ‘Poros Riganiou’ and third-order polynomial equations for ‘Kastania’ and ‘Karpenisi’ (Table 2). In the cases of ‘Agnanta’ and ‘Elati’, the linear relationship between tank volume and rainwater collection area due to the very high and uniformly distributed rainfall allows linear exploitation of the collection area.

Table 2

Curve fitting equations, Vtank(A), on calculated rainwater tank volumes in relation to collection area using the daily water balance model (curves shown in Figure 4)

Rainfall stationVtank(A)R2
Armenoi Vtank =0.0002A2 − 0.1775A + 115.92 0.994 
Kastania Vtank =− 0.000005A3 + 0.0052A2 − 1.9523A + 297.15 0.990 
Poros Riganiou Vtank = 0.0002A2 − 0.1569A + 89.933 0.996 
Karpenisi Vtank = −0.000007A3 + 0.0079A2 −3.0261A + 435.32 0.995 
Elati Vtank = −0.051A + 65.375 0.977 
Agnanta Vtank = −0.0425A + 51.198 1.000 
Rainfall stationVtank(A)R2
Armenoi Vtank =0.0002A2 − 0.1775A + 115.92 0.994 
Kastania Vtank =− 0.000005A3 + 0.0052A2 − 1.9523A + 297.15 0.990 
Poros Riganiou Vtank = 0.0002A2 − 0.1569A + 89.933 0.996 
Karpenisi Vtank = −0.000007A3 + 0.0079A2 −3.0261A + 435.32 0.995 
Elati Vtank = −0.051A + 65.375 0.977 
Agnanta Vtank = −0.0425A + 51.198 1.000 

In order to investigate the effect of demand level on tank volume for each station studied, the required rainwater tank volumes in relation to collection area to fully meet five different water demand levels are presented in Figure 6. Results showed that, for all demand levels, tank volumes increase as demand increases while they decrease as collection area increases. Greater decrease occurs in the 250–350 m2 range of the collection area, while in most of the cases an increase beyond 450 m2 has a minor effect on tank volume decrease.

Figure 6

Required rainwater tank volumes in relation to rainwater collection area to fully meet five different water-demand levels of a goat–sheep livestock holding in six regions of Greece with different rainfall regimes.

Figure 6

Required rainwater tank volumes in relation to rainwater collection area to fully meet five different water-demand levels of a goat–sheep livestock holding in six regions of Greece with different rainfall regimes.

More specifically, for the high demand level (576 L/d), the tank volume decrease, for the 250–450 m2 collection area range, is 115 m3 (60.5%), 104 m3 (57.4%), 84.4 m3 (43.0%), 41.8 m3 (33.6%), 20.8 m3 (27.6%), and 21.4 m3 (23.2%) in ‘Karpenisi’, ‘Kastania’, ‘Armenoi’, ‘Poros Riganiou’, ‘Agnanta’ and ‘Elati’ stations, respectively. In the case of the low demand level (320 L/d), the tank volume decrease, for the 250–450 m2 collection area range, is 8.5 m3 (28.0%), 7.1 m3 (18.0%), 5.5 m3 (13.3%), 3.4 m3 (7.6%), 3.2 m3 (4.9%), and 0.0 m3 (0.0%) in ‘Agnanta’, ‘Elati’, ‘Karpenisi’, ‘Poros Riganiou’, ‘Armenoi’ and ‘Kastania’ stations, respectively.

Tank sizes can be reduced when expected reliability is below 100%. In Figure 7, reliability curves for three demand levels (320, 400 and 576 L/day) and three collection areas (250, 350 and 450 m2) for all six stations studied are presented. As shown in this figure, reliability values for each station increase as the collection area increases and decrease as the demand level increases with the last one playing the dominant role over the collection area. The highest values of reliability are observed in the ‘Agnanta’ curve and the lowest in the ‘Armenoi’ curve. The required tank volume decreases as the reliability decreases. So, if a satisfactory reliability level below 100% is adopted, then the tank size can be reduced to more feasible levels.

Figure 7

Reliability curves of rainwater harvesting systems in relation to tank size to meet the water demands of 320, 400 and 576 L/day using rainfall collection areas of 250, 350 and 450 m2.

Figure 7

Reliability curves of rainwater harvesting systems in relation to tank size to meet the water demands of 320, 400 and 576 L/day using rainfall collection areas of 250, 350 and 450 m2.

As shown in Figure 7, very high values of reliability (95%) can be obtained with a 10–20 m3 tank for low water demand (320 L/day), a 15–40 m3 tank for medium water demand (400 L/day) and a 25–80 m3 tank for high water demand (576 L/day), irrespective of the collection area for all stations studied, except for ‘Armenoi’. In the case of ‘Armenoi’, due to the local rainfall regime (long dry period), 95% reliability can be obtained with tank volumes of 44–48 m3, 57–63 m3 and 85–110 m3, respectively.

In all stations studied, tank volumes at 95% reliability are much lower than those at 100% (Figure 7). The gain in volume is in the range 10–90 m3. Lower gain is observed in short dry period stations and in the low demand–high area scenario (‘Agnanta’, 320 L/day, 450 m2), while the highest gain is observed in the opposite case: long dry period station and high demand–low area scenario (‘Armenoi’, 576 L/day, 250 m2).

Overall, the required tank volumes for an achievable collection area of 450 m2 and reliability value of 95% range from 10 to 50 m3 for all stations except ‘Armenoi’ where the range is from 44 to 85 m3.

CONCLUSIONS

From the presented analysis, one can conclude that the rainwater tank capacity, to meet the water needs of a livestock holding, is strongly affected by local conditions of the study area and cannot be fully standardized. The longest dry period is the dominant factor against rainfall in rainwater tank sizing, i.e., the longer the dry period, the greater the required volume of the rainwater tank.

In all areas and at all demand levels studied, for 100% reliability, the required volume of the rainwater tank was decreased by increasing the rainwater collection area. Greater decrease occurs in the 250–350 m2 range of the collection area and mainly in the high demand scenario, while in most of the cases an increase beyond 450 m2 has a minor effect on tank volume decrease. The volume reduction for the 250–450 m2 collection area ranged from 23.2% to 60.5% in the high demand scenario and from 0% to 28% in the low demand scenario.

Required tank volumes at 95% reliability are much lower than those at 100%, irrespective of the collection area. The gain in volume ranges from 10 to 90 m3. Lower gain is observed in short dry period stations and in the low demand–high area scenario, while the highest gain is observed in the opposite case: long dry period station and high demand–low area scenario.

REFERENCES

REFERENCES
Aladenola
,
O. O.
&
Adeboye
,
O. B.
2010
Assessing the potential for rainwater harvesting
.
Water Resources Management
24
(
10
),
2129
2137
.
doi: 10.1007/s11269-009-9542-y
.
Angelakis
,
A. N.
2016
Evolution of rainwater harvesting and use in Crete, Hellas, through the millennia
.
Water Science and Technology: Water Supply
16
(
6
),
1624
1638
.
doi: 10.2166/ws.2016.084
.
Basinger
,
M.
,
Montalto
,
F.
&
Lall
,
U.
2010
A rainwater harvesting system reliability model based on nonparametric stochastic rainfall generator
.
Journal of Hydrology
392
,
105
118
.
http://dx.doi.org/10.1016/j.jhydrol.2010.07.039
.
Campisano
,
A.
&
Modica
,
C.
2012a
Optimal sizing of storage tanks for domestic rainwater harvesting in Sicily
.
Resources, Conservation and Recycling
63
,
9
16
.
http://dx.doi.org/10.1016/j.resconrec.2012.03.007
.
Campisano
,
A.
&
Modica
,
C.
2012b
Regional scale analysis for the design of storage tanks for domestic rainwater harvesting systems
.
Water Science and Technology
66
(
1
),
1
8
.
doi: 10.2166/wst.2012.171
.
Campisano
,
A.
,
Butler
,
D.
,
Ward
,
S.
,
Burns
,
M. J.
,
Friedler
,
E.
,
DeBusk
,
K.
,
Fisher-Jeffes
,
L. N.
,
Ghisi
,
E.
,
Rahman
,
A.
,
Furumai
,
H.
&
Han
,
M.
2017
Urban rainwater harvesting systems: research, implementation and future perspectives
.
Water Research
115
,
195
209
.
https://dx.doi.org/10.1016/j.watres.2017.02.056
.
Cowden
,
J. R.
,
Watkins
,
D. W.
Jr
&
Mihelcic
,
J. R.
2008
Stochastic rainfall modeling in West Africa: parsimonious approaches for domestic rainwater harvesting assessment
.
Journal of Hydrology
361
,
64
77
.
http://dx.doi.org/10.1016/j.jhydrol.2008.07.025
.
Ghisi
,
E.
2010
Parameters influencing the sizing of rainwater tanks for use in houses
.
Water Resources Management
24
,
2381
2403
.
doi: 10.1007/s11269-009-9557-4
.
Ghisi
,
E.
&
Ferreira
,
D. F.
2007
Potential for potable water savings by using rainwater and greywater in a multi-story residential building in southern Brazil
.
Building and Environment
42
(
4
),
2512
2522
.
doi: 10.1016/j.buildenv.2006.07.019
.
Gikas
,
G. D.
&
Tsihrintzis
,
V. A.
2012
Assessment of water quality of first-flush roof runoff and harvested rainwater
.
Journal of Hydrology
466–467
,
115
126
.
http://dx.doi.org/10.1016/j.jhydrol.2012.08.020
.
Guo
,
Y.
&
Baetz
,
B. W.
2007
Sizing of rainwater storage units for green building applications
.
Journal of Hydrologic Engineering
12
(
2
),
197
205
.
http://dx.doi.org/10.1061/(ASCE)1084-0699(2007)12:2(197)
.
Hellenic Statistical Authority
2015
.
Imteaz
,
M. A.
,
Shanableh
,
A.
,
Rahman
,
A.
&
Ahsan
,
A.
2011
Optimisation of rainwater tank design from large roofs: a case study in Melbourne, Australia
.
Resources, Conservation and Recycling
55
,
1022
1029
.
http://dx.doi.org/10.1016/j.resconrec.2011.05.013
.
Kinkade-Levario
,
H.
2007
Design for Water: Rainwater Harvesting, Stormwater Catchment and Alternate Water Reuse
.
New Society Publishers
,
Gabriola Island
,
Canada
.
Londra
,
P. A.
,
Theocharis
,
A. T.
,
Baltas
,
E.
&
Tsihrintzis
,
V. A.
2015
Optimal sizing of rainwater harvesting tanks for domestic use in Greece
.
Water Resources Management
29
(
12
),
4357
4377
.
doi: 10.1007/s11269-015-1064-1
.
Melidis
,
P.
,
Akratos
,
C. S.
,
Tsihrintzis
,
V. A.
&
Trikilidou
,
E.
2007
Characterization of rain and roof drainage water quality in Xanthi, Greece
.
Environmental Monitoring and Assessment
127
(
1–3
),
15
27
.
doi: 10.1007/s10661-006-9254-1
.
Mendez
,
C. B.
,
Klenzendorf
,
J. B.
,
Afshar
,
B. R.
,
Simmons
,
M. T.
,
Barrett
,
M. E.
,
Kinney
,
K. A.
&
Kirisits
,
M. J.
2011
The effect of roofing material on the quality of harvested rainwater
.
Water Research
45
,
2049
2059
.
doi: 10.1016/j.watres.2010.12.015
.
Mitchell
,
V. G.
,
McCarthy
,
D. T.
,
Deletic
,
A.
&
Fletcher
,
T. D.
2008
Urban stormwater harvesting – sensitivity of a storage behaviour model
.
Environmental Modelling and Software
23
,
782
793
.
doi: 10.1016/j.envsoft.2007.09.006
.
Official Gazette of the Hellenic Republic
2004
Second issue, No. 1396
.
Palla
,
A.
,
Gnecco
,
I.
&
Lanza
,
L. G.
2011
Non-dimensional design parameters and performance assessment of rainwater harvesting systems
.
Journal of Hydrology
401
,
65
76
.
http://dx.doi.org/10.1016/j.jhydrol.2011.02.009
.
Palla
,
A.
,
Gnecco
,
I.
,
Lanza
,
L. G.
&
La Barbera
,
P.
2012
Performance analysis of domestic rainwater harvesting systems under various European climate zones
.
Resources, Conservation and Recycling
62
,
71
80
.
http://dx.doi.org/10.1016/j.resconrec.2012.02.006
.
Souza
,
E. L.
&
Ghisi
,
E.
2012
Potable water savings by using rainwater for non-potable uses in houses
.
Water
4
,
607
628
.
doi: 10.3390/w4030607
.
Tsihrintzis
,
V. A.
&
Baltas
,
E.
2014
Determination of rainwater harvesting tank size
.
Global NEST Journal
16
(
5
),
822
831
.
Ward
,
D.
&
McKague
,
K.
2007
Water Requirements of Livestock. OMAFRA Factsheet Order No. 07-023
,
Ontario Ministry of Agriculture, Food and Rural Affairs, Canada
.
Yannopoulos
,
S.
,
Antoniou
,
G.
,
Kaiafa-Saropoulou
,
M.
&
Angelakis
,
A. N.
2016
Historical development of rainwater harvesting and use in Hellas: a preliminary review
.
Water Science and Technology: Water Supply
17
(
4
),
1022
1034
.
doi: 10.2166/ws.2016.200
.
Yaziz
,
M. I.
,
Gunting
,
H.
,
Sapari
,
N.
&
Ghazali
,
A. W.
1989
Variations in rainwater quality from roof catchments
.
Water Research
23
(
6
),
761
765
.
https://doi.org/10.1016/0043-1354(89)90211-X
.