3.1. The rainfall distribution
The present research article utilizes daily rainfall data obtained from 157 rain monitoring stations for the period between 1987–2018. The data were obtained from the Ministry of Water and Irrigation and were compiled into 12 groups based on their locations to represent rainfall distribution in twelve Jordanian governorates. The average monthly rainfall distribution in the study area is presented in Table 3. It is worth noting that the rainy season in Jordan lasts from October to May, with peak rainfall occurring during January and February, except for the Aqaba Governorate, where the rainy season lasts from April to July. The annual rainfall rates were found in the Ajlun, Jarash, Balqa, and Amman governorates, with precipitation rates of 471, 386, 396, and 283 mm, respectively. These data suggest that the four governorates have considerable potential for rainwater harvesting.
Table 3
Temporal distribution of mean monthly rainfall in the Jordanian Governorates.
Governorate
|
Temporal distribution of mean monthly rainfall (1987–2018) (mm/month)
|
Oct
|
Nov
|
Dec
|
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Ajlun
|
12
|
46
|
110
|
123
|
92
|
66
|
16
|
6
|
0
|
0
|
0
|
0
|
Irbid
|
10
|
36
|
77
|
91
|
86
|
55
|
19
|
5
|
0
|
0
|
0
|
0
|
Jarash
|
6
|
35
|
85
|
95
|
93
|
58
|
11
|
3
|
0
|
0
|
0
|
0
|
Mafraq
|
5
|
15
|
30
|
39
|
32
|
20
|
5
|
15
|
9
|
0
|
13
|
0
|
Amman
|
7
|
23
|
60
|
71
|
63
|
37
|
9
|
13
|
0
|
0
|
0
|
0
|
Balqa
|
7
|
37
|
90
|
87
|
89
|
64
|
13
|
9
|
0
|
0
|
0
|
0
|
Madaba
|
4
|
20
|
51
|
70
|
68
|
35
|
11
|
3
|
0
|
0
|
0
|
0
|
Zarqa
|
4
|
13
|
29
|
35
|
32
|
18
|
6
|
5
|
0
|
0
|
0
|
0
|
Aqaba
|
2
|
5
|
3
|
10
|
6
|
10
|
1
|
31
|
24
|
0
|
27
|
0
|
Karak
|
3
|
16
|
47
|
63
|
58
|
46
|
11
|
3
|
0
|
0
|
0
|
0
|
Ma’an
|
6
|
7
|
18
|
32
|
29
|
20
|
7
|
14
|
10
|
10
|
10
|
0
|
Tafilah
|
5
|
17
|
38
|
53
|
47
|
32
|
10
|
2
|
0
|
0
|
0
|
0
|
3.2. Population Growth and water consumption
The population distribution data and analysis of the population growth rate in Jordan are important for understanding the current situation. The 2020 census conducted by the Jordanian Department of Statistics indicates that Jordan’s total population was 10.8 million (“Jordan Statistical Yearbook 2020 – Department of Statistics,” 2020). As shown in Table 4, the population is distributed across the governorates, with approximately 63.4% of the total population living in the Central Region (Amman, Al-Zarqa, Madaba, Al-Balqa), 28.7% in the Northern Region (Irbid, Mafraq, Jarash, Ajlun), and 7.9% in the Southern Region (Ma’an, Kark, Al Tafeila and Aqaba). The availability of water plays a crucial role in determining the settlement patterns in Jordan, with the majority of the population clustered in the relatively more water-rich Central Region.
3.3. The total of rooftop areas
Table 4. lists the total roof area of each governorate. Amman Governorate has the highest roof area (34.2 ×106 m2); the lowest roof area (1.42 ×106 m2) is in Tafeila Governorate. The calculated total roof area values were used to estimate the volume of rainwater that could be harvested in each governorate.
3.4. Water consumption rate
The average water consumption per capita (liter/capita/day) for the 12 governorates was given in Table 4. along with the average annual rainfall rate, total roof area, and population according to the 2020 census. The average consumption rates per capita ranged from 80 L in Irbid to 274 L in Ma’an, indicating a significant variation in water consumption rates per capita in different governorates. The total annual per capita use in a governorate or Total Amount per Governorate (TAPG) was calculated using Eq. 1, and the results of these calculations are presented in Table 4. TAPG values were affected by two main factors: the population and water consumption per capita. Amman had the highest TAPG of 218 Mm3 since its population was the highest (4,536,500), but it had low water consumption per capita compared to other governorates like Ma’an. Ma’an had a low TAPG (18 Mm3) compared to Amman, but the water consumption per capita was relatively higher in Ma’an. The same phenomena could be seen again when Ajlun and Irbid were compared. The water consumption per capita value was 106 L/day in Ajlun, and it was 80 L/day in Irbid. Nevertheless, Irbid’s TAPG was 7.5 times that of Ajlun; because the population
density in Irbid was more than ten times greater compared to Ajlun. Therefore, the size of the population in a governorate was the main factor affecting the TAPG value compared to the water consumption per capita. However, a high TAPG value did not reflect a higher water consumption per capita.
Table 4
Rainfall, Total rooftop areas, Demographic Data Total water consumption, Potential water harvesting, and potential saving for the 12 Jordanian Governorates.
Governorate
|
Annual rainfall (mm/year)
|
Roof Area (m2)
|
Population (2020)
|
% of the total population
|
Per capita water use (l/d/capita)
|
Potential water harvesting (Mm3/year)
|
Total water consumption (Mm3/year)
|
Water-saving (%)
|
Ajlun
|
471
|
1,838,920
|
199,400
|
1.8
|
106
|
0.693
|
7.7
|
9
|
Amman
|
283
|
34,248,500
|
4,536,500
|
42.1
|
132
|
7.754
|
218
|
3.6
|
Aqaba
|
80
|
1,606,400
|
213,000
|
2
|
191
|
0.103
|
14.8
|
0.7
|
Balqa
|
396
|
6,028,300
|
556,600
|
5.2
|
239
|
1.91
|
48.5
|
3.9
|
Irbid
|
379
|
27,879,500
|
2,003,800
|
18.5
|
80
|
8.453
|
58.4
|
14.5
|
Jarash
|
386
|
2,716,900
|
268,300
|
2.5
|
110
|
0.839
|
10.7
|
7.8
|
Karak
|
230
|
4,646,500
|
358,400
|
3.3
|
192
|
0.855
|
25
|
3.4
|
Ma'an
|
110
|
1,506,600
|
179,300
|
1.7
|
274
|
0.133
|
18
|
0.7
|
Madaba
|
262
|
2,638,600
|
214,100
|
2
|
160
|
0.553
|
12.4
|
4.5
|
Mafraq
|
183
|
7,036,400
|
622,500
|
5.8
|
140
|
1.03
|
31.8
|
3.2
|
Tafilah
|
204
|
1,417,200
|
109,000
|
1
|
244
|
0.231
|
9.7
|
2.4
|
Zarqa
|
152
|
9,792,900
|
1,534,577
|
14.2
|
120
|
1.191
|
67.7
|
1.8
|
3.5. The potential water harvesting
Table 4. shows estimated potential water harvesting volumes in Jordan’s 12 governorates, along with PSP values. The potential amount of rainwater harvesting (VR) is estimated based on the total surface area of the roofs, average annual rainfall rate, and runoff coefficient using Eq. 3 (Abdulla and Al-Shareef, 2009). The potential saving percentage (PSP) was calculated by dividing the potential harvested rainfall volume by annual water demand using Eq. 4. The total potential for rooftop rainwater harvesting was estimated at 23.74 Mm3 per year. Comparing the governorates, Irbid was ranked first with a harvesting potential of 8.453 Mm3/year, followed by Amman, which had a water harvesting potential of 7.754 Mm3/year. The VR values of the other ten governorates were markedly lower compared to these top two governorates. The total rooftop area to harvest rainwater in Irbid was high, and this governorate has a higher average rainfall rate compared to other governorates.
The precipitation rates in Amman were not the greatest among the 12 governorates, but it had the largest total rooftop area. The lowest possible water harvesting rate of 0.103 Mm3/year was calculated for Aqaba, which had a low rooftop area and the lowest annual rainfall rate. In Jordan, the percentage of the domestic annual water requirements that can be met by rooftop rainwater harvesting was presented as a percentage of water-saving (PSP). The highest PSP of 14.5% was calculated for Irbid, suggesting that 14,5% of the total consumption could be covered in Irbid. PSP for the whole of Jordan was 4,5%.
Other studies have found similar results. In countries with low rainfall rates, it was reported that 3.4–8.5% of the domestic annual water requirements could be met through rainwater harvesting (Abu-Zreig et al., 2013; Baby et al., 2019; Şahin and Manioğlu, 2019; Summerville and Sultana, 2019). Jordan’s total rooftop rainwater harvesting potential of 23.74 Mm3/year was equivalent to 4.5% of the total domestic water demand in 2020. The total potential volume calculated here was significantly higher than the results from other older studies conducted in Jordan. This potential volume increase was most likely due to Jordan’s urban expansion throughout the last 15 years, reflecting the increase in the total rooftop area. Based on 2004 data, Abu-Zreig et al. (2013) reported that the national potential for rooftops was 14.7 Mm3/year corresponding to 6% of the total domestic water demand (Abu-Zreig et al., 2013). Abdulla and Al-Shareef (2009) found that the potential rooftop harvested rainwater was 15.5 Mm3 based on 2005 data, equivalent to 5.6% of the total domestic water demand (Abdulla and Al-Shareef, 2009).
The potential of rooftop harvesting volume varies from country to country, depending on the rainfall rate and rooftop area. Shadeed and Alawna (2021) found that the potential rainwater harvesting in the West Bank of Palestine is 37 Mm3/year (Shadeed and Alawna, 2021). Traboulsi and Traboulsi (2015) found that the potential rainwater harvesting from rooftops was 23 Mm3/year in Lebanon (Traboulsi and Traboulsi, 2017). Mourad and Berndtsson (2011) reported that the potential rainwater harvesting from urban areas in Syria was 35 Mm3/year (Mourad and Berndtsson, 2011).
3.6. The optimal tank size and cost
By utilizing the Ripple Method, Fig. 3 illustrates the most favorable tank sizes that can be constructed based on interrelated factors, including the rooftop area, monthly water consumption, and monthly rainfall rate, as represented by Eq. 5. The findings reveal the optimum tank sizes for rainwater harvesting across varying rooftop areas within four governorates in Jordan. The optimal tank size for each rooftop area is the minimum value between the cumulative rainfall volume (D-VR) and the harvested rainwater volume (VR).
In Ajlun Fig. 3-(a), the optimal tank size increases with an increase in rooftop area, from 19 m3 for an area of 50 m2 to 75 m3 for an area of 200 m2. The tank size is determined based on the amount of harvested water (VR). Then, the tank size decreases from 61 m3 for a rooftop area of 250 m2 to 8 m3 for an area of 500 m2, based on the Cumulative (D-VR). Therefore, it is important to consider both factors when determining the optimal tank size for rainwater harvesting in Ajlun.
In Irbid Fig. 3-(b), the optimal tank size increases with an increase in rooftop area, from 15 m3 for 50 m2 to 61 m3 for 200 m2. The tank size is determined based on the amount of harvested water (VR). Then, the tank size decreases from 54 m3 for a rooftop area of 250 m2 to 7 m3 for an area of 500 m2, based on the Cumulative (D-VR). Therefore, it is important to consider both factors when determining the optimal tank size for rainwater harvesting in Irbid.
In Jarash Fig. 3-(c), the optimal tank size increases with an increase in rooftop area, from 15 m3 for 50 m2 to 82 m3 for 300 m2. The tank size is determined based on the amount of harvested water (VR). Then, the tank size decreases from 66 m3 for a rooftop area of 350 m2 to 20 m3 for an area of 500 m2, based on the Cumulative (D-VR). Therefore, it is important to consider both factors when determining the optimal tank size for rainwater harvesting in Jarash.
In Amman Fig. 3-(d), the results were different, where the optimal tank size depends on the amount of harvested water (VR) from a rooftop area of 50 m2 to 450 m2, which are 11 m3 and 102 m3, respectively. As for houses with rooftop areas larger than 450 square meters, determining the optimal tank size relies on the Cumulative (D-VR).
The results also show that the optimal tank sizes for the remaining eight governorates (see Appendix A.3) equal the volume of rainwater (VR) that can be harvested for the whole rooftop area. This suggests that the cumulative rainfall volume (D-VR) was not large enough to influence the optimal tank size, and hence the optimal tank size was determined solely by the volume of rainwater that can be harvested (VR) for the whole rooftop area. Although a previous study found that harvested water was compatible with ‘Jordanian Drinking Water Standards’. However, best management practices should ensure that harvested rainwater is pathogen-free (Assayed et al., 2013). The harvested rainwater must be treated for drinking and cooking; it can also be used in agriculture, cleaning, and flushing without treatment (Abdulla and Al-Shareef, 2009) the harvested rainwater can be stored in tanks separate from existing tap water tanks and used in flushing.
The Ripple method has wide applicability across all governorates. Figure 4 was generated using the Ripple method and illustrates the optimal tank sizes for Ajlun, taking into account rooftop areas ranging from 50–500 m2 and monthly water consumption rates of 2–22 m3/month (The calculations in the Appendix A.5). Multiple possible consumption rates were considered to accommodate the wide range for a single family, as described by Abu-Zreig et al. (2019). This figure will prove useful for future research investigating how rainwater can be utilized and treated to meet monthly demand rates. Rainwater harvested using this method can be used directly for irrigation and flushing, as previously established by Assayed et al. (2013), Awawdeh et al. (2011), and Matos et al. (2014).
Rainwater harvested from rooftops may contain air-borne pollutants and contaminants such as dust, foliage, and dead birds, depending on the rooftop’s surface conditions and the location of the rainfall. Harvested rainwater is best suited for garden irrigation and flushing without prior treatment to avoid health risks and treatment costs while reducing pressure on other water sources. Even if the harvested water is only used for toilet flushing, a significant amount of tap water can be conserved, especially when the rooftop area and rainfall rate are sufficient.According to Sha’ban et al. (2011), water used for flushing comprises 17.7% of total water usage in Jordan (Sha’ban et al., 2012).
This study utilized the Water Balance Model to compute the ideal tank capacity for fulfilling flushing needs. The optimal tank size for every governorate was determined by constructing Fig. 5 using the Water Balance Model, and these figures can be employed to identify the perfect tank capacity based on flushing water requirement and rooftop area (50–500 m2) (see Appendix A.4.).
According to Fig. 5-(a), the rainwater that can be collected has the potential to fulfill the flushing requirements for all rooftop areas (50–500 m2) in Irbid and Ajlun. This means that the collected rainwater can be used to meet the flushing needs in Ajlun and Irbid, with the optimal tank sizes being 2.7 m3 and 2 m3 per household, respectively. In Jarash, the rainwater collected from a rooftop area of 100 m2 can meet the flushing requirements, but only 2 m3/month per household can be collected from a rooftop area of 50 m2. In Mafraq Governorate, the flushing needs can be partially fulfilled for households with a rooftop area of 150 m2; and completely met for those with a rooftop area of 200 m2 or more. A notable disparity was observed in the PSP values of Irbid (14.5% and 17.7%) due to the fact that in the former case, the PSP was calculated based on the actual population of Irbid. In contrast, in the latter case, a PSP of 17.7% (sufficient to meet the flushing consumption fully) was estimated using the average family size per household in Jordan (4.8 members). This discrepancy implies that the average number of family members in Irbid was higher than the national average in Jordan. For houses in Amman, Balqa, and Madaba with a rooftop area of 100 m2, the potential harvested rainwater could partially fulfill the flushing consumption. However, the entire flushing consumption could be met for households with a roof area of 150 m2 or more (Fig. 5-(b)). In Zarqa, households with a surface area of 200 m2 and more were the only ones able to access water for flushing. On the other hand, for households in the middle governorates with a roof area of 50 m2, the potential harvested rainwater could not meet the required amount for flushing. Figure 5-(c) illustrates that households in the Southern Governorates had the lowest performance compared to those in the Middle and Northern Governorates. In Ma’an, the entire flushing consumption could only be met for households with rooftop areas of 450 m2 or more. In Aqaba, households with roof areas of 350 m2 and above could
fully meet their flushing consumption through rainwater harvesting. In Tafiela, households with 250 m2 or more roof areas could fully meet their flushing consumption, while in Karak, households with 150 m2 or more roof areas could do so. The researchers used Eq. 9 to calculate the monetary value of the estimated water savings from rooftop rainwater harvesting based on the cost of groundwater extraction (0.35$/m3) and seawater desalination (1.4$/m3). The results showed that the estimated water savings across Jordan could range from 170 million to 678 million dollars annually.
3.7. Regularization study for a hypothetical building in Swieleh district of Amman governorate
Table 5
Flushing consumption of a house in Sweilih District with a 300 m2 rooftop area and a family of six members..
Months
|
Average monthly rainfall (mm /month)
|
Monthly flushing consumption (m3 /month)
|
Rooftop area (m2)
|
VR (m3)
|
Vi (m3)
|
Column 5 -
Column
3 (m3)
|
Vf (m3)
|
Oct
|
12
|
4.2
|
300
|
2.88
|
0
|
-1.32
|
-1.32
|
Nov
|
44
|
4.2
|
300
|
10.56
|
-1.32
|
6.36
|
5.04
|
Dec
|
93
|
4.2
|
300
|
22.32
|
6.36
|
18.12
|
24.48
|
Jan
|
117
|
4.2
|
300
|
28.08
|
18.12
|
23.88
|
42
|
Feb
|
106
|
4.2
|
300
|
25.44
|
23.88
|
21.24
|
45.12
|
Mar
|
60
|
4.2
|
300
|
14.4
|
21.24
|
10.2
|
31.44
|
Apr
|
14
|
4.2
|
300
|
3.36
|
10.2
|
-0.84
|
9.36
|
May
|
4
|
4.2
|
300
|
0.96
|
-0.84
|
-3.24
|
-4.08
|
Al-Qawasmi (2021) recommended that the optimal tank size should be estimated for single houses and villas since the installation of large storage tanks is usually not possible for multistory buildings in densely packed urban areas due to a lack of space between buildings. Even if required land for a large storage tank is available, installing a harvesting system in those apartment buildings is usually not practical due to the low rooftop area per person. In order to present results for more accurate calculations, a hypothetical case was regularized. Rainfall data from a single station was used instead of an average calculated for an entire governorate area using multiple stations. A family of six living in the Swieleh District of Amman Governorate (there is a precipitation measuring station in Swieleh) in a single house with 300 m2 of rooftop area was selected as the case. The cost of this family’s optimal tank has been estimated to meet the consumption only for flushing toilets in the rainy season from October to May using Eq. 3. The toilet flushing consumption was accepted to be equal to 17.7% of the total average consumption per family. The total monthly demand, in this case, was calculated using Eq. 6, and the family size was taken as 6 instead of 4.8. The consumption due to toilet flushing was estimated using Eq. 7. Based on the calculations, the total water demand was estimated at 23.76 m3/month per family, and the flushing consumption was estimated at 4.2 m3/month. The optimal tank size was estimated using the Water Balance Model. The results of the method are presented in Table 5. Vi value in October was assumed zero; the tank at the beginning of October was assumed empty. The maximum value of Vf in column (8) was 45.12 m3, and the flushing consumption (Df ) in Ajlun was 4.2 m3. According to Eq. 8 (Vf > Df, then S = Df ), the optimal tank size for flushing use (S) was determined to be 4.2 m3 per month.
Table 6
The payback results of flushing water consumption in Sweilih District for a 300 m2 rooftop area with a family of six members.
Rainwater harvesting system cost (JOD)
|
Cost (JOD/m3)
|
Annual flushing consumption (m3)
|
Payback Period (Years)
|
Maximum
|
798
|
1.00
|
33.6
|
P(FG)=
|
24
|
0.25
|
33.6
|
P(FG)=
|
95
|
Minimum
|
419
|
1.00
|
33.6
|
P(AD)=
|
12.5
|
0.25
|
33.6
|
P(AG)=
|
50
|
The cost of the tank was estimated using Eq. 10 (Tank Cost = 4.2 m3 * 95 JOD/m3 = 399 JOD). The minimum cost for installing the rainwater harvesting system was estimated as 419 JOD, and the maximum cost was 798 JOD. The payback period of the rainwater harvesting system was predicted using Eq. 11, and P(FG), P(FD), P(AG), and P(AD) were calculated (Table 6). The payback period for the selected case in Sweilih ranged between 12.5–90 years based on groundwater extraction cost and 50–95 years based on seawater desalination cost. For this regularized case, PSPr was the ratio of the total flushing consumption over the potential harvested rainwater in the rainy season between October and May (see Section 2.9). The PSPr for this case was estimated at 321%, indicating that installing a rainwater harvesting system to meet the flushing consumption was highly recommended.