Journal of Hydrology 498 (2013) 287291

Contents lists available at SciVerse ScienceDirect

Journal of Hydrology
journal homepage: www.elsevier.com/locate/jhydrol

Enhanced drinking water supply through harvested rainwater treatment

Vincenzo Naddeo, Davide Scannapieco , Vincenzo Belgiorno
Sanitary Environmental Engineering Division SEED, Dept. of Civil Engineering, University of Salerno, Via Ponte don Melillo, 84084 Fisciano, SA, Italy

a r t i c l e

i n f o

Article history:
Received 14 December 2012
Received in revised form 3 June 2013
Accepted 8 June 2013
Available online 17 June 2013
This manuscript was handled by Laurent
Charlet, Editor-in-Chief, with the assistance
of P.J. Depetris, Associate Editor
Keywords:
Developing countries
Drinking water treatment
Household water supply
Rainwater
Small communities
Water management

s u m m a r y
Decentralized drinking water systems represent an important element in the process of achieving the
Millennium Development Goals, as centralized systems are often inefcient or nonexistent in developing
countries. In those countries, most water quality related problems are due to hygiene factors and pathogens. A potential solution might include decentralized systems, which might rely on thermal and/or UV
disinfection methods as well as physical and chemical treatments to provide drinking water from rainwater. For application in developing countries, decentralized systems major constraints include low cost,
ease of use, environmental sustainability, reduced maintenance and independence from energy sources.
This work focuses on an innovative decentralized system that can be used to collect and treat rainwater
for potable use (drinking and cooking purposes) of a single household, or a small community. The experimented treatment system combines in one compact unit a Filtration process with an adsorption step on
GAC and a UV disinfection phase in an innovative design (FAD Filtration Adsorption Disinfection). All
tests have been carried out using a full scale FAD treatment unit. The efciency of FAD technology has
been discussed in terms of pH, turbidity, COD, TOC, DOC, Escherichia coli and Total coliforms. FAD technology is attractive since it provides a total barrier for pathogens and organic contaminants, and reduces turbidity, thus increasing the overall quality of the water. The FAD unit costs are low, especially if compared
to other water treatment technologies and could become a viable option for developing countries.
2013 Elsevier B.V. All rights reserved.

1. Introduction
Decentralized approaches to water supply issues have been already successfully applied in many parts of developing and transition countries. These decentralized solutions deal with both
quality and availability problems and include the direct use of
alternative water sources (groundwater, rivers or rainwater),
household-based water treatment units, dual tap water systems
and distribution and sale of ready-to-use treated water (Gadgil,
1998; Mintz et al., 2001; Belgiorno and Napoli, 2000).
As water shortages occur more often, the search for alternative
water sources and ways to promote its rational use is relevant not
only to water-stressed regions but also to secure a stable water
supply that allows for rising water demand, rapid urbanization
and climate change (Ghisi et al., 2006; Villareal and Dixon, 2005;
Mun and Han, 2012; Belgiorno et al., 2013; Naddeo et al., 2013).
In some semi-arid areas of the world, rainwater harvesting has
been promoted for a long time as a useful technology, able to provide local settlements with water. For example, in 50% of the Tanzania area, people completely rely on rainwater for their survival
(Mbilinyi et al., 2005).

Corresponding author. Tel.: +39 089969337; fax: +39 089969620.

E-mail address: dscannapieco@unisa.it (D. Scannapieco).
0022-1694/$ - see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jhydrol.2013.06.012

The efciency and feasibility of rainwater-based supply systems

have been studied by many authors (Eroksuz and Rahman, 2010;
Ghisi et al., 2007; Ghisi and Ferreira, 2007; Jones and Hunt,
2010; Khastagir and Jayasuriya, 2010; Li et al., 2010; Rahman
et al., 2012). Domnech and Saur (2011) argue that for this kind
of treatment units the economic feasibility can be determined
through a detailed analysis of the end-user needs, usually restricted to a few options. In addition, in water stressed areas the
economic feasibility threshold value tends to be lower.
Rainwater harvesting and treatment provides water directly to
households allowing family members to have full control of their
own water system, which greatly reduces centralized operation
and maintenance costs. There are also examples of rainwater harvesting systems developed for entire settlements, in which water is
withdrawn from roads or elds (Gould and Nissen-Petersen, 1999).
The main disadvantages of rainwater harvesting are the dependence on rainfall seasonal variability, the uncertainty of precipitations and also the rainwater quality, which is characterized by a
uctuating behaviour; in addition, diseases may spread over a
community since rainwater has to be stored, sometimes for a long
period. Several techniques used to collect precipitation runoff over
roads, elds or roofs after dry periods may provide nal users with
contaminated water supply due to deposited pollutants that are
ushed away during precipitation (Zhu et al., 2004).

288

V. Naddeo et al. / Journal of Hydrology 498 (2013) 287291

Several technologies have already been explored or used, alone

or in combination, as end-of-pipe systems to treat the fraction of
rainwater that is to be used for drinking purposes (Sobsey, 2002).
Some of these methods, such as boiling water, are traditionally
and widely used, although they may not always be the optimal
solution (Mintz et al., 2001) in terms of nancial issues as well
as nal water quality.
Solar water disinfection (SODIS) is a basic technology used to
improve the microbiological quality of drinking water by using
solar radiation as to destroy pathogens (Mintz et al., 2001). A
potential limitation of SODIS, besides its dependence on sunlight
for disinfection, is that the treatment process is rather complex.
UV irradiation with lamps has raised renewed interest in recent years because of its well-documented ability to extensively
(>99.9%) inactivate two waterborne, chlorine-resistant protozoans, Cryptosporidium parvum oocysts and Giardia lamblia cysts, at
relatively low irradiation doses. However, UV lamp disinfection
has some disadvantages when used as a drinking water disinfectant at household level. Organic matter, turbidity and certain dissolved contaminants can interfere with or reduce the efciency of
microbial inactivation. These lamps require periodic cleaning,
especially if placed in a submerged conguration; moreover, they
have a nite lifespan and must be periodically replaced (Gadgil,
1998).
Chemical treatment is widely used for disinfection purposes at
full scale facilities. Of the drinking water disinfectants, free chlorine is the simplest, most widely used and the most affordable
one. It is highly effective against nearly all water-borne pathogens,
with the notable exception of C. parvum oocysts and the Mycobacteria species (Sobsey, 2002; Mintz et al., 2001). However, the sociocultural acceptance of disinfection with chlorine-containing reagents is low in some cases, due to taste and odour impact problems (Murcott, 2005); in addition, chlorine gas storage poses
signicant health risks and is therefore used only at large water
facilities
Slow sand ltration has been adapted for household use and is
also known as Biosand ltration (BSF). Biosand lters are tanks
lled with sand in which a bioactive layer is allowed to grow as
a means of eliminating disease-causing microorganisms. Several
studies show that BSF removes bacteria consistently, if not completely, on average in the range 81100%, and protozoa by
99.98100%. However, these lters have limited viruses removal
efciency (Lantagne et al., 2007; Naddeo and Belgiorno, 2007).
Furthermore, paper, ber or fabric lters may be applied at
household level. They can be effective in the removal of larger
water-borne pathogens such as free-swimming larval forms (cercariae) of schistosomes and Faciola species, guinea worm larvae within
their intermediate crustacean host (Cyclops), and bacterial pathogens associated with relatively large copepods and other zooplankton in water, such as the bacterium Vibrio cholerae (Sobsey, 2002).
However, these lters are not recommended for the treatment of
household water supply because their pores are too large to significantly retain viruses, bacteria and smaller protozoan parasites
(Sobsey et al., 2008).
Activated carbon lters, often in the form of pressed blocks, followed by UV disinfection or pre-coated with silver (Ag), are used as
table-top units for additional tap water treatment (Abbaszadegan
et al., 1997). However, they have a only limited operating life (6
8 months) and relatively high costs.
This work focuses on an innovative decentralized system to collect and treat rainwater for potable use (drinking and cooking) of a
single household or a small community. The tested unit is composed of a Filtration phase, followed by Adsorption on Granular
Activated Carbon (GAC) and UV Disinfection, in an innovative design (FAD Filtration Adsorption Disinfection).

2. Materials and methods

2.1. FAD treatment unit
The FAD system has been reported in Fig. 1. All tests have been
carried out using a full scale FAD treatment unit at xed ow rate
(30 L/h) on harvested rainwater for a total treatment time of 25 h.
The unit is composed of two separate elements, and it can work
both with and without pumping. The pre-ltration unit is provided
with a membrane, characterized by a porosity of 75 lm. The FAD
unit combines adsorption on GAC, microltration at 0.5 lm and
UV disinfection. The microltration step is located immediately
after GAC treatment and there the feed is exposed to UV light irradiation (Fig. 1), therefore this zone is hereafter referred as hybrid
FAD zone. UV light is produced by a 15 W low pressure lamp made
of hard quartz glass.
The rain has been harvested from the rooftop of a small building
located in the experimental wastewater treatment station at the University of Salerno (Italy) in a conventional water tank of 2500 L, during the period ranging from December 2008 to May 2009. The water
was then treated on site with the FAD unit. The collected rainwater
was stored in the tank for no more than 19 consecutive days.
Tests have been carried out with and without UV light. In order
to compare efciencies, another set of experiments has been completed, using only the GAC adsorption step. The latter experimental
setup has been characterized by identical operating conditions of
the FAD unit in terms of GAC volume and available surface for
adsorption; it has been housed into the twin reactor, in which both
the membrane and the UV lamp were removed.
2.2. Analytical methods
Analytical measurements were made at the Environmental
Engineering Laboratory of the University of Salerno, Fisciano (SA),
Italy. The USEPA Standard Methods have been successfully used
for microbiological analyses. Acetate cellulose lters (0.45 lm pore
size, Millipore, USA) were employed for sample ltration while mendo medium (Oxoid, Italy) and Tryptone Bile X-Glucuronide (TBX)
medium (Oxoid, Italy) were used respectively for Total coliform
and Escherichia coli retention.
The results were expressed in terms of Colony Forming Units
per 100 mL (CFU/100 mL). Absorbance measurements were performed using a k12 UVVis spectrophotometer (Perkin Elmer,
USA). Turbidity was detected using a turbidimeter (Model
2100 N, Hach Lange AG, Germany). Total Organic Carbon (TOC)
was determined using a Shimadzu TOC-5000A analyzer. Chemical
Oxygen Demand (COD) and Total Suspended Solids (TSS) were
measured following the AWWAAPHAWEF Standard Methods
(AWWAAPHAWEF, 1998). The measurement of pH, conductivity
and redox potential were carried out using a multiparametric
probe (Hanna Instruments, USA).
3. Results and discussion
3.1. Rainwater characterization
During the harvesting period, rainwater quality has been constantly monitored, and results showed a fair stability in terms of
chemical-physical parameters, as reported in Table 1 and accordingly plotted in Fig. 3. High turbidity and a consistent concentration of organic matter have been found; this is in accord with
other previously published studies (Gould and Nissen-Petersen,
1999; Zhu et al., 2004). Finally, pH ranged from 5.5 to 7.1, with
an average value of 6.3.

289

V. Naddeo et al. / Journal of Hydrology 498 (2013) 287291

Fig. 1. Experimental setup of the FAD unit (left) and its section.

Table 1
Comparison of pollutants removal by the FAD process and the separate conventional GAC adsorption unit.
Analytical parameters

Raw water

Turbidity (NTU)
COD (mg/L)
TOC (mg/L)
DOC (mg/L)
UV254 (cm 1)
Escherichia coli (CFU/100 mL)
Total coliforms (CFU/100 mL)

25.88 3.62
4.79 0.56
5.952 0.71
5.398 0.517
0.086 0.008
328 126
152 32

FAD process

Separate GAC

Efuent

Removal (%)

Efuent

Removal (%)

0.42 0.06
2.39 0.35
3.383 0.43
3.383 0.436
0.052 0.003
00
01

98.3 1.6
49.9 7.5
42.8 6.9
37.3 5.9
38.3 6.7
99.99 0.0
99.99 0.0

2.13 0.42
3.70 0.63
4.278 0.49
3.169 0.419
0.060 0.005
108 92
61 29

59.7 12.7
22.8 8.7
27.8 7.1
29.3 6.0
29.9 4.7
67.9 14.3
59.8 19.1

Coliforms are often employed as the microbiological parameter

for the pathogenic safety of treated water. In the FAD efuent,
the concentration of Total coliforms was equal to 0 0 CFU/
100 mL, while in raw rainwater the same concentration attained
the value of 328 126 CFU/100 mL (Fig. 2). The pre-ltration unit
membrane nominal pore size was bigger than the average diameter of coliforms, therefore the complete removal of Total coliforms in
the FAD process might be attributed to both the separation effect of
the GAC lter and disinfection-oxidation effect of the UV lamp. Disinfection efciency was determined through E. coli monitoring as
well. A total removal of E. coli by the FAD process is also reported
in Table 1, which can be compared with the 151 32 CFU/100 mL
value in raw water (Fig. 2).
Fig. 2. Bacteriological raw rainwater characterization in terms of Total coliforms
and Escherichia coli (respectively, from the lower to higher value in terms of
detected colonies: min, 25 percentile, average, 75 percentile, max).

Water quality has also been monitored in terms

form and E. coli, which signicantly scattered over
This behaviour can be imputable to the presence of
the rooftop; their concentration was considerably
precipitation occurred after several dry days.

of Total colitime (Fig. 2).

bacteria over
higher when

3.2. Turbidity and bacteria removal

Turbidity is often used to represent the presence of particles in
water, which is a major parameter of water quality in drinking
water treatment. During the experiments, the turbidity in the
raw water was 25.88 3.62 NTU, on average. A moderate removal
efciency of 22.7 8.7% was achieved by the pre-ltration unit, so
that 20.33 2.16 Nephelometric Turbidity Units (NTU) level still
remained in the efuent. However, the FAD unit has proven itself
to be able to reduce turbidity as much as to 0.42 0.06 NTU, that
can be attributed to both the GAC adsorption and microltration
steps. Approximately 99.9% of the feed suspended solids has been
removed in the process.

3.3. Organic matter removal

Organic matter can be divided into a Particulate Organic Matter
(POM) and Dissolved Organic Matter (DOM) fraction. POM could be
easily removed even through conventional water treatment processes. On the other hand, DOM is one of the most critical concerns
in drinking water treatment, since this fraction is potentially hazardous and difcult to be eliminated (Sobsey, 2002).
The performance of the FAD process in terms of DOM removal
as DOC and UV254 was investigated and illustrated in Fig. 3. It
can be seen that Dissolved Organic Carbon (DOC) in the raw water
was reduced by 26.3 6.0% in the FAD unit without any UV lamp;
on the other hand, the total removal efciency increased to
37.3 5.9% thanks to the UV irradiation and the subsequent oxidation effect. As for UV254, the recorded removal efciency is equal to
29.9 4.7% and 38.3 6.7% after the treatment, without and with
UV irradiation, respectively. Thus, it can be concluded that the
inuent DOM is mainly removed through GAC adsorption.
The contribution of UV irradiation over the Natural Organic
Matter (NOM) removal can be attributed to oxidative effects on
dissolved organic matter and to the disinfection process that takes
place in the reactor. On the one hand, although the raw rainwater

290

V. Naddeo et al. / Journal of Hydrology 498 (2013) 287291

Fig. 3. Removal of DOC (a), TOC (b), SCOD (c) and UV254 (d) by the FAD process.

had already gone through the pre-ltration step, there still was a
remaining fraction of DOM. On the other, the GAC adsorption surface and microltration layer in the FAD reactor can provide an
excellent surface for the attachment of microbial communities
(Guo et al., 2008), thus enhancing the DOM removal through biodegradation of the adsorbed DOM by the microorganisms on GAC
surface. Consequently, the UV irradiation still contributed to
10.0% of the DOC and 8.4% of the UV254 removal even after GAC micro-ltration, as discussed above.
COD and TOC are widely used in water treatment as surrogate
parameters to represent the content of organic matter. As illustrated in Fig. 3, COD and TOC were removed by 22.8 8.7% and
27.8 7.1% as well as 49.9 7.5% and 42.8 6.9% after the treatment without and with UV irradiation, respectively. There was still
POM in the GAC efuent with an average concentration of
0.309 mg/L as TOC, which was probably due to the presence of biolm residues developed over the GAC surface. However, the POM
concentration can be easily reduced by the micro-ltration-layer
and UV oxidation if placed in the same unit.
A synergetic effect has been observed among the GAC, microltration and UV irradiation in the FAD process referring to the dissolved and total organic matter removal simultaneously, i.e. the
GAC was able to remove a large amount of DOM in water, the
downstream microltration further eliminated the POM and nally
UV irradiation provided a complete disinfection with a stable fouling formation on the quartz glass, which is dependent on the organic matter concentration of the feed (Liu et al., 2002; GurReznik et al., 2008). As a result, the total organic matter in raw
rainwater was effectively removed through the FAD process.
3.4. Comparison of pollutants removal by FAD and conventional
adsorption
The performance of the FAD process in terms of the removal of
several pollutants has been discussed in the previous sections, and
summarized in Table 1. To further illustrate the enhanced removal

of pollutants by the FAD process, a separate and conventional GAC

adsorption unit has been tested as a reference. The GAC unit had
the same operating conditions as the FAD plant and it was placed
in the same twin reactor, in which both the microltration and
UV irradiation steps were removed.
The removal efciency of the GAC unit for all the tested pollutants has been summarized and listed in Table 1. As pointed out
elsewhere, although good DOC removal was achieved by the separate GAC unit, the removal efciency of total organic matter and
bacteria by FAD was much higher than that by only GAC. The
FAD process exhibited excellent capacity for organic matter removal thanks to the synergetic effect of GAC adsorption, microltration and UV irradiation. In the hybrid process, adsorption by
GAC, three stages of biodegradation, separation by microltration
and oxidation by UV jointly contributed to the elimination of organic matter in the raw rainwater.
4. Conclusions
The FAD process has been tested for drinking water treatment
from harvested rainwater. The following conclusions might be
drawn from the results of our study:
 Pre-ltration exhibited a good total solid removal efciency in
respect to a moderate turbidity removal capacity and preserved
FAD unit performance over a long time.
 The FAD unit is able to produce pure water in terms of microbiological quality; it provides an absolute barrier for pathogens
and several major contaminants, also reducing turbidity. The
cost of FAD units is relatively low, which makes the system
accessible to small communities and households located in
developing, water stressed countries.
More extensive studies are needed to understand the FAD efciency on other types of bacteria and chemical pollutants and to
verify costs and applicability in developing countries.

V. Naddeo et al. / Journal of Hydrology 498 (2013) 287291

Acknowledgments
FAD treatment plant was designed and patented by Procom
S.R.L. (Firenze, Italy) and was provided for this research. Technical
assistance provided during the research activities by F. Santoriello,
P. Napodano, S. Giuliani, D. Ricco and M. Landi is highly
appreciated

References
Abbaszadegan, M., Hasan, M.N., Gerba, C.P., Roessler, P.F., Wilson, B.R., Kuennen, R.,
Van Dellen, E., 1997. The disinfection efcacy of a point-of-use water treatment
system against bacterial, viral and protozoan waterborne pathogens. Water Res
31 (3), 574582.
AWWAAPHAWEF, 1998. Standard Methods for the Examination of Water and
Wastewater, 20th ed., American Public Health Association/American Water
Works Association/Water Environment Federation. Washington, DC, USA.
Belgiorno, V., Napoli, R.M.A., 2000. Groundwater quality monitoring. Wat. Sci. Tech.
42 (12), 3741.
Belgiorno, V., Naddeo, V., Scannapieco, D., Zarra, T., Ricco, D., 2013. Ecological status
of rivers in preserved areas: Effects of meteorological parameters. Ecol. Eng. 53,
173182.
Domnech, L., Saur, D., 2011. A comparative appraisal of the use of rooftop
rainwater in single and multi-family buildings of the metropolitan area of
Barcelona (Spain): social experience, drinking water savings and economic
costs. J. Clean. Prod. 19, 598608.
Eroksuz, E., Rahman, A., 2010. Rainwater tanks in multi-unit buildings: a case study
for three Australian cities. Resour. Conserv. Recy. 54, 14491452.
Gadgil, A., 1998. Drinking water in developing countries. Ann. Rev. Energy Environ.
23, 253286.
Ghisi, E., Ferreira, D.F., 2007. Potential for potable water savings by using rainwater
and greywater in a multistory residential building in southern Brazil. Build.
Environ. 42, 25122522.
Ghisi, E., Montibeller, A., Schmidt, R.W., 2006. Potential for potable water savings by
using rainwater: An analysis over 62 cities in southern Brazil. Build. Environ. 41,
204210.
Ghisi, E., Bressan, D.L., Martini, M., 2007. Rainwater tank capacity and potential for
potable water savings by using rainwater in the residential sector of
southeastern Brazil. Build. Environ. 42, 16541666.
Gould, J., Nissen-Petersen, E., 1999. Rainwater Catchment Systems for Domestic
Supply: Design, Construction and Implementation. Intermediate Technology
Publications, London, UK.
Guo, W.S., Vigneswaran, S., Ngo, H.H., Xing, W., Goteti, P., 2008. Comparison of the
performance of submerged membrane bioreactor (SMBR) and submerged
membrane adsorption bioreactor (SMABR). Bioresour. Technol. 99 (5), 1012
1101.

291

Gur-Reznik, S., Katz, I., Dosoretz, C.G., 2008. Removal of dissolved organic matter by
granular-activated carbon adsorption as a pretreatment to reverse osmosis of
membrane bioreactor efuents. Water Res. 42 (67), 15951605.
Jones, M.P., Hunt, W.F., 2010. Performance of rainwater harvesting systems in the
southeastern United States. Resour. Conser. Recy. 54, 623629.
Khastagir, A., Jayasuriya, N., 2010. Optimal sizing of rainwater tanks for domestic
water conservation. J. Hydrol. 381, 181188.
Lantagne, S., Quick, R., Mintz, E.D., 2007. Household water treatment and safe.
Storage Options in Developing Countries. A Review of Current Implementation
Practices. Woodrows, Wilson International Center for Scholars, USA.
Li, Z., Boyle, F., Reynolds, A., 2010. Rainwater harvesting and greywater treatment
systems for domestic application in Ireland. Desalination 260, 18.
Liu, W., Wu, H., Wang, Z., Ong, S.L., Hu, J.Y., Ng, W.J., 2002. Investigation of
assimilable organic carbon (AOC) and bacterial regrowth in drinking water
distribution system. Water Res. 36 (4), 891898.
Mbilinyi, B.P., Tumbo, S.D., Mahoo, H.F., Senkondo, E.M., Hatibu, N., 2005.
Indigenous knowledge as decision support tool in rainwater harvesting. Phys.
Chem. Earth, Pts. A/B/C. Integrated Water Resources Management (IWRM) and
the Millennium Development Goals. Managing Water for Peace and Prosperity
30 (1116), 792798.
Mintz, E., Bartram, J., Lochery, P., Wegelin, M., 2001. Not just a drop in the bucket:
expanding access to point-of-use water treatment systems. Am. J. Public Health
91, 15651570.
Mun, J.S., Han, M.Y., 2012. Design and operational parameters of a rooftop rainwater
harvesting system: denition, sensitivity and verication. J. Environ. Manage.
93, 147153.
Murcott, S., 2005. Decentralized Household Water Treatment and Sanitation
Systems MIT Kenya Project. Report by the Department of Civil and
Environmental Engineering, Massachusetts Institute of Technology.
Naddeo, V., Belgiorno, V., 2007. Tertiary ltration in small wastewater treatment
plants. Water Sci. Technol. 55, 219225.
Naddeo, V., Belgiorno, V., Zarra, T., Scannapieco, D., 2013. Dynamic and embedded
evaluation procedure for strategic environmental assessment. Land Use Pol. 31,
605612.
Rahman, A., Keane, J., Imteaz, M.A., 2012. Rainwater harvesting in Greater Sydney:
water savings, reliability and economic benets. Resour. Conser. Recy. 61, 16
21.
Sobsey, M.D., 2002. Managing water in the home: accelerated health gains from
improved water supply. Water, Sanitation and Health. Department of Protection
of the Human Environment, World Health Organization, vol. 1, pp. 170.
Sobsey, M.D., Stauber, C.E., Casanova, L.M., Brown, J.M., Elliott, M.A., 2008. Point of
use household drinking water ltration: a practical, effective solution for
providing sustained access to safe drinking water in the developing world.
Environ. Sci. Technol. 42 (12), 42614426.
Villareal, E.L., Dixon, A., 2005. Analysis of a rainwater collection system for domestic
water supply in Ringdansen, Norrkping, Sweden. Build. Environ. 40, 1174
1184.
Zhu, K., Zhang, L., Hart, W., Liu, M., Chen, H., 2004. Quality issues in harvested
rainwater in arid and semi-arid Loess Plateau of northern China. J. Arid Environ.
57 (4), 487505.