G Model

RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS

Resources, Conservation and Recycling xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Resources, Conservation and Recycling

journal homepage: www.elsevier.com/locate/resconrec

Full length article

Developing surface water quality standards in China

Jing Su a,d , Danfeng Ji a , Mao Lin b , Yanqing Chen a , Yuanyuan Sun a , Shouliang Huo a ,
Jianchao Zhu a , Beidou Xi a,c,∗
a
State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing 100012, China
b
School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China
c
College of Environmental and Municipal Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China
d
CNHOMELAND Environmental Protection Water Pollution Governance Academician Workstation, Guangzhou 510000, China

a r t i c l e i n f o a b s t r a c t

Article history: A methodology for setting lake nutrient standard was proposed in this study based on review of water
Received 28 January 2016 quality standards (WQS) in China and understanding of lake nutrient criteria. There are very few studies
Received in revised form 29 July 2016 about lake nutrient criteria and water body classification system, which may cause difficulties in nutri-
Accepted 1 August 2016
ent level assessment and management. This paper reviewed WQS system in the US, EU, and Japan, and
Available online xxx
then presented the development procedures for revising surface WQS in different periods in China. The
amendment includes indicators, standard values, and framework of the standard system. However, cur-
Keywords:
rent WQS set in light of the criteria and standards of the US and Europe couldn’t adapt to the natural
Water quality standards
Lake nutrient criteria
conditions of lakes in China. Therefore, by learning from the water quality criteria setting method of USA
Structure equation model and taking the regional differences into consideration, a methodology of combining structure equation
China model and expert elicitation was recommended for setting WQS. Furthermore, it considered to inves-
tigate the differences (climate, elevation and geographical condition, etc.) in each region, and set lake
nutrient criteria and standards on regional scale. Consequently, the proposed methodology for setting
standards could be applied to each region in China. Yungui ecoregion was taken as an example for setting
up the lake nutrient standards with considering TP, Chl-a, TN and CODMn four influence factors for des-
ignated use attainment, and a suitable TP standard 0.02 mg/L was recommended by using the proposed
methods.
© 2016 Published by Elsevier B.V.

1. Introduction
Water quality is currently deteriorating rapidly due to water
pollution and human activities all over the world. The domestic and
industrial wastewater has contributed significantly to water pollu-
tion. Meanwhile, the water treatment plants are not upgraded in
a timely fashion for adequately treating the source water in com-
pliance with the water standards. WQS is an effective approach for
water quality protection and water basin system management.
Some developed countries proposed WQS to protect the water
body a long time ago. In June 1975, the Council of European Com-
munities (EC) adopted a Council Directive for the Member States
concerning the required quality of surface water intended for the
abstraction of drinking water (EC, 1999). The EC Directive divided
∗ Corresponding author at: State Key Laboratory of Environmental Criteria and
Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing
100012, China.
E-mail address: xibeidou@263.net (B. Xi).
raw water quality criteria into three classes according to the respec-
tive water treatment methods. Class I water treatment involves
simple purification processes including filtration and disinfection.
Class II water treatments include coagulation, flagellation, sedi-
mentation, filtration and disinfection. Class III, the most extensive
method, includes Class II processes and the activated carbon pro-
cess. For each class of source water criteria, there are two sets
of limit values. In order to meet with the high demand of good
water quality and improvement of wastewater treatment technol-
ogy, Water Framework Directive (WFD) was adopted in 2000 by
EU, and it has become a key piece of EU legislation that brings
together many aspects of water protection and management. The
purpose of the Directive is to establish a framework for the protec-
tion of inland surface waters (rivers and lakes), transitional waters
(estuaries), coastal waters and groundwater from a long-term view.
The WFD emphasizes river basin management and states that the
release of untreated water is prohibited throughout the EU to
ensure “good status” for water ecosystems. EU water directives
have brought considerable changes to national legislative statutes,
even in the countries with most developed environmental regula-

http://dx.doi.org/10.1016/j.resconrec.2016.08.003
0921-3449/© 2016 Published by Elsevier B.V.

Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
2 J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx
tions. The measures to achieve the goals will be coordinated at the
geographical/administrative level of the river basin district (WHO,
2003). River basin agencies may be based on existing authorities
but should not be based on hydrological administrative barriers
(Kallis and Butler, 2001). River basin districts should correspond to
large catchment basins incorporating the smaller sub-basins. The
latest plan for sustainable development of the Rhine was agreed in
accordance with the ecological and sustainable focus of the WFD
and is embodied in the “Rhine 2020” agreement of International
Commission for the Protection of the Rhine (ICPR).
In April 1970, the Water Quality Committee of the Japanese
Living Environment Council published WQS for water resources
area to protect the source water quality as well as public health
(Rook, 1974). Similar to the EC Directive, Japanese source water
standards were divided into three classes according to the type
of water purification process. In 1971, at the request of the US
Environmental Protection Agency (USEPA), the United States Pub-
lic Water Supplies Panel proposed water quality criteria for source
water based on water treatment processes, which were similar to
Class II of the EC Directive stated previously. In the USEPA, WQS
defined the water quality goals of a water body by designating
water its uses or treatments, by setting criteria necessary to protect
the uses, and by preventing degradation of water quality through
anti-degradation provisions (USEPA, 1980). States adopt WQS to
serve the purposes of the Clean Water Act, enhance the water qual-
ity, and protect public health or welfare (Cheremisinoff, 2002). EPA
first published a WQS regulation in 1975as a part of EPA’s water
quality management regulations. The first WQS Regulation did not
specifically address toxic pollutants or any other criteria. It simply
required “appropriate” water quality criteria necessary to support
designated uses. Due to the problem of Lake Eutrophication, Nutri-
ent Criteria Technical Guidance Manual was published in 2000, as
a reference for setting nutrient standard on the state scale (Gibson
et al., 2000). There were no general water standards in the USA, but
USEPA provided water quality criteria on the regional scale, and
standards were determined by state government with the consid-
eration of designated usages of water body and characteristics of
local economy and environment (USEPA, 2002).
Different countries have different approaches for water environ-
mental protection. Economic policy and market intervention play
an important role for solving environmental problems in the USA
and European countries, while basic approach for pollution control
is legislation in Japan where environmental protection is mainly
relying on law regulations and administrative guidance. However,
water environmental quality standard plays an important part in
water environmental management system in China. Water qual-
ity criteria and standards from the USA, and WQS from Japan and
European countries (WHO, 1993, 2003) were taken as references for
setting water body standards in China. The first water environmen-
tal quality standard (GB3838-83) in China was promulgated in 1983
by the Ministry of Environmental Protection of People’s Republic
of China (MEP). However, due to the conflicts between increasing
population and limited water resource, economic development and
environmental degradation, and the technique promotion and new
pollution emerging, GB3838-83 could not satisfy and effectively
support water environmental management. Therefore, GB3838-83
was converted to a new version, and its latest version was published
in 2002. The standard values in the standard of GB3838-2002 were
mostly set based on references from developed countries, which
could not reflect the water environmental condition in China, so
the GB3838-2002 needs to be further revised and refined in recent
years.
As there is no specific and profound research on the develop-
ment of China’s WQS, it is wise to review the development of WQS
system in China, so as to know China’s water quality management
concept and demand, and seek for a methodology for establish-
Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
ing WQS framework and determining standard levels satisfying the
demand. Therefore, the main purpose of this study is to, 1) review
the development of WQS in China, especially emphasizing on the
step changes and revision in order to understand future needs in
water environmental management; 2) develop a new technique
for setting WQS in China. The development procedures of WQS in
China are discussed in Chapter 2. Then Chapter 3 describes different
approaches for determining WQS in the USA for reference. Eventu-
ally a new methodology for setting WQS based on experiences in
the USA is proposed and discussed in Chapter 4 in this study.
2. Review of the development of WQS in China
According to the demand of surface water quality of national
environmental policy, the first national WQS “Surface Water Envi-
ronmental Quality Standards” was issued in 1983 under the name
of GB3838-83. Ever since then, it has been revised for 3 times. The
first revision was made in 1988 to form the version of GB3838-88.
It was then converted to GHZB 1-1999 in 1999 after 11 years of its
implementation. In 2002, the standards experienced the third revi-
sion and became the latest version (GB3838-2002) being effective
until now. After these 3 revisions, National Surface WQS developed
a comprehensive system in which nutrient indicator values in lakes
and reservoirs have been identified.GB 3838-2002 has become the
core for national water environmental monitoring, and plays a very
important role in water pollution prevention regulation and water
environmental management system. The development procedure
of surface WQS is listed in Table 1, and revised details are explained
below.
2.1. Water environmental quality standards (GB3838-83)
GB3838-83 was promulgated by the Ministry of Urban and Rural
Construction and Environmental Protection of the People’s Repub-
lic of China renamed as MEP in 2008 in light of “Environmental
Protection Law of the People’s Republic of China”. There are 3
classes corresponding to various water quality levels in GB3838-
83 (SEPA, 1983). Class I: water quality is in good conditions so
that the water could be used for various functions. Class II: water
quality is at middle level, and the water can be served as drinking
water resource and for fisheries. Class III is the lowest demand for
surface water pollution regulation. And 20 comprehensive indica-
tors were recommended in this standard to indicate different water
conditions.
This standard effectively acted as preventing water pollution,
ensuring human health, protecting resource, maintaining ecologi-
cal balance, and conserving sustainable economic development. It
could be applied to rivers, lakes, reservoirs, and other water basins,
and became a major regulation for environmental planning, envi-
ronmental management, water quality assessment, and pollutant
discharge.
2.2. From GB3838-83 to GB3838-88
GB3838-83 played a significant role in establishing water qual-
ity planning and setting objective for water pollution prevention
in 2000. However, with the economic development and scientific
improvement, the enforcement of GB3838-83 could not effectively
support environmental management any more.
Two driving factors contributed to the revision from GB3838-
83 to GB 3838-88. Firstly, the WQS of GB3838-83 was subdivided
into 3 classes but the division system was not closely related to
different functions of water body. Secondly, there was no clear
expression for the implementation scope and detection method.
Therefore, it was imperative to make necessary revision to GB3838-
83 so as to prevent water environmental deterioration caused by
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx 3

Table 1
Development Procedure of Surface WQS.

Standard GB3838-83 GB3838-88 GHZB 1-1999 GB3838-2002

Classification 3 Classes 5 Classes 5 Classes 5 Classes

Indicators Basic parameters 20 30 31 24

Organic chemical standards 0 0 40 80
Lake eutrophication 0 0 4 5
Assessment method No Single-factor evaluation Single-factor evaluation Multi-factors comprehensive evaluation

urbanization and industrialization. The water body in new version two concerns during the revision process. 1) Some toxic chemi-
(GB3838-88) was classified to natural reserve (Class I), drinking cal parameters were missing in GB3838-88. GB3838-88 had only
water source (Class II), industrial use (Class III), recreation (Class IV), COD and BOD for organic pollutants assessment (Charnley, 2008;
and irrigation (Class V) by water function, which could reflect envi- Greer et al., 2002); 2) There were no nutrient related indicators in
ronmental management demand and water functional protection GB3838-88. Lake Eutrophication had been erupting at a higher fre-
(SEPA, 1988). The revision details are as follow: quency over the past 30 years. According to the surveyed, only 5%
of the 34 lakes were eutrophicated in the 1970s, and the proportion
of eutrophicated lakes increased to 35.76% during 1986–1989. Thus
(1) Changes of indicators. The number of indicators changed from
eutrophication indicators needed to be added into the standard
20 in GB 3838-83 to 30 in GB3838-88. Three indicators were
system to control the eutrophication.
combined, one was replaced, and other 14 indicators were
According to the above-mentioned two main issues, GHZB
added into the new version. Narrative text was induced into
1-1999 was promulgated. There were totally 75 indicators in
GB3838-88 instead of numerical statement for some indica-
GHZB1-1999, of which, 31 were basic parameters, 4 were associ-
tors which could hardly be determined by specific values. Total
ated with lake eutrophication, and 40 organic chemical substances
nitrogen (TN) was replaced by Kjeldahl nitrogen (KTN), and
related indicators were supplemented (SEPA, 1999). It also pro-
NO3 -N, NO2 -N and ammonia were induced into GB3838-88,
vided a scientific approach for water quality assessment, which was
so as to effectively respond to nitrogen circulation. In order to
based on single criterion analysis where each water quality indica-
protect drinking water source, eight indicators (sulfate, anionic
tor was evaluated independently for water quality assessment. The
surfactants, benzo a-pyrene, chlorine, dissolved iron, Mn, Zn,
promulgation of GHZB 1-1999 served as a mark for promoting sur-
and Se4+ ) were added based on the requirement of protecting
face WQS. The following is the detailed revision from GB3838-88
aquatic life and human health. For the purpose of making con-
to GHZB 1-1999.
sistent meaning of COD in WQS and discharge standards, CODCr
and its detective method were supplemented.
(2) Changes of standard values. With water classes extending from
(1) Change of framework. GHZB 1-1999 still used the basic frame-
3 to 5 in GB3838-88, most water bodies were reclassified from
work in GB3838-88, which was divided into 5 classes based on
Class III to Class IV. After revision, a number of criteria and
water function. When water bodies have a variety of functions,
some of the standards GB3838-88 adopted for toxic substances
the most sensitive one would be taken as the basis function for
were stricter than those in developed countries, but the WQS
determining the conservation objectives. Regarding that some
for some degradable toxic substances (e.g. volatile phenol)
areas with abundant and high quality water resources were
might be less stringent compared to criteria. Such considera-
still under development, it was required to analyze and discuss
tion allowed for and promoted artificial treatment and natural
that whether the standards in these areas should be lowered
purification.
due to economic demand. The new revision clearly defined the
(3) Application of scientific results. Water quality criteria in devel-
appropriate application boundaries between WQS and other
oped countries were widely invested in this revision. The
standards, such as the effluent standard and standards of dif-
correlation coefficient between toxicity and hardness of heavy
ferent functions. This was to ensure that each standard would
metal was quite high, and revised equation was applied in the
be able to maintain its specific function.
new version of standard. It was difficult to reflect the effect
(2) Highlighting total amount control for pollutant discharge.
on heavy metal standard value from hardness (If the hardness
Water environmental capacity could be calculated based on the
increases 10 times, the heavy metal standard value could be
input response relationship between pollution and protection
extended to over 10 times), so revised hardness equation was
objective. Therefore, water pollution control objective could
introduced as a reference for local water quality assessment.
be calculated through analyzing water environmental capac-
(4) Minimizing the conflict among different standards. Other than
ity and pollutant amount fed into the water body. The water
GB3838-88, there were some standards for specific function,
environmental protection objective, which was used for deter-
such as “Drinking Water Sanitary Standard” (GB5749-85),
mining WQS values, became the key performance indicators.
“WQS for Fisheries” (TJ35-79), and “Standards for Irrigation
COD, NH3 -N, and sulfide were selected as water quality indi-
Water Quality” (GB5084-85). There might be some conflicts
cators in discharge outlet area; DO, BOD5 , un-ionized ammonia
during the enforcement process. GB3838-88 took this conflict
were chosen for main water body. The standard contained both
into consideration and ensured the enforcement of other stan-
un-ionized and NH3 -N based on the latest research result on
dards while implementing GB3838-88.
water quality criteria for aquatic.
(3) Regional differences were reflected. In GHZB 1-1999, there
2.3. From GB3838-88 to GHZB 1-1999 were 31 basic indicators being compelled to be monitored and
applied for water quality assessment; and some special indica-
GB3838-88 provided essential support for environmental man- tors were alternative. Local Environmental Protection Bureau
agement and related policy regulation during 1980s −1990s. As could select different special indicators in WQS regarding to
the accelerated economic development after the reform and open- regional characteristics. This revision effectively reflected dif-
ing policy intensified public’s demand on environmental quality, ferent features in different regions, thus powerfully promoted
MEP further revised GB3838-88 to GHZB 1-1999. There were the function and effect of local water resource management.

Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
4 J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx
The change was based on the consideration of the following fac-
tors. 1) There were thousands of lakes and reservoirs located in
different regions with different environmental conditions. For
some eutrophicated lakes, the factors of causing eutrophica-
tion were different in each area. Some of them were caused
by phosphorus, and some were dominated by nitrogen; 2) Each
region had different organic chemical substances due to various
historical, regional, and social background. In order to con-
trol organic chemical substances in drinking water sources, 40
indicators were recommended in GHZB 1-1999 and used for
regional water environmental management.
(4) Revision on toxic chemical substances and Lake Eutrophication
control. 1)14 sorts and 68 specifics priority pollutants were put
into priority list. Pollutants such as persistent organic pollu-
tants (POPs) which had been listed in treatment priority were
increasingly found in sewage water. 2) Four nutrient related
indicators (TN, total phosphorus (TP), chlorophyll-a (Chl-a) and
Secchi Depth (SD)) were induced into GHZB 1-1999. Different
from rivers, nutrients in lakes could easily be accumulated and
became the primary cause for lake eutrophication (Huo et al.,
2014). Consequently, it was imperative to select and screen
some indicators which could reflect the response between
nutrient concentration and ecological effect.
(5) Relaxed standards of organic substances with low oxygen con-
sumption. The standard values of CODCr and CODMn were
relaxed in GHZB 1-1999 for the level of Class III, IV, and V.
Representative aquatic, such as elver, eel, and Eriocheir sinen-
sis were applied as detectors to renew the indicator values.
Three factors contributed to this result: 1) There was no water
quality criterion on CODCr and CODMn which could be taken
as reference for setting WQS in the worldwide. In light of bio-
logical toxicology experiment in recent years, 42 mg/L (CODCr )
was suggested as the level for sensitive aquatic. 2) WQS estab-
lished by the European Economic Community was the only
standard using COD as an indicator in WQS of drinking water
source. The value of COD was suggested as 30 mg/L for pro-
tecting human health, while it was not in the indicator list in
drinking water sanitary standard in China. 3) Some toxic indica-
tors were added, and the values of oxygen consumption related
indicators (DO, BOD5, NH3 -N) were specified. COD was a factor
used for reflecting the amount of low oxygen consumption and
organic substance without biodegradability. The application
scope of COD became smaller in GHZB 1-1999. Standards val-
ues of CODCr for Class III–V were relaxed from 15 mg/L, 20 mg/L,
25 mg/L in GB3838-88 to 20 mg/L, 30 mg/L and 40 mg/L, respec-
tively in GHZB 1-1999. Meanwhile, CODMn in Class III, IV, and
V was changed from 6 mg/L, 8 mg/L, and 10 mg/L to 8 mg/L,
10 mg/L, and 15 mg/L, respectively.
(6) Water assessment methods. Single-criterion assessment was
applied in surface WQS system. Since water body might have
different characteristics in various stages (wet season, dry sea-
son and level period), Seasonal assessment ought to be applied
for water bodies whose water quality changed with seasons.
Thus seasonal data (at least 2 sets of monitoring data ought to be
given in each season) needed to be collected instead of instanta-
neous values and anural average values. Ten indexes, including
DO, COD, AR-OH, NH3 -N, cyanide, fluoride, total Hg, As, Pb, Cr6+
were supposed to 100% meet the demand of the standards in
each water stage; while 80% for that of other indexes.
(7) Clarify the relationship between “Surface Water Environmental
Standards” GHZB 1-1999 and other standards. “Surface Water
Environmental Standards” was a general standard which was
supposed to be respected and followed by all surface water
resources in China. 1) GHZB 1-1999 and Sea WQS: these two
standards were both regarded as water environmental quality
standards but with different scopes. “Surface Water Environ-
Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
mental Standards” was applied to water body with functions,
such as river, lake, canal, channel and reservoirs, While, “Sea
Water Quality Standards” was applied to maritime water, sur-
face water and entrance which were close to offshore waters.
2) Standards for irrigation water quality: Different from GHZB
1-1999, Irrigation Water Quality Standard was used for judging
whether the water quality could meet the standard for irriga-
tion.
2.4. From GHZB 1-1999 to GB3838-2002
During the implementation of GHZB 1-1999, problems emerged
with high demand for water quality and water environmental pro-
tection. The monitored data showed that some of the standard
values were not reasonable, since they could not be achieved with
current treatment technologies and management measures. The
main issues in GHZB 1-1999 are as follows:
1) Nutrient standard values were different from those in inter-
national nutrient criteria. USEPA Nutrient Criteria Technical
Guidance Manual for Lakes and Reservoirs was published in
April 2000 after the revision of GHZB 1-1999 (USEPA, 2001,
2008). The EPA water quality criteria published at this time
contained over 10 indicators on lake nutrient criteria. Without
the reference of these nutrient criteria, eutrophication related
indicators were mainly determined by considering related stan-
dards in other countries. Some indicators, for instance of TN and
TP, were stricter than those in Japan and South Korea. Hence it
provided an appropriate opportunity to update eutrophication
related indicators.
2) Change in drinking water resource standard values. Water at
the quality level of Class II and III could both be used for drink-
ing water source and aquatic life protection in GHZB 1-1999. In
Class II and III with the new version of “Drinking Water Health
Standards”, issued in September 2001, some indicators in GHZB
1-1999 was required to be revised, so as to facilitate water
quality assessment and environmental management of drink-
ing water sources. The modified standard effectively supported
the enforcement of WQS in drinking water sources.
3) Problem in nitrogen related parameters. There were 6 nitro-
gen related indicators involved in GHZB1-1999, including 5
basic indicators (NO3 -N, NO2 -N, NH3 -N, non-ionic ammonia,
and TKN), and one special indicator (TN) for lakes and reservoirs.
It was understandable to have different indicators for prevention
from various nitrogen related pollutants, while it might lead to
redundant problems caused by overlaying different indicators.
4) USEPA promulgated a national water quality criteria guidance,
which contained 157 indicators related to human health and
aquatic life protection in April 1999. In this new version, non-
ionic ammonia was replaced by NH3 -N regarding to its toxic
under various conditions (different pH levels and temperatures).
Moreover, it revised some indicator values, and clarified the def-
inition of heavy metal, thus it was necessary to integrate the new
points into GHZB 1-1999, and promote surface WQS. Based on
the above concerns, revisions were made to form GB3838-2002.
Following are some explanations of GB3838-2002.
(1) WQS framework. There are still 5 classes in GB3838-2002 based
on different designated uses and water protection objectives.
Parameters values were determined in accordance with the
water environmental protection objective of aquatic life and
human health for both water body and the recharge sources.
(2) Emphasizing the coordination with other standards. WQS for
5 classes were supposed to be integrated. Each sort of water
body ought to comply with basic indicators, while drinking
water source standards including supplemented indicators and
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx 5

specific indicators, should be coordinated with drinking water
health standards. The specific indicators demonstrated that
drinking water source would be taken as a significant part in
water environment protection in the future, and local govern-
ment could make their decision based on assessment indicators
to improve water pollution control, and to protect water envi-
ronmental resource and human health, so that to promote
drinking water source management.
(3) Main revision. GB3838-2002 supplemented the “Water body
with a specific function” to express the relationship between
water body function and classification. “Water body function
and standards classification” revised to “Water body function
classification”. “Rare aquatic habitat” was replaced by “Precious
fish reserves” in class II, and “Larvae feeding ground” was added
into class II as well. In class III, “General fish protected areas”
was replaced by “Wintering grounds of fish and shrimp, migra-
tion channels, aquaculture areas of fishing waters”.
There are 109 parameters in GB3838-2002, including 24 basic
parameters, 5 supplemented parameters for drinking water source,
and 80 specific parameters. NO2 -N, non-ionic ammonia, and KTN
were deleted from the old version on drinking water source (SEPA,
2002). Not only TN was added, but also the standard values of
TN and TP for lake and reservoirs were specified (Zan et al.,
2012). Seven indicators, including pH, DO, NH3 -N, CODMn , Pb, and
Escherichia coli in dung were revised in the new version. The num-
ber of organic chemical substances related indicators was increased
to 80, of which except for the standard value for carbon tetra-
chloride being stricter, other standards have been relaxed, and the
standard values are shown in Table 2.
Basic indicators in surface WQS were implemented for rivers,
lakes, canals, channels, and reservoirs for different functions. Sup-
plemented and specific indicators for drinking water resource were
applied (Dugan et al., 2009). Specific indicators could be determined
and selected by local governments considering local environmental
characteristics and protection demand.
2.5. Inadequacies in GB3838-2002
Although the GB3838-2002 made corresponding improvement
in view of the increasingly serous eutrophication, there are still
some insufficiencies existing in this latest version, as eutrophica-
tion control management is incomplete compared in developed
countries. Firstly, a gap remains in WQS system between China
and developed countries. Lake Eutrophication management mainly
relies on 2 nutrient related indicators (TN and TP) in GB3838-2002.
Since there were no lake nutrient criteria related studies during
the revision period from GHZB 1-1999 to GB3838-2002, nutri-
ent indicator standards were mainly based on the references from
developed countries (Zhou et al., 2014). Some of the standard val-
ues are more stringent than those in Japan and South Korea, while
some are relaxed. Secondly, GB3838-2002 is applied generally in
all the lakes in China without considering the differences in differ-
ent regions, including various climates, elevation, geography, and
other regional factors (Ding et al., 2015). A methodology is needed
to transform the lake nutrient criteria to standards considering
regional differences.
The crucial step of this transformation is to find a measurable
indicator to express the probability of attainment of designated
use. It is significant to achieve tradeoff between water environ-
mental protection and economy development. These two issues
are in accordance with the USEPA. Meanwhile, USEPA using the
eco-regionalization technique divides lakes into different regions in
view of various ecological factors, and sets lake nutrient criteria and
standards in regional scale. Therefore, the methodology of USEPA
could be taken as a reference for setting lake nutrient standards
based on nutrient criteria in China. Nevertheless, the methodol-
ogy for setting nutrient standards in individual state is different
from USEPA. The approach used by USEPA with four indicators
could not reflect the relationship between nutrient standard and
designated use, while individual state considers the water body
type and its designated uses by combining expert elicitation and
structure equation model. So it’s advisable to know about differ-
ent approaches for setting lake nutrient standard in the USEPA and
different states in the USA for comprehensive reference.
3. Methodology for setting water quality criteria and
standard in the USA
3.1. USEPA water quality criteria and standards
Nutrient over-enrichment is a major source of water pollu-
tion in the United States. The link between eutrophication and
the over-enrichment of surface waters has long been presumed
but not definitively documented (Cunha et al., 2012). Reference
conditions are statuses observed in the absence of human distur-
bance or alteration, corresponding to very low pressure, with only
minimal human impacts from industrialization, urbanization and
intensive agriculture. However, because some parts of the coun-
try have naturally higher soil matrixes and different precipitation
regimes, the application of that general consensus approach has
to be adjusted by region. Therefore, an eco-region and reference
condition approach is necessary to develop nutrient criteria appro-
priate to different geographical and climatological areas of the
country. Initially, the continental United States has been divided
into 14 separate eco-regions with similar geographical charac-
teristics, and criteria and standards are developed for each. The
reference conditions help set the upper bounds of what can be con-
sidered as the most natural and attainable lake conditions for an
ecoregion, and thus contribute to setting the criteria and standards
at appropriate levels (Huo et al., 2012).
USEPA recommended four variables, TN, TP, Chl-a, and SD, to
develop nutrient criteria that reflects both causal and response
conditions of eutrophication. The criterion levels were set by
estimating the median value of the 25th percentile of seasonal fre-
quency distributions for all lakes and reservoirs within an ecoregion
(Walker et al., 2007). There was no relationship between criteria
value and designated use in this approach. WQS is set with con-
sideration of designated uses in individual stage based on regional
water quality criteria in the USA. However, the methodology for
setting WQS is different in various states. The criteria will serve
as the base-line lake nutrients at national level, all the States will
need to set their lake nutrient standards with the consideration of
nutrient criteria.
3.2. Nutrient standards setting method in Minnesota, US
The method used for setting WQS in Minnesota is mainly
based on the USEPA ecoregion reference water body (Heiskary and
Wilson, 2005). Minnesota not only classified their lakes and reser-
voirs by reference conditions and eco-region, but also subdivided
the water bodies via natural lakes, reservoirs, designated uses, and
depth (Walker et al., 2007). The most sensitive designated use was
selected to be protected because it was believed that if the most
sensitive use was guaranteed, the other uses then could also be
appropriately protected. Instead of using a single criterion, such as
the current TP criterion set in 1988, Minnesota chose to develop
standards using three criterion variables: TP, Chl-a, and SD. Nitro-
gen was not included since the water bodies were phosphorus
limited based on many studies on eutrophication in Minnesota.
Meanwhile, Chl-a and SD were selected as criteria indicators as they

Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
6 J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx

Table 2
Comparison of Standard Values between GHZB 1-1999 and GB3838-2002.

Indicators Standard Classes

I II III IV V

pH 02 6–9
99 6.5–8.5 6–9
DO (mg/L) 02 90% or 7.5 6 5 3 2
99 90% 6 5 3 2
NH3 -N (mg/L) 02 0.15 0.5 1.0 1.5 2.0
99 0.5 0.5 0.5 01 1.5
TP (mg/L) River 02 0.02 0.1 0.2 0.3 0.4
Lake and reservoirs 99 0.02 0.1 0.1 0.2 0.2
02 0.01 0.025 0.05 0.1 0.2
99 0.002 0.01 0.025 0.06 0.12
TN (mg/L) 02 0.2 0.5 01 1.5 2.0
99 0.04 0.15 0.3 0.7 1.2
Pb (mg/L) 02 0.01 0.01 0.05 0.05 0.1
99 0.01 0.05 0.05 0.05 0.1
Coliformin (per L) 02 200 2000 10000 20000 40000
99 200 1000 2000 5000 10000
CODMn (mg/L) 02 2 4 6 10 15
99 2 4 8 10 15

were both closely related to TP. The weight-of-evidence approach
was applied to set criterion levels. Minnesota’s method used the
wealth of scientific knowledge regarding eutrophication in Min-
nesota lakes to refine their current nutrient standards. Although
the goal of nutrient standards was to protect the designated use,
the standards variable was not been explicitly linked to the specific
designated uses. Additionally, the standards levels were weak with
the trade-offs underlying this policy decision (Kenney, 2007).
3.3. Methodology for water quality standards in Arizona, US
Arizona translated the state’s primary narrative nutrient stan-
dards into numeric nutrient criteria (USEPA, 2008). They used a
weight-of-evidence approach to compare monitoring data with tar-
gets provided by sources such as scientific literatures, the trophic
state indexes developed for Arizona waters, and statistical analysis
results. Then criterion Chl-a was recommended, with secondary
consideration to SD, TP, TN, and total Kjeldahl nitrogen (Walker
et al., 2007; Kenney, 2007). A water body would be classified as
impaired if there were two exceedances of the nutrient standards
within a 2–5 year assessment period.
3.4. Methodology for water quality standards in North Carolina,
US
The method adopted by North Carolina is different from other
states. They considered the method used by USEPA and other states
was difficult to assess whether a specific nutrient criterion was
appropriate, overprotective, or under-protective. North Carolina
used expert elicitation to quantify designated-use attainment for
lakes, and they applied structural equation model (SEM) to iden-
tify good WQS as the best predictors of the designated use (Li
et al., 2015). Setting the standard level required explicit considera-
tion of the decision-makers’ value trade-off between water quality
and mitigation cost (Huo et al., 2013b). They used a combination
of water quality modeling and multi-attribute value analysis to
determine the standards variables and thresholds that maximized
decision makers’ overall value.
This predictive nutrient standards method successfully trans-
ferred the narrative designates use into measureable standards. It
enabled us to select the most significant indicator of designated use
attainment probability, and established the relationship between
the criterion level and the designated use attainment. This process
explicitly considered the attainment of water quality goal and pre-
sented a procedure to quantify the appropriateness of candidate
nutrient criteria. This method provided decision-makers with an
estimate of risk associated with the criterion level, and facilitated
the selection of appropriate WQS. The method successfully estab-
lished the relationship among nutrient indicator, designated use
and attainable percentage. Designated use, which could be diffi-
cult to be measured, was expressed by indicators and numeric, so
it was possible to evaluate whether the water designated use was
attained or not. Meanwhile, the approach balanced environmental
benefit and pollution control cost so as to provide references for
decision-makers.
4. Methodology for setting lake nutrient standards in China
Based on the review of the methods used in the US for setting
lake nutrient standards, there are generally three approaches for
setting numeric standard to address nitrogen/phosphorus pollu-
tion (USEPA, 2000): reference condition approaches, mechanistic
modeling, and stressor-response analysis (USEPA, 2010). The pro-
posed predictive nutrient criteria method is a component of the
stress-response analysis. The CWA requires that surface waters
should meet standards necessary for human recreational demands,
as well as aquatic breeding. Thus selecting the most sensitive des-
ignated use to set the standard could satisfy the needs. Different
from the USA, the water use is not aiming at human health and
water ecological system safety protecting, but at water resources
use in China. The water quality standard is set according to different
use function (Li et al., 2015). Thus, there is a demand of establishing
a method to link standard variables and levels to different desig-
nated uses. The predictive nutrient criteria method used in North
Carolina could successfully transfer the narrative designates use
into measureable standards. It is a referable standard establishing
method to China (Ji et al., 2013).
4.1. Structural equation modeling (SEM)
SEM has been widely used in various research fields (Hayduk,
1996; Po et al., 2005). And it has been applied to water environment
(La Peyre et al., 2001; Russell and Hampton, 2006; Sutton-Grier
et al., 2010). SEM is often considered as superior to multivariate
regression, as it measures indirect effects between two explanatory
variables explicitly, and where the effects between two variables
can be mediated by another intermediary variable (Kline, 1998).

Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx
Fig. 1. Distribution of the TP Concentration in China.
4.2. Expert elicitation introduced in SEM
Although science was thought of as objective, judgment is
indeed necessary throughout a scientific study, from the state-
ment of hypotheses, the specification of a model, the design of
experiments or monitoring programs, the selection of methods of
analysis, to the final inferences and conclusions. All of these tasks
generally involve expert judgment intermingled with the objective
analysis of data (Reckhow et al., 2005). The need to link measured
standard with narrative designated-use statements unavoidably
requires expert judgment. The goal of expert elicitation is to extract
subjective judgments from experts in a systematic procedure. The
improvement resulting from the use of expert elicitation is that
it makes these subjective judgments transparent (Reckhow et al.,
2005).
Designated use attainment cannot be directly measured. There-
fore, an expert group was convened to elicit the probability of
designated use attainment given the correlated water quality data.
Similar to Reckhow’s approach, a SEM was formulated to describe
the interplay between designated use attainment and physical,
chemical, as well as biological factors. It was useful for determin-
ing the most predictive measured water quality variables of the
probability of attainment of the designated use (designated use
attainment) for a region of lakes. Using SEM, these predictive vari-
ables were statistically determined by combining water quality
Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
7
data with assessments from experts on the probability of desig-
nated use attainment.
Lake eutriphication is becoming a big problem in China and
other countries as well. TP and TN are the key indicators causing
lake eutrophication (Elklit, 2012; Hauck et al., 2016). Fig. 1 shows
the distribution of TP in China. It indicates that the concentration
increases with the color getting dark. The lakes with the highest
concentration are almost 10 times of the lakes with the lowest
ones (Huo et al., 2013a). Therefore, it is necessary to define WQS on
regional scale so as to reflect the driving differences.
4.3. Procedures for setting methodology for lake nutrient
standards in China
Methodology applied by North Carolina, US could be referred as
important lessons for China. Combining the elicited judgments with
the water quality data, a data set could be created which allows us to
use SEM to identify the best predictive criterion for designated use
(Reckhow et al., 2005). In addition, we can reformulate the resul-
tant SEM to estimate the probability of attainment associated with
various levels of the criterion. In China, the aim of standard classifi-
cation system was to protect the designated use of different water
bodies. Expert elicitation was superior to experimental and statisti-
cal approaches to quantify the narrative designated use statement
(Ji et al., 2013).Due to that elicited judgments are based on sub-
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
8 J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx

Statistical
Base line
Lake Nutrient

Criteria
Characteristic Investigation

Structure Equation Model

Recommended Lake
Designated Use
Nutrient Standards
Expert elicitation

Technique and economic

evaluation

Lake Nutrient Standards

Fig. 2. Flow Chart for Setting Lake Nutrient Standard in China.

jective results from experts, statistical analysis was also induced

to seek distribution of water monitoring data. And based on an
“a priori” statistical technique (Hurlimann et al., 2008), SEM could
determine and test a predictive criterion for designated use. There-
fore, this was a good methodology being appropriate for Chinese
actual situation. Detailed flow chart for setting lake nutrient stan-
dard is shown in Fig. 2.
Step 1: Water body classification. The division is based on var-
ious water body functions. There are 5 classes regarding to the
current surface water environmental standards system: Class I for
natural reserve; class II for drinking water source, precious fish
reserves, spawning grounds; class III for fishery and direct contact
recreation; class IV for recreation, shipping and flood protection;
class V for irrigation and recreational environment use.
Step 2: Indicator screening. There are two causal variables for
lake eutrophication: TP and TN, which are suggested as basic indi-
cators for nutrient standards (Xu et al., 2014; Wu et al., 2014).
However, TN and TP could not appropriately and timely reflect
Fig. 3. Map of the Study Area.
eutrophic state (Jiang et al., 2014; William et al., 2011). Therefore,
other indicators need to be screened through integration of elicited
judgments and SEM. An investigation form needs to be provided Step 4: Modeling and standard determination. Firstly, provid-
to expert in limnology, and familiar with the surveyed lakes. Over ing lake water quality monitored data (monitored data in different
200 monitored data sets for 20 indicators would be involved in the monitoring sites) to experts in limnology, and familiar with the
form as a database, and the experts need to score the probability of surveyed lakes. They would then be asked to estimate the proba-
attaining the designated use for each data set. In view of SEM, the bility of the designated use attainment for each data set. Secondly,
indicator with the highest coefficient with attainment probabilities structural equation model would be established and applied to
of the designated use would be recommended as nutrient standard determine the most predictive indicator of the use attainment as
indicators. the criteria. Finally, revising (or reformulating) the structural equa-
Step 3: Monitored data distribution analysis. Based on the tion model to the best-fit model, and assessing the probability of
review and analysis of historical water quality data, statistic use attainment associated with the level of the criterion.
method would be applied to analyze data distribution. Understand-
ing the water quality distribution is helpful for setting reasonable 4.4. Case study
WQS. On account of the subjective result from integration of elicited
judgments and SEM, historical data statistic analysis could be used This method was used for setting lake nutrient standards in
for remedy so as to achieve more reasonable result. It was reported Yungui Ecoregion in China, including Yunnan province, Guizhou
that the 50th percentile of the distribution of the data might be province, and some areas of Guangxi and Sichuan province. It is in
recommended as the main value of the water body, thus it could the subtropical zone with the area of 7.31 × 105 km2 (Fig. 3). Cov-
be taken as a reference for determining WQS, and standard values ered by dense jungle, rivers and canyons, most natural lakes in this
ought to be lower than the 50th percentile. plateau are located within the altitude between 1280 m and 3270 m
(above sea level, ASL).

Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx
Table 3
Designated Use Attainment Versus the Candidate Criterion for TP.
TP (mg/L) 0.007
Designated use attainment (%) 100
Fig. 4. SEMs from Multi-expert.
Lakes serving for drinking water sources in the Yungui Ecoregion
were selected as the study case. Totally, 160 independent moni-
tored data from year 1990 to 2010 were collected from the 12 lakes
from the Department of Environmental Protection of the Yunnan
and Sichuan Provinces, China. The data consisted of WT, TP, TN,
Chl-a, SD, pH, DO, BOD, and COD.
The previewers studies suggested that result trom the
multi-experts is more practical than that from single expert.
Therefore,four limnologic experts were selected in this study to for-
mulate the connections between water quality and the designated
use of the water. The multi-expert models fairly reflected the com-
mon knowledge of different experts. This model explains 25% of
the variation in Chl-a and 22% of the variation in the probability of
designated-use attainment (Fig. 4).
Determination of the standard level associated with the attain-
ment and nonattainment transition ideally requires the integration
of science and values (Reckhow et al., 2005). A standard level with
100% assurance of designated use attainment is generally impracti-
cal. Therefore, the standard level decision should contain scientific
uncertainty and attitude toward risk of nonattainment. Hence, the
decision should be made considering what probability (or, perhaps,
what percentage of space-time) of nonattainment is acceptable
(Reckhow et al., 2005). USEPA guidance (USEPA, 1997) suggested
that if more than 10% of samples substandard, the water body will
be identified as not fully supporting the designated use. Different
from the developed countries, a rather low assurance of designated
use attainment (85%) is allowed for short-term objective in devel-
oping countries like China. The standard level associated with the
higher expected utility (90%) might then be considered as the long-
term objective for LE control. Using a graph like that in Fig. 3, the
different Chl-a and TP levels could be test based on willingness to
accept risk of nonattainment (Ji et al., 2013).
Designated use attainment was affected directly by TP, and
Chl-a, and indirectly by temperature and CODMn (through Chl-a).
Increase levels of temperature, TP, CODMn and Chl-a may lead to
decrease of the probability of designated use attainment. The model
can also be used for estimating the probability of use attainment
associated with various levels of candidate standards. The proba-
bility of designated use attainment would reach 100% when the TP
concentration was lower than 0.01 mg/L; it remained high (more
than 90%) when the TP level was less than 0.018 mg/L; it then
Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
9
0.020 0.025 0.048 0.079 0.100
85 80 60 40 30
decreased along with the increase of TP, and decline to 85% when
the TP concentration was 0.02 mg/L (Table 3). It is recommended
that the probability of designated use over 85% can be used as the
corresponding lake nutrient standards, and it also need to consider
the lake nutrient criteria. It means that the standards value needs to
be looser than the criteria values. The Criteria value of TP in Yungui
Province is 0.01 mg/L, thus it could be feasible if the TP concen-
tration was 0.02 with the percentage of attainment of designated
use.
5. Conclusions
With the source water quality being deteriorating and growing
eruption of eutrophication, setting water quality standard becomes
an effective approach for water quality protection and water basin
system management. There are few studies about lake nutrient
criteria or water boy classification in China. A methodology for set-
ting lake nutrient standard has been proposed in this study based
on reviewing of water quality standard in China and taking the
experience of developed countries.
Water quality standard system in the US, European Union, and
Japan were discussed. Then it reviewed the modification processes
of surface water quality standard in different period in China,
involved in indicators, standard values, and framework of the stan-
dard system. There are two main problems exit in the current
water quality standards: (1) a gap between developed countries
and China in water quality standard system; (2) without consid-
ering the regions differences. Taking the nutrient criteria setting
method as a reference, this study combined the structure equa-
tion model and expert elicitation to get a methodology for setting
water quality standard. By investigating the differences in each
region, and setting lake nutrient criteria and standards on regional
scale, this methodology for setting standards could be applied for
each region in China. A suitable TP standard 0.02 mg/L in Yungui
ecoregion was recommended by using the proposed methods.
Acknowledgments
This work was supported by the Mega-projects of Science
Research for Water Environment Improvement (No. 2012ZX07101-
002) and National Natural Science Foundation of China (No.
71103166). We are grateful to Dr. Ming Xu, experts and two anony-
mous reviewers for their insightful comments and suggestions
which improved the quality of the original manuscript.
References
Charnley, G., 2008. Perchlorate: overview of risks and regulations. Food Chem.
Toxicol. 46, 2307–2315.
Cheremisinoff, N.P., 2002. Handbook of Water and Wastewater Treatment
Technologies. Butterworth-Heinemann, Woburn.
Cunha, D.G.F., Ogura, A.P., Calijuri, M.C., 2012. Nutrient reference concentrations
and trophic state boundaries in subtropical reservoirs. Water Sci. Technol. 65,
1461–1467.
Ding, J.T., Cao, J.L., Xi, B.D., Xu, Q.G., Su, J., et al., 2015. Spatial heterogeneity of lake
eutrophication caused by physiogeographic conditions: an analysis of 143
lakes in China. J. Environ. Sci. 30, 140–147.
Dugan, N.R., Williams, D.J., Meyer, M., et al., 2009. The impact of temperature on
performance of anaerobic biological treatment of perchlorate in drinking
water. Water Res. 43, 1867–1878.
EC, 1999. EU Focus on Clean Water. Office for Official Publications of the European
Communities, Luxembourg.
Elklit, A., 2012. Interaction of adult attachment and division-of-labour. Clean 40,
1145–1153.
G Model
RECYCL-3324; No. of Pages 10 ARTICLE IN PRESS
10 J. Su et al. / Resources, Conservation and Recycling xxx (2016) xxx–xxx
Gibson, G., Carlson, R., Simpson, J., et al., 2000. Nutrient Criteria Technical Guidance
Manual: Lakes and Reservoirs. United States Environment Protection Agency,
Washington DC.
Greer, M.A., Goodman, G., Pleus, R.C., Greer, S.E., 2002. Health effect assessment for
environmental perchlorate contamination: the dose response for inhibition of
thyroidal radioiodine uptake in humans. Environ. Health Prospects 110,
927–937.
Hauck, M., Maalcke-Luesken, F.A., Jetten, M.S.M., Huijbregts, M.A.J., 2016. Removing
nitrogen from wastewater with side stream anammox: what are the trade-offs
between environmental impacts? Resour. Conserv. Recycl. 107, 212–219.
Hayduk, L.A., 1996. LISREL: Issues, Debates and Strategies. John Hopkins University
Press, Baltimore.
Huo, S.L., Zan, F.Y., Chen, Q., Xi, B.D., et al., 2012. Determining reference conditions
for nutrients, chlorophyll a and Secchi depth in Yungui Plateau ecoregion
lakes, China. Water Environ. J. 26, 324–334.
Huo, S.L., Ma, C.Z., Xi, B.D., Su, J., et al., 2013a. Establishing eutrophication
assessment standards for four lake regions, China. J. Environ. Sci. 10,
2014–2022.
Huo, S.L., Xi, B.D., Ma, C.Z., Liu, H.L., 2013b. Stressor-response models: apractical
application for the development of lake nutrient criteria in China. Environ. Sci.
Technol. 47, 11922–11923.
Huo, S.L., Ma, C.Z., Xi, B.D., Gao, R.T., et al., 2014. Lake ecoregions and nutrient
criteria development in China. Ecol. Indic. 11, 1–10.
Hurlimann, A., Hemphill, E., McKay, J., Geursen, G., et al., 2008. Establishing
components of community satisfaction with recycled water use through a
structural equation model. J. Environ. Manage. 88, 1221–1232.
Ji, D.F., Xi, B.D., Su, J., Huo, S.L., et al., 2013. A model to determine the lake nutrient
standards for drinking water sources in Yunnan-Guizhou Plateau Ecoregion,
China. J. Environ. Sci. 9, 1773–1783.
Jiang, T.T., Huo, S.L., Xi, B.D., Su, J., et al., 2014. The influences of land-use changes
on the absorbed nitrogen and phosphorus loadings in the drainage basin of
lake Chaohu, China. Environ. Earth Sci. 5, 4165–4176.
Kallis, G., Butler, D., 2001. The EU water framework directive: measures and
implications. Water Policy 3, 125–142.
Kenney, M.A., 2007. Which Nutrient Criteria Should States and Tribes Choose to
Determine Waterbody Impairment? Using Science and Judgments to Inform
Decision-Making Dissertation. Duke University.
Kline, R.B., 1998. Principles and Practice of Structural Equation Modeling. The
Guilford Press, New York.
La Peyre, M.K., Mendelssohn, I.A., Reams, M.A., Templet, P.H., Grace, J.B., 2001.
Identifying determinants of nations’ wetland management programs using
structural equation modeling: an exploratory analysis. Environ. Manage. 27,
859–868.
Li, J., Tam, V.W.Y., Zuo, J., Zhu, J., 2015. Designers’ attitude and behaviour towards
construction waste minimization by design: a study in Shenzhen, China.
Resour. Conserv. Recycl. 105, 29–35.
Po, M., Nancarrow, B.E., Leviston, Z., Poter, N.B., Syme, G.J., Kaercher, J.D., 2005.
Predicting Community Behavior in Relation to Wastewater Reuse: What Drives
Decisions to Accept of Reject? CSIRO, Perth.
Reckhow, K.H., Arhonditsis, G.B., Kenney, M.A., et al., 2005. A predictive approach
to nutrient criteria. Environ. Sci. Technol. 9, 2913–2919.
Rook, J.J., 1974. Formation of haloforms during chlorination of natural water.
Water Treat. Exam. 23, 234–242.
Russell, S., Hampton, G., 2006. Challenges in understanding public responses and
providing effective public consultation on water reuse. Desalination 187,
215–227.
Please cite this article in press as: Su, J., et al., Developing surface water quality standards in China. Resour Conserv Recy (2016),
http://dx.doi.org/10.1016/j.resconrec.2016.08.003
SEPA, 1983. Environmental Quality Standard for Surface Water (GB3838-83). State
Environmental Protection Administration of China, Beijing.
SEPA, 1988. Environmental Quality Standard for Surface Water (GB3838-88). State
Environmental Protection Administration of China, Beijing.
SEPA, 1999. Environmental Quality Standard for Surface Water (GHZB 1-1999).
State Environmental Protection Administration of China, Beijing.
SEPA, 2002. Environmental Quality Standard for Surface Water (GB3838-2002).
State Environmental Protection Administration of China, Beijing of Regula.
Sutton-Grier, A.E., Kenney, M.A., Richardson, C.J., 2010. Examining the relationship
between ecosystem structure and function using structural equation
modelling: a case study examining denitrification potential in restored
wetland soils. Ecol. Modell. 221, 761–768.
USEPA, 1980. Ambient Water Quality Criteria. Office of Regulation Standards.
Criteria and Standards Division, Washington, DC.
USEPA, 1997. Guidelines for Preparation of the Comprehensive State Water Quality
Assessments. Environmental Protection Agency, Office of Water, Washington,
DC.
USEPA, 2000. Nutrient Criteria Technical Guidance Manual: Lakes and Reservoirs.
Environmental Protection Agency, Office of Water, Washington, DC,
EPA822-B00-001.
USEPA, 2001. Ambient Water Quality Criteria Recommendations: Information
Supporting the Development of State and Tribal Nutrient Criteria for Lakes and
Reservoirs in Nutrient Ecoregion IX. Office of water, Washington, DC.
USEPA, 2002. Edition of the Drinking Water Standards and Health Advisories.
Office of water, Washington, DC.
USEPA, 2008. States Adoption of Numeric Nutrient Standards (1998–2008). Office
of water, Washington, DC.
USEPA, 2010. Using Stressor-response Relationships to Derive Numeric Nutrient
Criteria. Environmental Protection Agency, Office of water, Washington, DC.
WHO, 1993. Guidelines for Drinking-water Quality, 2nd ed. Office of water,
Washington, DC.
WHO, 2003. Guidelines for Drinking-Water Quality, 3nd ed. World Health
Organization, Washington, DC.
Walker, J.L., Younos, T., Zipper, C.E., 2007. Nutrients in Lakes and Reservoirs-a
Literature Review for Use in Nutrient Criteria Development. Virginia Water
Resources Research Center, Virginia.
William, M.L., Wayne, A.W., Hans, W.P., 2011. Rationale for control of
anthropogenic nitrogen and phosphorus to reduce eutrophication of inland
waters. Environ. Sci. Technol. 45, 10300–10305.
Heiskary, S.A., Wilson, C.B., 2005. Minnesota Lake Water Quality Assessment
Report: Developing Nutrient Criteria, 3rd ed. Minnesota pollution control
agency, Minnesota.
Wu, H., Yuan, Z., Zhang, Y., Gao, L., Liu, S., 2014. Life-cycle phosphorus use
efficiency of the farming system in Anhui Province, Central China. Resour.
Conserv. Recycl. 83, 1–14.
Xu, Y.F., Ma, C.Z., Huo, S.L., Zhang, D.Y., et al., 2014. Establishing reference
conditions for lake water quality: a novel extrapolation approach. Water Res.
Manag. 6, 2161–2178.
Zan, F.Y., Huo, S.L., Xi, B.D., Zhu, C.W., et al., 2012. A 100-year sedimentary record of
natural and anthropogenic impacts on a shallow eutrophic lake, Lake Chaohu,
China. J. Environ. Monit. 14, 804–816.
Zhou, Y., Khu, S.T., Xi, B.D., Su, J., et al., 2014. Status and challenges of water
pollution problems in China: learning from the European experience. Environ.
Earth Sci. 72, 1243–1254.