Abstract

Lake Dianchi is severely polluted with nitrogen (N) and phosphorus (P). The effects of implementing environmental policies and technologies on future lake quality are not well understood. We analyse effects of environmental policies and technologies on future river export of nutrients into Lake Dianchi. We develop scenarios for 2050 and analyse these with the existing MARINA-Lakes model (Model to Assess River Inputs of Nutrient to lAkes). The scenarios differ in assumptions about future nutrient management in agriculture, sewage systems and mining. In the SSP3 (Shared Socio-economic Pathway 3) scenario, river export of nutrients to Lake Dianchi is projected to increase 1.4–4.4 times between 2012 and 2050. In the Current Policies scenario, rivers may export fewer nutrients than in SSP3, but this may not avoid eutrophication. Effects of improved nutrient management on river export of nutrients differ among nutrient forms, sub-basins and sources (e.g., urbanization in the north, agriculture in the middle and south). Pollution levels can be reduced below the 2012 level in an Optimistic scenario assuming advanced wastewater treatment, improved nutrient management in agriculture and no mining. However, even this may not completely prevent eutrophication. Preventing eutrophication requires even more efforts, for example, in implementing circular-oriented management options.

INTRODUCTION

Lake Dianchi is located in Yunnan province, China. Lake Dianchi is the sixth largest lake in China and the largest freshwater lake in Yunnan. The lake is an important source of drinking water for many cities. Water from the lake is also used for industrial and agricultural activities (Gu et al. 2016). However, the water quality of the lake has been deteriorating since the 1970s (Liu & Qiu 2007). As a result, the lake has been suffering from serious eutrophication problems such as harmful algal blooms over the past years. Algal blooms happen frequently in this lake due to high concentrations of phosphorus (P) and nitrogen (N) (Tong et al. 2017).

High concentrations of N and P in the lake are largely related to intensive human activities on land. These human activities include residential and industrial sewage, fertilizers usage in agriculture and P mining (Huang et al. 2014; Li et al. 2019). Rivers export these nutrients to Lake Dianchi. However, large amounts of nutrients stay in the lake because of the limited outflow from the lake. As a result, eutrophication in the lake has occurred (Huang et al. 2014; Zhou et al. 2014; Gao et al. 2015; Wang et al. 2016).

Many studies have been carried out for Lake Dianchi. They mainly focus on monitoring and modelling the current water quality status (Liu et al. 2004, 2009; Gao et al. 2014; Zhou et al. 2015). However, few studies exist on how river export of nutrients from sub-basins influence the water quality of the lake in the future. A study quantified river export of nutrients by source from sub-basins to Lake Dianchi for the year 2012 (Li et al. 2019). However, future trends in river export of nutrients are not well studied. In addition, the effects of implementing environmental policies and technologies on the lake quality in the future are not well understood.

The national and local government have introduced a number of policy documents to reduce the eutrophication problem of the lake. The lake is included in a national environmental project called ‘Three rivers and three lakes’ since 1996. National plans are published every five years. These plans set goals to improve the water quality of the lake to meet the national standard of Water quality for Surface Water (GB3838-2002) (Liu & Wang 2016). In 1999, a local policy ‘Zero o'clock action’ was published to control wastewater emissions from companies around the lake watershed (Zhang et al. 2014). Since 2006, projects such as wastewater collection and treatment, and sediments dredging have been introduced to improve the water quality of the lake. These projects amounted to about 2.8 billion USD (Liu & Wang 2016).

The Chinese government has already realized the importance of controlling nutrient inputs to water systems from agriculture. For example, in 2015, the Ministry of Agriculture published a policy ‘Zero Growth in Fertilizer Use by 2020’ to reduce over-fertilization of cropland and increase recycling of animal manure to replace fertilizers (MOA 2015). In 2018, Kunming City introduced a local policy ‘The implementation plan of the three-year action on the protection and management of Lake Dianchi’. This plan is valid from 2018 to 2020. Approximately 2.05 billion USD is set to be invested in the plan. A better understanding of the effects of these environmental policies on the lake quality is still required.

Several models exist to quantify nutrient flows from land to rivers, seas and lakes in China (Lewis et al. 2007; Ma et al. 2010; Gao et al. 2014; Strokal et al. 2016). Water quality models exist with different spatial and temporal levels of detail (e.g., SWAT, SPARROW, IMAGE-GNM) (Parton et al. 1998; Beusen et al. 2015). The Global NEWS-2 model (Nutrient Export from WaterSheds) is an example that quantifies river export of N and P by source for over 6,000 rivers in the world (Mayorga et al. 2010; Seitzinger et al. 2010). Another example is the MARINA model (Model to Assess River Inputs of Nutrients to seAs), developed for China based on the modelling approaches of Global NEWS-2 (Strokal et al. 2016). The MARINA model quantifies past and future river export of nutrients at the sub-basins scale for large Chinese rivers. MARINA has been used for the future analysis of coastal eutrophication.

Recently, the MARINA-Lakes model has been implemented to Lake Dianchi for the year 2012 (Li et al. 2019). MARINA-Lakes has been used to better understand the impact of nutrient loadings on the lake in a spatially explicit way. This has been done for different nutrient forms: dissolved inorganic (DIN, DIP) and dissolved organic (DON, DOP) N and P. Results of the MARINA-Lakes model indicate that agriculture, urbanization and mining are important causes of the lake pollution with nutrients today. There is a need for future analyses to reduce this pollution.

The main objective of this study is to analyse the effects of environmental policies and technologies to reduce river export of nutrients to Lake Dianchi in 2050. To this end, we develop five scenarios and implement them into the MARINA-Lakes model for 2050. We analyse future trends in river export of nutrients by source from sub-basins to Lake Dianchi for the period of 2012–2050. Our analyses focus on river export of DIN (dissolved inorganic N), DIP (dissolved inorganic P), DON (dissolved organic N) and DOP (dissolved organic P).

METHODS

Study area

Lake Dianchi is located in southwest of China in Yunnan province. The lake is surrounded by highly urbanized areas, intensive agricultural and mining activities. The surface area of the lake is about 309 km2, with 39 km of length and 13.5 km of width. The annual average depth of the lake is 4.9 m. The maximum depth is 10.1 m, and the volume of the lake is around 13.9 billion m3 of water (Li et al. 2019). The watershed area of the lake comprises 15 sub-basins (Gao et al. 2014). These 15 sub-basins are used in the MARINA-Lakes model (see below). In this study, we define Panlong, Caohai, Daqing, Haihe, Baoxiang sub-basins as the northern sub-basins, Xian, Maliao, Luolong, Laoyu, Nanchong and Yuni sub-basins as the middle sub-basins, and Gucheng, Dongda, Cigang and Baiyu sub-basins as the southern sub-basins (Figure 1).

Figure 1

Lakes Dianchi and its sub-basins. Source: MARINA-Lakes model (Li et al. 2019).

Figure 1

Lakes Dianchi and its sub-basins. Source: MARINA-Lakes model (Li et al. 2019).

The MARINA-Lakes model

MARINA-Lakes quantifies annual river export of DIN, DIP, DON and DOP by source from 15 sub-basins to Lake Dianchi. This is done as a function of human activities on land taking into account sub-basin characteristics (e.g., land use, hydrology) and nutrient retentions in soils and rivers. Full explanations and equations are in Appendix A and Li et al. (2019).

The main equation of the MARINA-Lakes model is: 
formula
(1)
where,

F is the form of nutrients: dissolved inorganic (DIN, DIP) and dissolved organic (DON, DOP) nitrogen (N) and phosphorus (P).

is the annual river export of nutrient form F by source y from sub-basin j (kg/year).

is the input of nutrient form F to rivers by source y from sub-basin j (kg/year). is divided into diffuse () and point () sources.

is the fraction of nutrients exported to the outlet of sub-basin j (0–1).

is the fraction of nutrients exported from sub-basin outlets to the river mouth (into the lake) (0–1).

  • Nutrient inputs to rivers from diffuse sources ()

Diffuse sources include the use of synthetic fertilizers, animal manure and human waste on land. These are diffuse sources of DIN, DON, DIP and DOP in rivers. Atmospheric N deposition and biological N2 fixation by vegetation are diffuse sources of DIN in rivers. Weathering of P-contained minerals and P mining are diffuse sources of DIP in rivers. Leaching of organic matter from soils is a diffuse source of DON and DOP in rivers (Table 1). is the sum of nutrient inputs to rivers in the sub-basins from all diffuse sources (kg/year, see Equation (2)). 
formula
(2)
where,
Table 1

Summary of the sources for model inputs and parameters for 2012 and 2050

Model inputs and parametersUnitSources for 2012Sources for 2050
Sub-basin area km2 Gao et al. (2014)  Gao et al. (2014)  
Population people Chen (2013)  Samir & Lutz (2017); Cuaresma (2017)  
Synthetic fertilizer use kg/year Wang et al. (2017)  Estimated based on Wang et al. (2017)  
Animal manure applied on land kg/year Strokal et al. (2016); Wang et al. (2017)  Estimated based on Wang et al. (2017)  
Human waste applied on land kg/year Strokal et al. (2016)  Estimated 
Atmospheric N deposition kg/year Xu et al. (2015)  Xu et al. (2015)  
Biological N2 fixation by crops kg/km2/yeara Strokal et al. (2016)  Strokal et al. (2016)  
P weathering kg/km2/yeara Strokal et al. (2016)  Strokal et al. (2016)  
Organic leaching from soils kg/km2/yeara Strokal et al. (2016)  Estimated based on Strokal et al. (2016)  
Mining kg/year Li et al. (2019)  Li et al. (2019)  
Animal manure directly discharged (untreated) kg/year Wang et al. (2017)  Estimated based on Wang et al. (2017), see Table 2  
Human waste directly discharged (untreated) kg/year Wang et al. (2017)  Wang et al. (2017)  
Sewage systems kg/year Li et al. (2019)  Estimated based on Li et al. (2019) b 
Actual river discharges km3/year Jin (2003)  Jin (2003)  
Water removal from the river system 0–1 Jin (2003)  Jin (2003)  
Nutrient retention in reservoirsc 0–1 Strokal et al. (2016)  Strokal et al. (2016)  
Nutrient retention in water systems ()c 0–1 Strokal et al. (2016)  Strokal et al. (2016)  
Model inputs and parametersUnitSources for 2012Sources for 2050
Sub-basin area km2 Gao et al. (2014)  Gao et al. (2014)  
Population people Chen (2013)  Samir & Lutz (2017); Cuaresma (2017)  
Synthetic fertilizer use kg/year Wang et al. (2017)  Estimated based on Wang et al. (2017)  
Animal manure applied on land kg/year Strokal et al. (2016); Wang et al. (2017)  Estimated based on Wang et al. (2017)  
Human waste applied on land kg/year Strokal et al. (2016)  Estimated 
Atmospheric N deposition kg/year Xu et al. (2015)  Xu et al. (2015)  
Biological N2 fixation by crops kg/km2/yeara Strokal et al. (2016)  Strokal et al. (2016)  
P weathering kg/km2/yeara Strokal et al. (2016)  Strokal et al. (2016)  
Organic leaching from soils kg/km2/yeara Strokal et al. (2016)  Estimated based on Strokal et al. (2016)  
Mining kg/year Li et al. (2019)  Li et al. (2019)  
Animal manure directly discharged (untreated) kg/year Wang et al. (2017)  Estimated based on Wang et al. (2017), see Table 2  
Human waste directly discharged (untreated) kg/year Wang et al. (2017)  Wang et al. (2017)  
Sewage systems kg/year Li et al. (2019)  Estimated based on Li et al. (2019) b 
Actual river discharges km3/year Jin (2003)  Jin (2003)  
Water removal from the river system 0–1 Jin (2003)  Jin (2003)  
Nutrient retention in reservoirsc 0–1 Strokal et al. (2016)  Strokal et al. (2016)  
Nutrient retention in water systems ()c 0–1 Strokal et al. (2016)  Strokal et al. (2016)  

Details are in Li et al. (2019) for model inputs for the year 2012. Table 2 provides details of how we derived model inputs for the year 2050 (see also Table B.2 in the Appendix).

aWe applied the rates for atmospheric N deposition and biological N2 fixation per km2 and multiplied with the sub-basin areas to get kg/year.

bWe estimated a number of people connected to sewage systems in 2050 taking into account the storylines of the scenarios; the connection rate of the population to sewage systems in 2012 is from Li et al. (2019).

cNutrient retentions for 2012 and 2050 are taken from Strokal et al. (2016) for the Jinsha sub-basin, and we applied it for all other sub-basins of Lake Dianchi because of data limitation.

is the input of nutrient form (F: DIN, DON, DIP or DOP) to rivers from anthropogenic (ant) diffuse (dif) source y in sub-basin j (kg/year). These diffuse sources include the use of synthetic fertilizers, animal manure and human waste on land, atmospheric N deposition over agricultural land and biological N2 fixation by crops. They are quantified as nutrient inputs to agricultural land that are corrected for nutrient export from agricultural land (crop harvesting and grazing) and retentions in soils (Li et al. 2019, Appendix B).

is the input of nutrient form (F: DIN, DON, DIP or DOP) to rivers from anthropogenic (ant) explicit (expl) diffuse source y in sub-basin j (kg/year). These anthropogenic explicit diffuse sources include weathering of P-contained minerals and leaching of organic matter over agricultural land. They are quantified using an export-coefficient approach (Li et al. 2019, Appendix B).

is the input of nutrient form (F: DIN, DON, DIP or DOP) to rivers from natural (nat) explicit (expl) diffuse (dif) source y in sub-basin j (kg/year). These natural explicit diffuse sources include weathering of P-contained minerals and leaching of organic matter over natural land. They are quantified using an export-coefficient approach (Li et al. 2019, Appendix B).

is the input of dissolved inorganic phosphorus (DIP) to rivers from mining activities (rock processing as diffuse source) in sub-basin j (kg/year). These inputs are quantified using the information on the production of DIP during mining activities (Li et al. 2019, Appendix B).

is the input of nutrient form (F: DIN, DON, DIP or DOP) to rivers from natural (nat) diffuse (dif) source y in sub-basin j (kg/year). These natural diffuse sources include atmospheric N deposition over natural areas and biological N2 fixation by natural vegetation. These inputs are quantified as inputs to natural land that are corrected for retentions in soils (Li et al. 2019, Appendix B).

  • Nutrient inputs to rivers from point sources (, kg/year)

Point sources of nutrients in rivers include direct discharges of untreated animal manure and human waste. These are point sources of DIN, DIP, DON and DOP in rivers. Mining is a point source of DIP in rivers (Table 1). Nutrient inputs to rivers from point sources are quantified using Equation (3). 
formula
(3)
where,

is the input of nutrient form (F: DIN, DIP, DON and DOP) to rivers from sewage systems ( direct discharges of untreated animal manure (and of untreated human waste (, kg/year), and from mining as a point source (, for DIP only, kg/year). Details about the equations to quantify these point source inputs to rivers are given in Appendix B (see also Li et al. 2019).

is the input of nutrient element (E: N or P) to rivers from point source y in sub-basin j (kg/year).

is the fraction of nutrient element that enters rivers from point sources in form F (DIN, DIP, DON and DOP) (0–1).

  • Nutrient retentions in rivers

The nutrient export fractions are quantified using Equation (4). 
formula
(4)
where,

is the fraction of nutrient form (F: DIN, DIP) retained in and lost from river systems in sub-basin j due to denitrification (for DIN) and sedimentation (for DIP) (0–1).

is the fraction of nutrient form (F: DIN, DIP) retained in reservoirs in sub-basin j (0–1).

is the fraction of nutrient form (F: DIN, DIP, DON, DOP) removed from sub-basin j through water consumption (0–1). Details are in Li et al. (2019).

In Lake Dianchi, all sub-basins drain directly into the lake, thus .

  • Model inputs

The nutrient export to Lake Dianchi in 2012 is quantified using the MARINA-Lakes model from Li et al. (2019). The model inputs and their sources are provided by Li et al. (2019) for 2012 (see also Table B.1 in the Appendix). Here, we summarize the sources of the main model inputs and parameters for Lake Dianchi in Table 1. Table 2 gives more details on how we derive model inputs for 2050 (see section on scenarios below).

Table 2

Description of the five scenarios to quantify nutrient export by rivers from sub-basins to Lake Dianchi in 2050

Human activities2012Scenarios for 2050
SSP3CPATIAMOPT
Agriculture Manure excretion (ton/km2/year) 3 for N and 1.5 for P 90% increase between 2012 and 2050a As SSP3 As SSP3 As SSP3 As SSP3 
Manure recycling on land (% of the total excretion) 52%a for N1 and P As 2012 57% for N1 and P As SSP3 62% for N1 and P As IAM 
Manure discharges to rivers (% of the total excretion) 10% As 2012 5%1 As SSP3 0%1 As IAM 
Synthetic fertilizer use (ton/km2/year) 8.5 for N and 1.3 for P 26% increase between 2012 and 2050a At the level of 2020 As SSP3 30% decrease between 2012 and 2050 As IAM 
Sewage systems Human waste directly discharged to rivers (untreated) from urban and rural population (% of the total human excretion) For N: 13% for urban and 23% for rural population; For P: 13% for urban and 47% for rural population As 2012 As 2012 As SSP3 As SSP3 As 2012 
Total population (103 people) 103 urban and 368 rural As 2012b As SSP3 As SSP3 As SSP3 As SSP3 
People connected to sewage systems (% of the population) 94% for urbanc and 55% for ruralc As 2012 97% for urban and 60% for rural 99% for urban, 70% for rural As SSP3 AS AT 
Nutrient removal efficiencies (%) 35% for Nc, 67% for Pc As 2012 65% for N, 83% for P2 90% for N and 95% for P As SSP3 AS AT 
Mining Mining activities (ton/year) 28 ton of DIPc As 2012 Ban As SSP3 As CP As IAM 
Human activities2012Scenarios for 2050
SSP3CPATIAMOPT
Agriculture Manure excretion (ton/km2/year) 3 for N and 1.5 for P 90% increase between 2012 and 2050a As SSP3 As SSP3 As SSP3 As SSP3 
Manure recycling on land (% of the total excretion) 52%a for N1 and P As 2012 57% for N1 and P As SSP3 62% for N1 and P As IAM 
Manure discharges to rivers (% of the total excretion) 10% As 2012 5%1 As SSP3 0%1 As IAM 
Synthetic fertilizer use (ton/km2/year) 8.5 for N and 1.3 for P 26% increase between 2012 and 2050a At the level of 2020 As SSP3 30% decrease between 2012 and 2050 As IAM 
Sewage systems Human waste directly discharged to rivers (untreated) from urban and rural population (% of the total human excretion) For N: 13% for urban and 23% for rural population; For P: 13% for urban and 47% for rural population As 2012 As 2012 As SSP3 As SSP3 As 2012 
Total population (103 people) 103 urban and 368 rural As 2012b As SSP3 As SSP3 As SSP3 As SSP3 
People connected to sewage systems (% of the population) 94% for urbanc and 55% for ruralc As 2012 97% for urban and 60% for rural 99% for urban, 70% for rural As SSP3 AS AT 
Nutrient removal efficiencies (%) 35% for Nc, 67% for Pc As 2012 65% for N, 83% for P2 90% for N and 95% for P As SSP3 AS AT 
Mining Mining activities (ton/year) 28 ton of DIPc As 2012 Ban As SSP3 As CP As IAM 

SSP3 is Shared Socio-economic Pathway 3. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Source: MARINA-Lakes model (see Methods section).

1Manure recycling rates are quantified as the amount of recycled manure on land divided by the amount of the total manure excretion. The recycled manure is quantified from the total manure excretion that is corrected for N losses to air and direct discharges to rivers. N losses to air are from Li et al. (2019). The fraction of direct manure discharges for 2012 is from Li et al. (2019) and assumed for 2050 depending on scenario. For CP scenario, we assumed that manure recycling will be increased as a result of decreased direct manure discharges to rivers compared to 2012. For IAM scenario, we assumed that direct discharges to rivers will be fully avoided. As a result, more animal manure will be applied on land.

2This nutrient removal efficiencies are derived using literature on the advanced technologies A²/O and ICEAS (Liu et al. 2007; Zhang et al. 2008) implemented by a wastewater treatment company in Kunming City, we assumed that the whole sub-basin will implement this technology by 2050 in CP scenario (see also Appendix C for more information about policies and technologies).

Scenarios

SSP3 (Shared Socio-economic Pathway 3) is used as a worst-case for Lake Dianchi (O'Neill et al. 2014; Fujimori et al. 2017; Riahi et al. 2017). Four alternative scenarios are developed based on the SSP3 scenario: CP, AT, IAM and OPT scenarios. These four alternative scenarios are described below. Table 2 summarizes the main descriptions of the scenarios.

SSP3 was interpreted for the Chinese food system for 2050 (Wang et al. 2017). For Lake Dianchi, SSP3 assumes that population will increase relatively fast until 2030. However, by 2050 the total population may decrease, reaching the level of 2012. Population in China is expected to decrease between 2030 and 2050 as the net effect of the ‘One-Child’ policy (Goodkind 2017; Zhang 2017). The production of animal manure is projected to increase by 90% between 2012 and 2050 in all sub-basins. This is because more people will move to cities, the demand for food in cities might increase. People are expected to have more preferences for meat products (Wang et al. 2017). In this scenario, policies for nutrient management in agriculture, sewage systems and mining are not effectively implemented. As a consequence, synthetic fertilizer use on land is expected to increase by 26% (Wang et al. 2017) between 2012 and 2050 for all sub-basins (Table 2). Manure recycling rates will stay at the level of 2012 (52% of the total excretion). Fractions of direct discharges of manure to rivers in this scenario are expected to stay as in 2012. The same holds for untreated human waste. The same fraction of the urban and rural population will be connected to sewage systems in 2050 as in 2012 with the higher number of people in cities. Wastewater treatment technologies might not be improved largely compared to the level of 2012. Mining activities will continue as in 2012.

The CP scenario reflects implementation of the Current environmental Policies for agriculture, mining and urbanization (Table 2). In this scenario, implementation of the following national policy is assumed for agriculture: ‘Zero growth in Synthetic Fertilizers Use by 2020’ (MOA 2015). This current policy also calls for a 60% recycling rate of manure on land. Thus, we assume that the percentage of direct manure discharges to rivers will decrease to 5% in 2050 (Table 2, see also Appendix C). As a result, the amount of manure use on land will increase in 2050 compared to SSP3. The amount of synthetic fertilizers will stay at the level of 2020. The total population will stay at the level of 2012 as in SSP3. However, the fraction of urban (97% in 2050) and rural (60% in 2050) population connected to sewage systems will increase; this is in line with the ‘Lake Dianchi protection regulations in Yunnan province’ (KLDA 2013). Improved technologies to treat wastewater will be implemented. This implies that nutrient removal efficiencies will be higher in 2050 than in 2012: 65% for N and >80% for P in all sub-basin. All mining activities will be banned (KGO 2017). More information about policies and technologies are in Appendix C (Tables C.1 to C.5).

The AT scenario assumes the full implementation of Advanced Technologies for wastewater treatment. In this scenario, the population size, urbanization, economy and agricultural activities are the same as in SSP3. The only difference from SSP3 is in advanced technologies for wastewater treatment and in an increasing number of people connected to sewage systems. Some technologies could also reach very high nutrient removal efficiencies in theory, such as reverse osmosis (RO) and anaerobic ammonium oxidation (anammox). RO is a pressure driven membrane process to separate dissolved components in permeates, and is more important in industrial applications (Kucera; Wenten 2016). Anammox is a process which directly oxidizes ammonium to dinitrogen gas under anoxic condition, it is promising for N removal from wastewater (Tang et al. 2017) (see also Table C.4 in the Appendix). The connection rate of population to sewage systems will be 99% for urban people and 70% for rural people. The nutrient removal efficiencies are 90% and 95% for N and P, respectively (Table 2).

The IAM scenario assumes Improved nutrient use efficiencies in Agriculture and stopping Mining activities. Improved nutrient management in agriculture implies high animal manure recycling on land and less synthetic fertilizer use with the same socio-economic development as in SSP3. This scenario differs from the CP scenario in that it has well-implemented policies in agriculture and mining industry (KPGOYP 2008; MOA 2015; KGO 2017). These policies are extended with advanced technologies to manage manure and nutrients in agriculture. Examples are injection of manure to avoid losses of nutrients to the environment during application (Wang et al. 2017). These nutrient management options can increase N and P use efficiencies in food production and thus reduce losses of N and P to rivers. For the sub-basins of Lake Dianchi, all available animal manure is assumed to be recycled on land (100% for N and P). Thus, direct discharges of manure will be avoided. Synthetic fertilizer use on land will be reduced by 30% compared to 2012 to avoid over-fertilization (Table 2). All mining activities along Lake Dianchi are assumed to be stopped.

The OPT scenario combines assumptions of the AT and IAM scenarios. OPT assumes an optimistic world with improved nutrient management in agriculture, no mining activities and implemented best available technologies. The population, urbanization and economy growth will stay as in SSP3. Almost all urban and over two-thirds of the rural population will be connected to sewage systems. Nutrient removal efficiencies will be high in wastewater treatment plants (90% for N, 95% for P). The use of synthetic fertilizer is expected to reduce by 30% in 2050 compared to 2012, and manure cycling rates could increase as in IAM. Mining activities will be forbidden in 2050.

RESULTS

Future trends in river export of nutrients by source

In 2012, rivers exported large amounts of nutrients to Lake Dianchi (Figure 2). We quantify that about 5,000 ton of DIN and over 1,000 ton of DON were exported to the lake from 15 sub-basins (Figure 1). In addition, we quantify around 300 ton of DIP and 100 ton of DOP in 2012 (Figure 2). Point sources accounted for over half of the nutrients in the lake except for DIN. Less than half of DIN in Lake Dianchi resulted from sewage systems and around 40% from synthetic fertilizers. Around half of DON in the lake (651 ton) was from sewage systems. For DIP, sewage systems were responsible for over 80% of DIP in the lake. Mining was an important source of DIP in individual rivers in the southern sub-basins. For DOP, direct discharges of animal manure and of untreated human waste (point sources) contributed around 80% of DOP to the lake.

Figure 2

Annual river export of dissolved inorganic (DIN, DIP) and dissolved organic (DON, DOP) nitrogen (N) and phosphorus (P) by source to Lake Dianchi in 2012 and 2050 according to the scenarios (t/year). SSP3 is a Shared Socio-economic Pathway 3 scenario. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Other: atmospheric N deposition and biological N2 fixation for DIN, leaching of organic matter for DON and DOP, and weathering of P-contained mineral for DIP, detergents for DIP and DOP.

Figure 2

Annual river export of dissolved inorganic (DIN, DIP) and dissolved organic (DON, DOP) nitrogen (N) and phosphorus (P) by source to Lake Dianchi in 2012 and 2050 according to the scenarios (t/year). SSP3 is a Shared Socio-economic Pathway 3 scenario. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Other: atmospheric N deposition and biological N2 fixation for DIN, leaching of organic matter for DON and DOP, and weathering of P-contained mineral for DIP, detergents for DIP and DOP.

In the future, river export of nutrients is projected to increase by 1.4–4.4 times between 2012 and 2050 in SSP3, depending on nutrient form (Figure 2). River export of DIN and DON is projected to increase by 1.4 times, reaching around 8,000 ton and 1,600 ton in 2050, respectively. River export of DIP is projected to increase by 4.4 times, reaching 1,400 ton in 2050. River export of DOP is, however, projected to decrease slightly during 2012–2050 in the SSP3 scenario. The increases in river export of the nutrients are associated with increasing urbanization and intensive agriculture in 2050. Point sources are simulated to remain the main cause of nutrient pollution in the lake (except for DIN where fertilizers are also important). For DIN, sewage systems and synthetic fertilizers are projected to remain important sources of DIN in the lake in 2050. These sources are projected to contribute around 2,700 ton (from sewage) and 2,000 ton (from fertilizers) of DIN to the lake in 2050. Above half of DON is simulated to result from sewage systems in 2050. Mining activities project to continue contributing to river export of DIP from individual sub-basins in the south. Sewage systems are projected to contribute more than 80% of DIP in the lake. For DOP, direct discharges of animal manure are expected to remain the dominant source together with human waste (treated and untreated).

Current Policies (CP) may reduce river export of nutrients to the lake by 2050, but not below the level of 2012 (CP scenario, Figure 2). In 2050, river export of DIN and DOP is slightly higher than in 2012, and river export of DON and DIP is 1.2 and 2.5 times as high as in 2012, respectively. Effects of the current policies on reducing river export of nutrients differ among nutrient forms. Current policies are projected to reduce more DIN and DOP export by rivers than river export of the other nutrient forms compared to SSP3. This is associated with the sources of the nutrients in rivers. Current policies aim largely at reducing over-fertilization of cropland and recycling of manure on land. These are important sources of DIN (fertilizers) and DOP (manure) in rivers. This is different for river export of DIP where sewage systems contribute considerably to DIP in rivers. This also holds for DON in rivers (Figure 2). Mining activities will not add DIP in rivers because they will be completely banned in this scenario.

Implementing advanced technologies for wastewater treatment is more effective to reduce future river export of DIP and DON than of DIN and DOP (AT scenario, Figure 2). In this AT scenario, river export of DON and DIP is lower than in the CP scenario. River export of DIN is comparable to the level in the CP scenario whereas river export of DOP is higher than in the CP scenario. River export of DON is projected to below the level of 2012 in 2050. These trends can be explained by the high nutrient removal efficiencies for N and P in the AT scenario (Table 2). For reducing DIN and DOP in rivers, improving wastewater treatment in sewage systems is not enough because DIN and DOP are largely originated from fertilizers and manure (Figure 2).

Implementing improved nutrient management in agriculture and banning mining is more effective to reduce future river export of DIN and DOP than of DIP and DON (IAM scenario, Figure 2). River export of DIN and DOP is expected to reach the level of 2012 in 2050. This is because DIN and DOP in rivers originate largely from agricultural-related sources such as direct discharges of manure (for DOP) and synthetic fertilizers (for DIN). Reducing manure discharges (point sources) and over-fertilization of cropland by synthetic fertilizers (diffuse sources) according to the improved nutrient management in agriculture will largely reduce river export of DIN and DOP. This is not the case for DON and DIP where sewage systems (point sources) are the main causes.

The OPT scenario indicates that it is possible to reduce future river export of all nutrients to below the level of 2012. This scenario assumes improved wastewater treatment and implemented improved nutrient management in agriculture (Figure 2). For example, river export of nutrients is projected to reduce by a factor of 36% for DIN, 61% for DON, 22% for DIP and 60% for DOP between 2012 and 2050 in the OPT scenario (Figure 2). These reductions are related to the net effect of the management options to reduce pollution from cities (advanced wastewater treatment), agriculture (improved nutrient management) and mining (banned).

Future trends in river export of nutrients by sub-basin

River export of nutrients differs among sub-basins and nutrient forms (Figures 3 and 4). Caohai and Daqing sub-basins (north) are the two most polluted sub-basins (Figure 1). For example, they exported 0.3–10 times higher DIN to the lake in 2012 than the other sub-basins (except for the Panlong sub-basin). This also holds for other scenarios for 2050 (Figures 3 and 4).

Figure 3

Annual river export of dissolved inorganic (DIN) and dissolved organic (DON) nitrogen (N) by sub-basin (kg/km2 of the sub-basin area/year) and the share of the sources in annual river export of DIN and DON (0–1) in 2012 and 2050 according to the scenarios. SSP3 is a Shared Socio-economic Pathway 3 scenario. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Source: MARINA-Lakes model (see Methods section).

Figure 3

Annual river export of dissolved inorganic (DIN) and dissolved organic (DON) nitrogen (N) by sub-basin (kg/km2 of the sub-basin area/year) and the share of the sources in annual river export of DIN and DON (0–1) in 2012 and 2050 according to the scenarios. SSP3 is a Shared Socio-economic Pathway 3 scenario. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Source: MARINA-Lakes model (see Methods section).

Figure 4

Annual river export of dissolved inorganic (DIP) and dissolved organic (DOP) phosphorus (P) by sub-basin (kg/km2 of the sub-basin area/year) and the share of the sources in annual river export of DIP and DOP (0–1) in 2012 and 2050 according to the scenarios. SSP3 is a Shared Socio-economic Pathway 3 scenario. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Source: MARINA-Lakes model (see Methods section).

Figure 4

Annual river export of dissolved inorganic (DIP) and dissolved organic (DOP) phosphorus (P) by sub-basin (kg/km2 of the sub-basin area/year) and the share of the sources in annual river export of DIP and DOP (0–1) in 2012 and 2050 according to the scenarios. SSP3 is a Shared Socio-economic Pathway 3 scenario. CP assumes the full implementation of the Current Policies relative to SSP3. AT assumes more urbanization and the full implementation of Advanced Technologies to treat wastewater relative to SSP3. IAM assumes the full implementation of Improved nutrient management in Agriculture and banned Mining relative to SSP3. OPT is an optimistic scenario combining the AT and IAM scenarios. Source: MARINA-Lakes model (see Methods section).

River export of DIN is projected to increase by 2050 in SSP3 (see the previous section). However, the sources of DIN in rivers differ among sub-basins in SSP3 (Figures 1 and 3). DIN is mainly from synthetic fertilizers and manure use on land in middle and southern sub-basins in 2050 SSP3. In the northern sub-basins, DIN in rivers is largely from sewage systems. In the CP scenario, all sub-basins are projected to export lower amounts of DIN compared to SSP3, but higher amounts of DIN compared to 2012 in 2050. In the AT scenario, improved sewage systems is more effective to reduce DIN from the northern sub-basins (Caohai and Daqing, Figure 3). In the IAM scenario, improved nutrient management in agriculture is effective to reduce DIN export from middle and southern sub-basins. In the OPT scenario, DIN reductions in river export from all sub-basins are higher compared to the other scenarios. For example, river export of DIN is projected to reduce by 20–65% (range for sub-basins) in OPT between 2012 and 2050.

Likewise, sources of DON in rivers differ among sub-basins (Figure 3). In SSP3, DON in rivers is mainly from sewage systems in northern sub-basins. In rivers of southern sub-basins, DON is largely from sewage systems and animal manure (untreated) (Figures 1 and 3). In the CP scenario, DON export is not effectively reduced. In this scenario, lower river export of DON is projected from only a few sub-basins compared to SSP3. This is because more efforts are needed to reduce direct manure discharges and improve wastewater treatment in the CP scenario. In the AT scenario, DON in rivers of sub-basins in the middle-northern part (Figure 1) is expected to reduce in 2050 compared to SSP3. This is because sewage systems in those sub-basins contribute largely to the river pollution. Advanced wastewater treatment in the AT scenario may reduce river pollution in those sub-basins. This is different for southern sub-basins (e.g., Cigang and Baiyu) where improved nutrient management in agriculture (IAM scenario) is more effective to reduce river pollution compared to SSP3. In the OPT scenario, river export of DON is projected to reduce from all sub-basins.

DIP in rivers of northern sub-basins is mainly from sewage systems in SSP3. Mining activities largely contribute to DIP in rivers of southern sub-basins in SSP3. Direct discharges of animal manure are important sources of DIP in rivers of southern sub-basins in SSP3. Overall, the CP scenario is less effective in reducing river export of DIP from most sub-basins. Exceptions are rivers of some southern sub-basins that may export less DIP to the lake. This is because CP assumes some improvements in nutrient management in agriculture and banning of mining that are sources of DIP in those rivers. In the AT scenario, river export of DIP from northern sub-basins (Figure 1) is a factor of 3–6 lower compared to SSP3 and CP. In the IAM scenario, policies (KPGOYP 2008; MOA 2015; KGO 2017) for better agricultural management and banning mining reduce river export of DIP from the southern sub-basins (Figure 4). In the OPT scenario, river export of DIP from all sub-basins is projected to be lower in 2050 than in 2012.

River export of DOP from southern sub-basins is generally higher than from northern sub-basins (Figure 4). This is different for other nutrient forms (see above). The main sources of DOP in rivers are untreated human excreta in northern sub-basins and direct discharges of animal manure to rivers in southern sub-basins. By 2050, river export of DOP from all sub-basins may increase in SSP3. Current policies are somewhat effective to reduce river export of DOP from some southern sub-basins. In the AT scenario, DOP export is much higher than in 2012 from most sub-basins. In the IAM scenario, river export of DOP is largely reduced compared to SSP3. This holds especially for some sub-basins in the southern part, where river export of DOP is projected to decrease by a factor of 2 between 2012 and 2050. In the OPT scenario, river export of DOP is 38% to 76% lower in 2050 than in 2012 for all sub-basins (range for the sub-basins).

DISCUSSION

Strengths of our study

Our study provides new insights in future trends in river export of different forms and their sources for Lake Dianchi. Existing studies focus often on the total N and P export to the lake (Chen et al. 2002; Cheng et al. 2008; Gao et al. 2015). Only a few studies distinguish the nutrient forms (Zhang et al. 2007; Qian et al. 2016; Liu 2017). However, the various nutrient forms in the lake may result from different sources. We show that synthetic fertilizers are important sources of DIN in the lake. This is different for DON, DIP and DOP. Sewage systems are important sources of DIP and DON and direct discharges of manure of DOP in the lake. Mining contributes DIP in southern rivers of the lake (Figure 4). Reduction of all nutrient forms in the lake may contribute to effective reduction of eutrophication (Pinckney et al. 2001). Our insights about the sources can help to reduce nutrient forms in the lake effectively. Furthermore, we account for most important sources of lake pollution. Existing studies often focus either on collected human waste (point source) or fertilizer use (diffuse source). For example, direct discharges of animal manure to rivers are often ignored in most of the studies (Wang et al. 2017).

Another strength of our study is our analysis of the future for sub-basins and their sources. We quantify river export of nutrients from 15 sub-basins and identify main sources of river pollution in those sub-basins. This helps to better understand the contribution of sub-basins to the total nutrient pollution of the lake. This information is relevant for identifying sub-basins specific management options. Sub-basin specific analyses for future trends in river pollution for Dianchi are limited in existing studies (Kong et al. 2012; Gao et al. 2014; Zhou et al. 2015). In this study, we perform such analyses using the MARINA-Lakes model. This model enables to quantify river export of different nutrient forms by sub-basin and source. To our knowledge, other models for Lake Dianchi that quantify simultaneously river export of nutrients in different forms from sub-basins and sources do not exist.

An important strength of our study is that we developed five different scenarios to analyse the future for Lake Dianchi. The lake has been eutrophic for over decades, and many studies tried to understand the causes and the consequences of the nutrient pollution. However, most of these studies are limited to analyses of past and current trends, and do not explicitly model future trends. National and local governments have invested considerably to improve water quality. However, the effects of the proposed and implemented environmental policies are not always well understood. Several existing studies discussed control measures for eutrophication (Wang 2002; Liu et al. 2012), but not at the sub-basin scale and not for different nutrient forms for 2050. We explore several future scenarios for river export of nutrients to Lake Dianchi. We focus, in particular, on effects of the environmental policies and advanced wastewater treatment to reduce nutrient pollution in the lake. We show possible effects of the current policies for agriculture (e.g., Zero Growth in Synthetic Fertilizers by 2020) and mining. These current policies are promising and are expected to reduce future river pollution compared to the SSP3 scenario. However, this may not be enough to avoid eutrophication in 2050. We argue that more effort might be needed in reducing both agricultural and urbanization related pollution. Our study may serve as an inspiration for further research on future trends.

Limitations of our study

Our study also has four main limitations. First, we may not include all existing environmental policies and technologies. Policies are often generic without explicit targets. Plenty of technologies exist and their nutrient removal efficiencies vary greatly (Oller et al. 2011; Foged et al. 2012; Jie et al. 2017). Local conditions (e.g., the capacity and the maintenance of the wastewater treatment plants) can influence the removal efficiencies. However, we believe that we considered most relevant environmental policies and accounted for most recent advanced technologies that are relevant for the study area.

Second, the MARINA-Lakes model has uncertainties. The sources of the uncertainties are largely associated with model inputs and parameters. For example, some model parameters (e.g., fractions of manure discharges) are generic for all sub-basins and based on expert knowledge and some literature. MARINA-Lakes has recently been implemented for Lake Dianchi for the year 2012 (Li et al. 2019). The model has been validated against observations for nutrients in the lake. Validation results are promising and indicate that the model can be used to analyse river export of nutrients to Dianchi (Li et al. 2019). We implemented the existing MARINA-Lakes model (Li et al. 2019) for 2050.

Third, a few sources might be missing in the MARINA-Lakes model. These include, for example, direct atmospheric N and P deposition on the lake. We account for atmospheric N deposition on land, but not on the lake water. Atmospheric deposition of N and P on water can potentially be a large source of nutrients in coastal water systems (Jassby et al. 1994; Paerl et al. 2002). However, atmospheric N and P deposition on the lake may not be a primary pollution source. However, this needs more investigation (Ren et al. 2019).

Fourth, future trends are uncertain. We developed five scenarios that have the same socio-economic development, but differ in nutrient management in agriculture, cities and mining. Socio-economic development is based on the existing SSP3 scenario (O'Neill et al. 2014; Fujimori et al. 2017; Riahi et al. 2017). For nutrient management, we interpreted relevant policies based on literature and expert knowledge. We assumed the full implementation of the policies and advanced technologies for wastewater treatment plants in all sub-basins in our OPT scenario. Such combinations of policies and the best technologies are simplified in the MARINA-Lakes model. In reality there may be synergies and conflicts that affect reduction options for different nutrient forms. One synergy could be that the recycling of animal manure leads to even more reductions in synthetic fertilizers than assumed in our IAM scenario, and thus fewer nutrients will be exported to rivers. Conflicts might happen when the implementation of advanced wastewater treatment technologies requires additional financial support on top of other management options (e.g., in our AT scenario). Implementing policies and advanced technologies in our optimistic scenario might, thus, be challenging. However, China has been developing fast over past decades, opening an opportunity for further implementation of the effective environmental policies and technologies. We quantify the technical feasibility of implementing solutions. Future research may focus on the political, societal and economic feasibility.

Implications for policies to reduce future eutrophication

Our study shows that it will be difficult to reduce coastal eutrophication in the future, even under the OPT scenario. We compare our quantified nutrient loadings to Lake Dianchi in 2050 with the critical nutrient loadings in 2012 (Li et al. 2019). These critical nutrient loadings are the levels of nutrients below which eutrophication may not occur. The critical nutrient loadings were quantified using the lake ecosystem model (PCLake) (Janse et al. 2008) and take into account the nutrient cycling in the lake and the food web (Li et al. 2019). For the northern part of Dianchi (Caohai sub-basin), these critical nutrient loadings are quantified at 3.06 mg N/m2/day and 0.34 mg P/m2/day in (Li et al. 2019). For the southern part of the lake (other 14 sub-basins), the critical nutrient loadings are quantified at 3.42 mg N/m2/d and 0.38 mg P/m2/d. We convert projected nutrient loadings for the year 2050 from kton/year to mg/m2/d using the surface area of the lake. We compare the N and P quantified in the OPT scenario to the critical loadings. Our comparison shows that the projected N and P loadings exceed their critical loadings by a factor of 27 for N and 25 for P for the northern part of the lake. For the southern part, this exceedance is a factor of 10 for N and 17 for P. It seems that these exceedances are high. However, they are much lower compared to the exceedance between the nutrient loadings in 2012 and the critical nutrient loadings (a factor of 17–82 depending on nutrients and lake part) (Li et al. 2019).

Our study indicates that advanced wastewater treatment technologies, improved nutrient management in agriculture and banned mining can help to reduce the gap between the projected nutrient loadings and their critical loadings in the lake (OPT scenario). This is a right direction towards reducing future eutrophication. However, more efforts are needed to avoid eutrophication in 2050. For example, more reductions are needed in the use of synthetic fertilizers than assumed in the OPT scenario. Decentralized wastewater treatment with high nutrient removal efficiencies for rural areas may help to avoid nutrient losses to rivers. We argue that there is a need to focus on circular-oriented management options (e.g., recover and recycle nutrients from waste). Furthermore, controlling internal release of nutrients from lake sediments is also needed. For example, studies show that a gypsum-based technique or the use of alum can prevent phosphorous releases from lake sediments (Varjo et al. 2003; Steinman et al. 2004). Besides, wetlands or buffer zones around the water systems can prevent nutrient releases to rivers and thus to Lake Dianchi (Yin & Lan 1995; Borin et al. 2001; Coveney et al. 2002). Efficient use of water from the lake and implementation of water-saving techniques can save water and reduce domestic water use (Xia et al. 2007).

CONCLUSIONS

In this study, we analysed the effects of environmental policies and technologies to reduce river export of nutrients to Lake Dianchi in 2050. SSP3 (Shared Socio-economic Pathway 3) is used to reflect a worst-case situation. Four alternative scenarios are developed relative to SSP3. The first alternative scenario assumes the implementation of the current environmental policies in agriculture and sewage (CP). The second and third alternative scenarios assume the implementation of advanced technologies to treat wastewater (AT) and improved nutrient management in agriculture and mining (IAM), respectively. The fourth alternative scenario (OPT) combines AT and IAM.

River export of nutrients is projected to increase 1.4–4.4 times between 2012 and 2050 according to SSP3. In general, agriculture is projected to be an important source of DIN (synthetic fertilizers) and DOP (direct discharges of manure) in rivers of the middle and southern sub-basins. Mining is projected to contribute to DIP in the southern rivers. Sewage systems are projected to contribute to DIN, DON, DIP and DOP in rivers of the northern sub-basins in SSP3. Northern sub-basins (especially Daqing and Caohai) are more polluted because of the high urbanization and high population density. In contrast, middle and southern sub-basins suffer from intensive agricultural activities. Mining occurs within individual sub-basins in the south.

Current policies may reduce river export of nutrients to the lake by 2050 compared to SSP3 (the CP scenario). However, this may not be enough to avoid coastal eutrophication in the future. The AT and IAM scenarios project lower river export of nutrients than in SSP3, but reduction effects differ largely among nutrient forms. River export of nutrients is expected to be below the level of 2012 in an optimistic scenario (OPT). This is because this scenario combines advanced wastewater treatment (AT), improved nutrient management in agriculture and banned mining (IAM). However, even under these optimistic assumptions, eutrophication will be difficult to completely avoid. This is because the projected nutrient inputs in 2050 exceed their critical loadings. This exceedance is, however, much lower than for 2012. Our results indicate that more efforts are needed in reducing both agricultural and urbanization related nutrient pollution to avoid eutrophication of Lake Dianchi in the future. Circular-oriented management options such as recover and recycle nutrients from waste could help to avoid nutrient losses to rivers. Controlling internal nutrient loadings from lake sediments may also be important in combination with the other policies.

Insights of our study can support the formulation of management options to reduce future pollution of the lake. For example, improving treatment is important to reduce pollution from the northern sub-basins whereas improving nutrient management in agriculture may be important to reduce pollution from the middle and southern sub-basins. Our study can serve as an example for other lakes experiencing similar environmental problems.

SUPPLEMENTARY MATERIAL

The Supplementary Material for this paper is available at https://dx.doi.org/10.2166/bgs.2020.923.

REFERENCES

REFERENCES
Borin
M.
Bonaiti
G.
Giardini
L.
2001
Controlled drainage and wetlands to reduce agricultural pollution
.
Journal of Environmental Quality
30
(
4
),
1330
1340
.
Chen
Y. D.
2013
云南统计年鉴 (Yunnan Statistical Yearbook)
.
Yunnan Bureau of Statistics
,
Yunnan Province, China
. .
Chen
J.-n.
Zhang
T.-z.
Du
P.-f.
2002
Assessment of water pollution control strategies: a case study for the Dianchi Lake
.
Journal of Environmental Sciences
14
(
1
),
76
78
.
Cheng
W.
Shi
J.
Xia
Y.
Zhang
N.
2008
Farmland runoff of nitrogen and phosphorus in Dianchi watershed
.
Journal of Soil and Water Conservation
5
,
52
55
.
Coveney
M.
Stites
D.
Lowe
E.
Battoe
L.
Conrow
R.
2002
Nutrient removal from eutrophic lake water by wetland filtration
.
Ecological Engineering
19
(
2
),
141
159
.
Foged
H.
Flotats Ripoll
X.
Bonmatí Blasi
A.
Palatsi Civit
J.
Magrí Aloy
A.
Schelde
K. M.
2012
Inventory of Manure Processing Activities in Europe
.
Fujimori
S.
Hasegawa
T.
Masui
T.
Takahashi
K.
Herran
D. S.
Dai
H.
Hijioka
Y.
Kainuma
M.
2017
SSP3: AIM implementation of shared socioeconomic pathways
.
Global Environmental Change
42
,
268
283
.
Gao
W.
Howarth
R.
Hong
B.
Swaney
D.
Guo
H.
2014
Estimating net anthropogenic nitrogen inputs (NANI) in the Lake Dianchi basin of China
.
Biogeosciences
,
11,16(2014-08-28)
,
11
(
3
),
4577
4586
.
Gao
W.
Howarth
R. W.
Swaney
D. P.
Hong
B.
Guo
H. C.
2015
Enhanced N input to Lake Dianchi Basin from 1980 to 2010: drivers and consequences
.
Science of the Total Environment
505
,
376
384
.
Gu
L.-p.
Huang
B.
Zhao
S.-m.
Yang
X.-x.
Pan
X.-j.
2016
Twenty-eight polychlorinated biphenyls in surface sediments of Dianchi Lake and its Estuaries, China
.
Journal of Water and Environment Technology
14
(
3
),
115
124
.
Janse
J. H.
Domis
L. N. D. S.
Scheffer
M.
Lijklema
L.
Van Liere
L.
Klinge
M.
Mooij
W. M.
2008
Critical phosphorus loading of different types of shallow lakes and the consequences for management estimated with the ecosystem model PCLake
.
Limnologica-Ecology and Management of Inland Waters
38
(
3–4
),
203
219
.
Jassby
A. D.
Reuter
J. E.
Axler
R. P.
Goldman
C. R.
Hackley
S. H.
1994
Atmospheric deposition of nitrogen and phosphorus in the annual nutrient load of Lake Tahoe (California-Nevada)
.
Water Resources Research
30
(
7
),
2207
2216
.
Jie
Y.
Buisonjé
F.
Melse
R.
2017
Livestock Manure Treatment Technology of the Netherlands and Situation of China: White Paper
.
Jin
X.
2003
Experience and lessons learned brief for Lake Dianchi
.
Environmental Science
1
,
1
36
.
KGO
2017
关于滇池流域和西山重点保护区域采石采砂点关停和治理修复的通知 (Notice on Closure, Management and Restoration of Stone and Sand Quarries in Lake Dianchi Valley and Xishan Key Protected Areas)
.
Kunming Government Office
,
Kunming City, Yunnan Province
,
China
. .
KLDA
2013
云南省滇池保护条例 (Lake Dianchi Protection Regulations in Yunnan Province)
.
Kunming Lake Dianchi Authority
,
Kunming City, Yunnan Province
,
China
. .
Kong
W.
Rao
W.
Wang
C.
Peng
M.
Dong
L.
Yang
S.
Luo
T.
Li
Q.
2012
基于PSR 模型的滇池流域农村生活污水空间分布特征和控制研究 (Spatial distribution characteristics and control of rural domestic sewage in Lake Dianchi sub-basin based on PSR model)
.
农业环境科学学报
31
(
7
),
1393
1403
.
KPGOYP
2008
滇池流域 ‘五矿区’ 重点区域植被恢复指导意见 (Guiding Opinions on Vegetation Restoration in Key Areas of the ‘Five Mining Areas’ in Dianchi Watershed)
.
Kunming People's Government of Yunnan province
,
Kunming City, Yunnan Province
,
China
. .
Lewis
G. N.
Auer
M. T.
Xiang
X.
Penn
M. R.
2007
Modeling phosphorus flux in the sediments of Onondaga Lake: insights on the timing of lake response and recovery
.
Ecological Modelling
209
(
2–4
),
121
135
.
Li
X.
Janssen
A. B.
de Klein
J. J.
Kroeze
C.
Strokal
M.
Ma
L.
Zheng
Y.
2019
Modeling nutrients in Lake Dianchi (China) and its watershed
.
Agricultural Water Management
212
,
48
59
.
Liu
P.
2017
Simulation on Agricultural Non-Point Source Pollution Under Different Fertilization Scenarios in Lake Dianchi Basin Based on SWAT Model 基于SWAT 模型的滇池流域不同施肥情景下农业非点源污染模拟研究
.
Doctoral dissertation
,
Yunan Normal University
.
Liu
W.
Qiu
R.
2007
Water eutrophication in China and the combating strategies
.
Journal of Chemical Technology & Biotechnology: International Research in Process, Environmental & Clean Technology
82
(
9
),
781
786
.
Liu
Y.
Chen
J.
Mol
A. P.
2004
Evaluation of phosphorus flows in the Dianchi watershed, Southwest of China
.
Population and Environment
25
(
6
),
637
656
.
Liu
Y.
Liu
R.
Li
H.
2007
ICEAS 工艺脱磷除氮的影响因素 (Influence factors of phosphorus and nitrogen removal of ICEAS)
.
水处理技术
33
(
4
),
57
59
.
Liu
Z.
Liu
X.
He
B.
Nie
J.
Peng
J.
Zhao
L.
2009
Spatio-temporal change of water chemical elements in Lake Dianchi, China
.
Water and Environment Journal
23
(
3
),
235
244
.
Liu
Y.
Yang
P.
Sheng
H.
2012
Watershed pollution prevention planning and eutrophication control strategy for Lake Dianchi
.
Huanjing Kexue Xuebao
32
(
8
),
1962
1972
.
Ma
L.
Ma
W.
Velthof
G.
Wang
F.
Qin
W.
Zhang
F.
Oenema
O.
2010
Modeling nutrient flows in the food chain of China
.
Journal of Environmental Quality
39
(
4
),
1279
1289
.
Mayorga
E.
Seitzinger
S. P.
Harrison
J. A.
Dumont
E.
Beusen
A. H.
Bouwman
A.
Fekete
B. M.
Kroeze
C.
Van Drecht
G.
2010
Global nutrient export from WaterSheds 2 (NEWS 2): model development and implementation
.
Environmental Modelling & Software
25
(
7
),
837
853
.
MOA
2015
到 2020 年化肥使用量零增长行动方案 (Zero Growth in Synthetic Fertilizer Use After 2020)
.
Ministry of Agriculture of the People's Republic of China
,
Beijing
,
China
. .
Oller
I.
Malato
S.
Sánchez-Pérez
J.
2011
Combination of advanced oxidation processes and biological treatments for wastewater decontamination – a review
.
Science of the Total Environment
409
(
20
),
4141
4166
.
O'Neill
B. C.
Kriegler
E.
Riahi
K.
Ebi
K. L.
Hallegatte
S.
Carter
T. R.
Mathur
R.
van Vuuren
D. P.
2014
A new scenario framework for climate change research: the concept of shared socioeconomic pathways
.
Climatic Change
122
(
3
),
387
400
.
Parton
W. J.
Hartman
M.
Ojima
D.
Schimel
D.
1998
DAYCENT and its land surface submodel: description and testing
.
Global and Planetary Change
19
(
1–4
),
35
48
.
Pinckney
J. L.
Paerl
H. W.
Tester
P.
Richardson
T. L.
2001
The role of nutrient loading and eutrophication in estuarine ecology
.
Environmental Health Perspectives
109
(
suppl. 5
),
699
706
.
Qian
W. B.
Zhang
L.
Wang
S. R.
Cao
C. C.
Yan-Ping
L. I.
Cheng
J.
Yang
J. C.
Wen-Zhang
L. I.
2016
Compositional characteristics of sediment dissolved organic nitrogen in typical lakes and its relationship on water trophic status
.
Spectroscopy & Spectral Analysis
36
(
11
),
3608
3614
.
Ren
J.
Jia
H.
Jiao
L.
Wang
Y.
Yang
S.
Wu
Q.
Gao
Q.
Cui
Z.
Hao
Z.
2019
Characteristics of nitrogen and phosphorus formation in atmospheric deposition in Dianchi Lake and their contributions to lake loading
.
Huan jing ke xue=Huanjing kexue
40
(
2
),
582
589
.
Riahi
K.
Van Vuuren
D. P.
Kriegler
E.
Edmonds
J.
O'neill
B. C.
Fujimori
S.
Bauer
N.
Calvin
K.
Dellink
R.
Fricko
O.
2017
The shared socioeconomic pathways and their energy, land use, and greenhouse gas emissions implications: an overview
.
Global Environmental Change
42
,
153
168
.
Seitzinger
S.
Mayorga
E.
Bouwman
A.
Kroeze
C.
Beusen
A.
Billen
G.
Van Drecht
G.
Dumont
E.
Fekete
B.
Garnier
J.
2010
Global river nutrient export: a scenario analysis of past and future trends
.
Global Biogeochemical Cycles
24
(
4
),
1
16
.
Steinman
A.
Rediske
R.
Reddy
K. R.
2004
The reduction of internal phosphorus loading using alum in Spring Lake, Michigan
.
Journal of Environmental Quality
33
(
6
),
2040
2048
.
Strokal
M.
Kroeze
C.
Wang
M.
Bai
Z.
Ma
L.
2016
The MARINA model (Model to assess river inputs of nutrients to seAs): model description and results for China
.
Science of the Total Environment
562
,
869
888
.
Tang
C.
Duan
C.
Yu
C.
Song
Y.
Chai
L.
Xiao
R.
Wei
Z.
Min
X.
2017
Removal of nitrogen from wastewaters by anaerobic ammonium oxidation (ANAMMOX) using granules in upflow reactors
.
Environmental Chemistry Letters
15
(
2
),
311
328
.
Tong
Y.
Zhang
W.
Wang
X.
Couture
R.-M.
Larssen
T.
Zhao
Y.
Li
J.
Liang
H.
Liu
X.
Bu
X.
2017
Decline in Chinese lake phosphorus concentration accompanied by shift in sources since 2006
.
Nature Geoscience
10
(
7
),
507
.
Wang
R.
2002
滇池水体富营养化特征分析及控制对策探讨 (Analysis of the eutrophication characteristics and the control strategy of the Lake Dianchi)
.
地理科学进展
21
(
5
),
500
506
.
Wang
M.
Kroeze
C.
Strokal
M.
Ma
L.
2017
Reactive nitrogen losses from China's food system for the shared socioeconomic pathways (SSPs)
.
Science of the Total Environment
605
,
884
893
.
Xia
J.
Zhang
L.
Liu
C.
Yu
J.
2007
Towards better water security in North China
.
Water Resources Management
21
(
1
),
233
247
.
Xu
W.
Luo
X.
Pan
Y.
Zhang
L.
Tang
A.
Shen
J.
Zhang
Y.
Li
K.
Wu
Q.
Yang
D.
2015
Quantifying atmospheric nitrogen deposition through a nationwide monitoring network across China
.
Atmospheric Chemistry and Physics
15
(
21
),
12345
12360
.
Zhang
D.
Tang
L.
Chen
Y.
Zhu
Y.
Yang
Y.
Zhu
Z.
2007
滇池流域典型城郊村镇排放污水氮, 磷特征分析 (Characteristics of Nitrogen and Phosphorus in Typical Suburban Villages and Towns in Lake Dianchi sub-Basins)
.
Zhang
H. M.
Zhang
J.
Xiao
J. N.
2008
Study of denitrification phosphorus removal in A ∼ 2O-MBR
.
Journal of Dalian University of Technology
48
(
4
),
490
495
.
Zhang
T.
Zeng
W.
Wang
S.
Ni
Z.
2014
Temporal and spatial changes of water quality and management strategies of Dianchi Lake in southwest China
.
Hydrology and Earth System Sciences
18
(
4
),
1493
1502
.
Zhou
J.
He
D.
Xie
Y.
Liu
Y.
Yang
Y.
Sheng
H.
Guo
H.
Zhao
L.
Zou
R.
2015
Integrated SWAT model and statistical downscaling for estimating streamflow response to climate change in the Lake Dianchi watershed, China
.
Stochastic Environmental Research and Risk Assessment
29
(
4
),
1193
1210
.

Supplementary data