Mitigating nitrogen pollution from global croplands with cost-effective measures


 Cropland is one of the major sources of global nitrogen pollution1, 2. Mitigating nitrogen pollution from global croplands is a grand challenge because of the nature of non-point source pollution from millions of farms and the lack of financial resources and scientific knowledge of farmers3. Here we synthesize 683 studies worldwide and identify 11 key measures which can reduce 30-70% of nitrogen pollution while increasing crop yield and nitrogen use efficiency by 10-30% and 20-60%, respectively. Adoption of these measures would produce 14 million tonnes (Tg) more crop nitrogen with 28 Tg less nitrogen fertilizer and 27 Tg less nitrogen pollution to the environment in global croplands in 2015. However, to achieve these potentials, innovative policies such as a nitrogen credit system (NCS) should be implemented to incentivize and subsidize the adoption of these measures given the mismatch between benefits for the whole society while the abatement cost only for farmers. Full implementation of the best-fitted measures could achieve 306 billion USD benefits on ecosystem, human health and climate globally, with net mitigation costs of only 21 billion USD given 35 billion USD fertilizer saving cost has offset 2/3 of the total mitigation cost. The large benefit-to-cost ratio suggests the feasibility and urgency to implement the NCS and Tier approaches could help to implement the most cost-effective measures on regional and local scales.

global food production 4 . To increase food supply, intensified agriculture has used more 51 and more nitrogen fertilizers and manure 2 . Nevertheless, over half of these nitrogen 52 inputs is lost to the environment, close to the annual total chemical nitrogen fertilizer 53 used globally, leading to severe air (e.g. fine particle matters, PM2.5) and water (e.g. 54 eutrophication) pollution, soil acidification, climate change, ozone depletion and 55 biodiversity loss 1,5 . Therefore, reducing nitrogen loss from croplands can not only 56 increase direct economic returns from fertilizer saving but also improve human health 57 and other welfare linked to ecosystem health and climate change 6 .

59
Mitigation of nitrogen pollution from croplands has attracted global attention 7 . Best 60 management practices have been developed such as 4R nutrient stewardship (right 61 fertilizer type, right amount, right placement, and right time) and soil testing to 62 precisely apply fertilizer to the soil 8 . However, these advanced management and 63 technologies are seldom fully implemented, a high heterogeneity of best-practices on 64 the local scale, and high implementation cost for farmers 2, 9 . Thus, we attempted to 65 solve these challenges through identifying the best measures to abate nitrogen pollution 66 and estimate the cost and benefit to facilitate the implementation of these measures. We Mitigation potential 74 Through conducting a meta-analysis of 683 papers published in the past two decades, 75 we identified 11 key measures that can mitigate nitrogen pollution from croplands 76 across global regions (Fig. 1, Fig. S1-S11). These measures can be divided into four  Reduction of Nr input and losses varied globally (Fig. 2) due to the optimization of nitrogen use (Fig. 2). While the NUE would be only slightly 116 increased in other global regions since both high-income countries such as US and low-117 income countries from Africa already have high NUE 13 . Detailed changes of nitrogen 118 budget on national and regional scales can be found in Extend Data Fig. 1-3.

120
Mitigation cost and benefit 121 The net mitigation cost of these measures mainly including the labor cost, material cost 122 and services is estimated at 21 billion USD in 2015 globally (Fig. 3a). These costs 123 represent the net benefits from fertilizer savings that are estimated at 35 billion USD, 124 which means the initial implementation cost is around 56 billion USD. Fertilizer saving 125 contribute to 2/3 of the total implementation cost of these measures. China alone would 126 require 5 billion USD to implement the mitigation measures, followed by India which 127 needs 3 billion USD; these two countries are the largest consumers of Nr fertilizers and 128 emitters of Nr to the environment (Fig. 4). The net mitigation cost of other countries is 129 less than 1 billion USD, mainly due to small amount of Nr loss and/or more advanced 130 agricultural machinery and well-trained farmers, allowing a low transaction cost to 131 implement these measures. These Nr abatement measures such as the 4R nutrient 132 stewardship would be implemented by farmers who would have to change management 133 practices on their lands 3 . Farmers must invest money to achieve these changes if there is 134 no government or public support. However, in competitive markets that do not 135 internalize the pollution damage costs, implementation of these mitigation measures, 136 which reduce farmers' profits, is unlikely 3 .

138
Despite large mitigation costs for farmers, the overall benefits on human health, 139 ecosystem, and climate change due to abatement of Nr losses from croplands are 140 approximate 15 times that of the cost, i.e., 306 billion USD for the whole society ( Fig.   141 3a, Table S2). About 43% of the benefit (132 billion USD) stems from improved human 142 health, most importantly through avoiding respiratory diseases due to PM2.5 pollution.

143
The remaining 55%, about 170 billion USD, stem mainly from reduced damages to 144 ecosystem services such as eutrophication, which reduces the recreation value. Climate

170
The NCS can not only be applied to Nr reduction but also to broader pollution control, 171 including other agricultural pollution such as phosphorus, potassium, heavy metals, and 172 other sectors such as livestock production 16 . A more comprehensive credit system such 173 as Environmental Credit System (ECS) to ensure environmentally-friendly food 174 production is critical for agricultural sustainability. The credit system is an initiative for 175 the solution to the tragedy of common-pool resource 17 . Environment is a typical 176 common-pool resource, while food supply is essential for human beings. We may have 177 to subsidize farmers to produce food without compromising environmental quality 178 given the essential of food production for human and it is costly for farmers to control 179 pollution. Thus, the polluter-pays-principle (i.e., pollution tax rather than production 180 subsidy) applied to reduce industrial pollution 18 , may be not suitable for agricultural 181 pollution control. Therefore, ensuring the synergy of food production and environmental 182 protection via the credit system is one of the key strategies for achieving the global 183 sustainable goals 4 .

185
Feasibility of implementation 186 The implementation barriers and costs of mitigation measures vary strongly. We 187 therefore grouped the mitigation measures into four Tiers with increasing 188 implementation challenges from lower to higher Tiers (Table 1, Table S4). The lower 189 the tiers, the easier and cheaper for farmers to adopt. Use of enhanced efficiency 190 fertilizers (EEFs) is a typical Tier 1 measure, government can subsidize EEFs to make 191 them the same or lower price compared to traditional fertilizers, and farmers would 192 prefer using these new EEFs given they can earn more with the same or lower input 19 .   Cropland NUE is defined here as the ratio of harvested crop N to total cropland N input.  The reduction potential of N fertilizer use 303 The estimation of global and regional fertilizer reduction potential is based on the 304 difference between the current and optimal fertilizer use with best cropland N 305 management practices.   (Table S1) (Table S2).

377
The climate impact of Nr management is assessed as showed in Eq. (11): where , represents the unit damage cost to climate in US $ per kg N (Table S2).    Table S3 and Fig. S1-S11.

Scenario Description BAU
The world develops into one that is unequal and fragmented. Lifestyles are materialistic and consumption-oriented, resulting in food waste and livestock consumption which is relatively high. There is little investment into farming practices and technology, leading to no improvements in agricultural production efficiency. Trade is low, and there is no efficient regulation of land-use change.

Tier 1
The three options are preferentially adopted by farmers regarding cost feasibility and efficiency.

Tiers 1+2
Based on Tier 1, 4R stewardships with efficient technologies in fertilizer application are further implemented to varying degrees by nation due to a high heterogeneity of best-practices on the local scale, and high implementation cost for farmers.

Tiers 1+2+3
Based on Tier 1+2 regulations, Tier 3 regulations will further implement regional specific tillage, irrigation, and new cultivar options leading to improvements in agricultural production efficiency. More advanced knowledge and facilities are required, which usually can only be adopted by professional farmers with large-scale farming.

Tiers 1+2+3+4
Based on Tier 1+2+3 regulations, Tier 4 requires more advanced knowledge and financial support to ensure the sustainability and resilience of the cropland system through such as buffer zone. BAU, business as usual. 4R refers to the right fertilizer type, right amount, right 573 placement, and right time on fertilizer application.