We thank Wang et al.(1) for their comments on our article. Wang et al.(1) convincingly show in their recent publication that aquaculture is another important source of nutrients in rivers locally, and that this source has been neglected by several studies, including ours. When we developed our model, their publication was not available yet. We agree that aquaculture can be an important source of nutrients in Chinese rivers at the local scale. However, for China as a whole, aquaculture is a relatively small source of nutrients in aquatic systems. Indeed, Wang et al.(1) conclude that nutrient release from freshwater aquaculture into surface water made up 6% of total nitrogen (TN) and 7% of total phosphorus (TP) delivery obtained from the IMAGE-GNM model(2) for the year 2010. We added their estimates of nutrient release from aquaculture to our total dissolved nitrogen (TDN) and total dissolved phosphorus (TDP) model simulations on the national scale.(3) We found that aquaculture contributes by 5% to N and 0.5% to P in rivers. At the local scale, aquaculture can be an important source of nutrients. However, we believe that aquaculture is a dominant source in only a relatively small percentage of the more than 2000 counties represented in our model. Figure 1 shows our modeled inputs of total dissolved nitrogen (TDN) (Figure 1A) and total dissolved phosphorus (TDP) (Figure 1B) to rivers at a grid of 0.5° × 0.5° from our study.(3) We may compare this to the gridded total nitrogen and total phosphorus release from aquaculture (see Figure 2 in Wang et al.(4)). This comparison suggests that our calculated levels of total dissolved N and P inputs to rivers are much higher (generally 10 times higher) than the levels of the total N and P release by aquaculture. Moreover, many hotspots of aquaculture seem to be overlapping with hotspots of industrialized livestock production.(3) This would reduce the relative share of aquaculture in the total pollution in these regions. On the other hand, we also see downstream areas in the Pearl basin, and middle- and downstream areas in the Yangtze basin, where aquaculture could indeed be a dominant source which cannot be ignored (Figure 1). Figure 1. Total dissolved nitrogen (TDN) (A) and total dissolved phosphorus (TDP) (B) inputs to rivers (Gg year–1) in China on 0.5° × 0.5°grid scale for the year 2012 calculated from Chen et al.(3) We consider our model system as a powerful tool for decision support on nutrient management in China. Our system includes the most relevant nutrient sources, and our study highlights the important contribution of direct manure discharges, which can be considered as another important source of nutrients inputs, which has been ignored by several studies. It goes without saying that if our model is used by county mayors for pollution management, the model first needs to be validated for the local conditions. A calibration against local data might be required, depending on the validation results. When relevant, our multiscale nutrient model should integrate more detailed local inputs, such as the nutrient inputs of aquaculture on the county scale and the actual locations of aquaculture farms as point sources. To conclude, it seems that another missing source of nutrients in Chinese surface water systems has been found. Like the direct discharge of manure,(5) we agree with Wang et al.(1) that aquaculture can indeed be considered as another point source of animal waste of nutrient pollution in rivers that should be accounted, in particular for local studies. We invite Wang et al.(1) to collaborate on a next generation of nutrient models, that account for all sources of animal waste in rivers (pigs, cows, chicken, and fish) to improve the decision making for effective pollution management strategies in China. The authors declare no competing financial interest. We gratefully acknowledge the sponsors of this research: Wageningen Institute for Environment and Climate Research (WIMEK) of Wageningen University & Research, National Key Research and Development Program of China (2016YFD0800106), the Hundred Talent Program of the Chinese Academy of Science, and the Distinguished Young Scientists Project of Natural Science Foundation of Hebei (D2017503023). This article references 5 other publications.

Figure 1. Process for EDA of unknown DBPs in drinking water. The authors declare no competing financial interest. Dr. Huiyu Dong was financially supported by the National Natural Science Foundation of China (51525806, 51878648), Chinese Academy of Sciences (QYZDY-SSW-DQC004), and Chinese Scholarship Council (201804910238). This article references 5 other publications.

It has recently been drawn to our attention that the original Figure 1G of this paper was incorrect. Here, we provide the correct Figure 1. This correction does not affect the results and conclusion of this paper. Figure 1. Corrected Figure This article has not yet been cited by other publications. Figure 1. Corrected Figure

In our manuscript, we reported that the pyrethroid metabolite measurements were specific-gravity adjusted. However, we had instead used unadjusted values for the statistical analyses. To examine the effects of the adjustment, we have now performed the analyses adjusting the pyrethroid values by specific gravity using the following formula: Many of the associations with pyrethroid metabolites remain the same when using specific-gravity adjusted measurements as with unadjusted measurements, whereas others diminished. Age was still associated with cis-DCCA, trans-DCCA, and 3-PBA. Working during pregnancy remained associated with cis-DCCA and 3-PBA. The association with high fat intake and cis-DBCA, cis-DCCA, and 3-PBA remained significant, but the association with trans-DCCA became only marginal (p < 0.1). Postdelivery BMI remained associated with cis-DCCA and 3-PBA, but was no longer associated with cis-DBCA and trans-DCCA. The application of IRS in the home during pregnancy was now only marginally associated with cis-DCCA (p < 0.1) and no longer associated with 3-PBA. The application of pesticides to food stocks was no longer associated with urinary concentrations of cis-DCCA, trans-DCCA, or 3-PBA. The presence of pesticide containers on the homestead was no longer associated with cis-DBCA, cis-DCCA, or trans-DCCA, although it remained associated with 3-PBA. The storage of pesticides in the yard remained associated with trans-DCCA and marginally associated with cis-DCCA and 3-PBA, but was no longer associated with cis-DBCA. While the household’s primary drinking water source was marginally associated with cis-DCCA, trans-DCCA, and 3-PBA in the unadjusted analyses, it was no longer associated when using specific-gravity adjusted concentrations. When using Bayesian Model Averaging, the most notable predictors remained the same, although the strength of the associations became weaker in most cases. This article has not yet been cited by other publications.

Figure 1. U.S. uranium imports, domestic uranium production, potential and historic uranium recovery from phosphates, and weighted-average uranium prices as well as projected minimum and maximum costs for uranium recovery using solvent extraction (a) and ion exchange (b) in the United States (Source: EIA(8) and Beltrami et al.(1)). The authors declare no competing financial interest. This article references 10 other publications.