将：3结果与讨论。31不同用水行为下硫稳定同位素变化规律分析。311硫稳定同位素变化规律分析。本轮监测分别以1-24 h第1天、25-48 h第2天、49-72 h第3天连续三个监测区间对不同用水行为下其相应污水的34S变化规律进行研究监测数据对比详见图2。根据统计结果不同用水行为下的污水中34S在连续三个监测区间内同样表现出了较为一致的变化规律：餐厨行为会造成34S的上升淋浴行为、舆洗室行

Results and discussion. 3.1 Analysis of sulfur stable isotope variation patterns under different water usage behaviors. 3.1.1 Analysis of sulfur stable isotope variation patterns. In this round of monitoring, the δ34S variation patterns of corresponding sewage under different water usage behaviors were studied in three consecutive monitoring intervals of 1-24 h (Day 1), 25-48 h (Day 2), and 49-72 h (Day 3). The comparison of monitoring data is shown in Figure 2. According to the statistical results, the δ34S in sewage under different water usage behaviors showed a consistent variation pattern in the three consecutive monitoring intervals: kitchen behavior caused an increase in δ34S, while showering and laundry behavior caused a decrease in δ34S. Tooth brushing and dishwashing behavior caused δ34S to fluctuate around the background level of tap water. Overall, the δ34S in sewage under tooth brushing behavior fluctuated upward, while the δ34S in sewage under dishwashing behavior fluctuated downward. The δ34S variation patterns under different water usage behaviors and monitoring intervals are shown in Figure 3. Based on the analysis of the upward magnitude of δ34S under different water usage behaviors, kitchen behavior caused the largest increase in δ34S, with a 72-h change ratio mean of 64.69% compared to tap water. Tooth brushing behavior caused the second-largest increase in δ34S, with a 72-h change ratio mean of 5.62%. Laundry behavior caused the largest decrease in δ34S, with a 3-day decrease ratio mean of -14.37% compared to tap water. Showering and dishwashing behavior caused a 3-day decrease ratio mean of -7.17% and -7.69%, respectively, compared to tap water. Based on the analysis of the dispersion degree of δ34S under different water usage behaviors, the sewage under showering behavior showed excellent stability, with a 72-h coefficient of variation of 0.036. The sewage under tooth brushing, laundry, and dishwashing behavior also showed good stability, with 72-h coefficients of variation of 0.074, 0.101, and 0.149, respectively. The sewage under kitchen behavior showed a moderate level of stability, with a 72-h coefficient of variation of 0.258. In addition, tap water showed excellent stability, with a 72-h coefficient of variation of 0.012. 3.1.2 Uncertainty evaluation. Based on the uncertainty evaluation formula for water quality indicators under different water usage behaviors, the uncertainty results of δ34S under different water usage behaviors under the experimental conditions of this chapter are shown in Table 4. The statistical results show that the uncertainty range of sulfur isotope indicators under different water usage behaviors is 0.28‰-2.26‰. Among them, the uncertainty of δ34S indicators in sewage under showering behavior is the smallest, at 0.28‰, while the uncertainty of δ34S indicators in sewage under kitchen behavior is relatively the largest, at 2.26‰. 3.2 Analysis of sulfur stable isotope variation patterns at the end of the drainage area. 3.2.1 Analysis of sulfur stable isotope variation patterns. Based on the 72-h hourly monitoring data of δ34S in sewage at the end of the drainage area, it was divided into three consecutive monitoring intervals of 1-24 h (Day 1), 25-48 h (Day 2), and 49-72 h (Day 3) for statistical analysis. The comparison of δ34S monitoring data and the distribution of relative deviation ratios by hour are shown in Figure 4. The results show that the δ34S at the end of the drainage area was generally higher than the δ34S baseline of tap water and locally decreased, and the indicator fluctuated uniformly with time in the three consecutive monitoring periods. Compared with tap water, the growth rate of δ34S in the three consecutive monitoring periods tended to be conservative, with mean increase ratios of 30.58%, 24.85%, and 22.54%, and a 72-h change ratio mean of 25.99%. In terms of dispersion degree, it showed a moderate to good level of stability in the three consecutive monitoring periods, with corresponding coefficients of variation of 0.236, 0.145, and 0.138, and a 72-h coefficient of variation of 0.186. The distribution of relative deviation ratios by hour reflects that the δ34S in sewage at the end of the drainage area showed a mild hourly fluctuation in the three consecutive monitoring periods. The statistical results show that the relative deviation ratio by hour fell within the main interval of -13.62% to 15.51%, -8.51% to 11.96%, and -18.06% to 14.87% (sample size>80%) in the three consecutive monitoring periods. To further consider the reproducibility of the δ34S fluctuation pattern, the two-way correlation of δ34S under different statistical periods (taking 3, 4, ..., 23 h) in the three consecutive monitoring periods was calculated, and the distribution is shown in Figure 5. The results show that the 24 h fluctuation pattern of δ34S has good reproducibility. The two-way correlation levels for the 23 h statistical periods in the three monitoring periods were 0.758 (Day 1 and Day 2), 0.721 (Day 2 and Day 3), and 0.743 (Day 1 and Day 3), with a mean level of 0.741. It showed a correlation level greater than 0.850 (some greater than 0.900) in the time period of 18:00-23:00, followed by the time period of 0:00-12:00, with a correlation level greater than 0.800 (some greater than 0.900). These two dominant correlation periods accounted for 73.91% of the total statistical period. 3.2.2 Analysis of the correlation between kitchen behavior and water quality at the end of the drainage area