Distinguished Prof. Jih-Hsing Chang
Department of Environmental Engineering and Management, Chaoyang University of TechnologySpeech Title: Cross-Seasonal Multivariate Analysis of Heavy Metal Partitioning and Nitrogen/Organic Pollution Control Mechanisms in Urban Waterways
Abstract: Urban waterways in New Taipei City receive combined loadings of domestic sewage, industrial wastewater, and surface runoff, leading to the accumulation of heavy metals (Cu, Zn) in sediment and elevated water-column concentrations of ammonia nitrogen (NH3-N) and chemical oxygen demand (COD). This study integrates three sampling campaigns (April 29, May 29, and June 17, 2026) across five sites (Longan Road, Tanti Ditch, Yakoukeng Creek, Zhongzheng Road, and Lujiao Creek; n = 9 per campaign), applying three complementary statistical approaches: multiple linear regression (MLR: C_w = β_0 +β_1 ·C_s +β_2 ·fine fraction + β_3·organic matter +β_4 ·pH + β_5·ORP), partition coefficient (Kd = Cs/Cw) analysis, and Pearson correlation. MLR results show that sediment total Cu remained a stable, dominant predictor of dissolved Cu across all three campaigns (R² ≈ 0.99), whereas the Zn model's explanatory power declined progressively (R² = 0.981→0.957→0.889) and lost overall significance by June. NH3-N and COD models exhibited an alternating source-term (sediment TN) - redox (ORP) - source-term (TN) control pattern, reaching their best fits in June (R² = 0.9987 and 0.9946, respectively). Partition coefficients confirmed Cu (log Kd 4.23-4.32) and Zn (log Kd 3.98-4.16) as consistently strongly sediment-bound, while NH3-N log Kd rose from 2.86 to 3.07, largely driven by a sharp NH3-N decline at the wetland site Lujiao Creek (11.0/11.2 → 1.64 mg/L). Site-level heterogeneity was notable: Cu Kd at Zhongzheng Road rose continuously (42,387→97,412 L/kg), while at Lujiao Creek it fell sharply after April. Pearson analysis consistently identified three co-source structures across campaigns: Cu-Zn (industrial metal co-source), organic matter-fine fraction (sediment physicochemical coupling), and NH3-N-sediment Cu (persistent cross-correlation), though the COD-NH3-N linkage was significant only in April and June. These findings indicate that heavy-metal partitioning is temporally stable, favoring sediment-focused remediation (e.g., dredging, reductive settling) as a management priority, whereas NH3-N/COD control requires combined point-source reduction and maintenance of oxidizing conditions (ORP > 0 mV).