Stream Water Quality Dynamics Across Multiple Temporal Scales in Two Adjacent Watersheds in North Central Texas

Understanding stream water-quality dynamics across multiple temporal scales is essential for characterizing their relationships with environmental conditions. This study investigated dissolved oxygen (DO), specific conductance (SC), and pH in two neighboring North Central Texas watersheds, Little Elm Creek (LEC) and Doe Branch (DB), which differ substantially in urban development. Using 15 min observations of DO, SC, pH, water temperature (WT), and streamflow (Q) from 2010 to 2025, together with hourly ERA5-Land 2 m air temperature (AT) and precipitation (P) data and annual land-use/land-cover (LULC) data, this study examined interannual, seasonal, daily, diurnal, and storm-event water-quality variations. At the interannual scale, WT and DO increased significantly in LEC (0.15 °C yr−1 and 0.09 mg L−1 yr−1, respectively), whereas SC increased significantly in DB (10.75 µS cm−1 yr−1), despite generally lower SC in DB than in LEC. Although annual mean SC was correlated with LULC percentages in both watersheds, these correlations weakened substantially in DB after removing common linear temporal trends, indicating that the original relationships were strongly influenced by shared temporal changes. At monthly and daily scales, DO was strongly negatively correlated with WT, with monthly correlation coefficients of −0.883 in LEC and −0.954 in DB. Diurnal DO fluctuations were greater in LEC (median fitted amplitude = 0.82 mg L−1) than in DB (0.26 mg L−1). Elevated pH (>8.5) occurred in 4.34% of observations in LEC compared with 0.33% in DB. During inter-storm periods, SC tended to increase in LEC (median slope = 0.34 µS cm−1 h−1) but decrease slightly in DB (−0.07 µS cm−1 h−1). During storm events, rising-limb SC decreases were more frequent in LEC (40.3%), whereas increases were more frequent in DB (41.3%). These findings reveal contrasting water-quality dynamics in neighboring watersheds under similar regional climatic conditions and highlight the importance of integrating long-term and high-frequency observations while considering watershed-specific hydrological conditions and external water inputs.