Contribution of preferential flow to tile drainage varies spatially and temporally

Abstract

Tile drainage of agricultural fields is a conduit for nutrient losses. Preferential flow in the soil can more directly connect surface runoff with tile drainage compared with matrix flow, which may also increase P losses. In this study, water temperature was monitored in surface runoff and tile drainage and electrical conductivity (EC) was measured in tile drainage at two sites in southern Ontario with different soil types (i.e., clay and loam). These data were used to estimate the percentage of preferential flow in tile drainage based on end member mixing. Estimates using temperature were compared with estimates using EC, and both were evaluated across seasons and hydrographs and compared with P concentration and load data. There was strong correlation (r = .83) between estimates of preferential flow using the two methods, but due to variability in surface temperatures, EC provided a less flashy estimate for preferential flow (Durbin-Watson statistics of 0.34 for temperature and 0.09 for EC). Preferential flow accounted for a higher percentage of tile drainage flow in clay soil than loam, but percentages were not significantly different between seasons or timing within events. Phosphorus concentrations and loads were weakly correlated with preferential flow, suggesting that P transport was influenced by other factors as well. Although further work is necessary to calibrate these methods for estimating preferential flow from continuously monitored temperature and EC, this technique can be applied to already collected data to model and test posited explanations of observed phenomena in P, other nutrients, and water transport from tile-drained agricultural land.

Testing the ‘two water worlds’ hypothesis under variable preferential flow conditions

“…More cumulative preferential flow generally corresponded to greater f PF values detected in the first 24 h of drainage of high drainage rates (Figure S4), yet preferential flow often persisted through the inter‐event drainage period. This result suggests that the preferential flow detected throughout each event had an inherent travel time distribution that was selective to rapid transport (e.g., more event water appearing in drainage in first 24 h) but nonetheless incorporated a relevant proportion of water that required multiple days to drain (Pluer et al, 2020).…”

One size does not fit all: Toward regional conservation practice guidance to reduce phosphorus loss risk in the Lake Erie watershed

“…The increased P losses in tile drainage that have been observed in the southwestern management region have not been observed in the loamier, undulating soils in the northeastern management region (Lam et al., 2016b). Although there is preferential flow in loam soils in this region (Pluer et al., 2020), there is a greater degree of interaction between macropores and soils (Grant, Macrae, & Ali, 2019) and more buffering of P by subsoils (Plach, Macrae, Ali, et al., 2018); consequently, the contribution of tile drains to edge‐of‐field P losses is small relative to their hydrologic contribution (Plach et al., 2019; Van Esbroeck et al., 2016). The fact that tiles are spaced more widely and are installed deeper in loams than in clay may also be a factor.…”

“…This primarily occurs on saturated soils (Macrae et al., 2019) rather than as infiltration‐excess runoff over frozen soils, which is more common in regions with more severe winters (Dumanski et al., 2015; Gray et al., 2001). Although tiles flow at their maximum capacity during these events (Macrae et al., 2007b), P losses through tiles are often less than those in the surface runoff that occurs simultaneously (Van Esbroeck et al., 2016; Plach et al., 2019) due to strong buffering potential in the subsurface (Plach, Macrae, Williams, et al., 2018; Plach, Macrae, Ali, et al., 2018) and less preferential connectivity than clay soils (e.g., Grant, Macrae, Rezanezhad, et al., 2019; Macrae et al., 2019; Pluer et al., 2020). Moreover, most runoff occurs before spring fertilizers have been applied.…”

Quantifying hydrologic pathway and source connectivity dynamics in tile drainage: Implications for phosphorus concentrations

“…Previous studies have either used total Q, preferential flow, or new/old water estimates to predict P concentrations and loading in tiles. For instance, Pluer et al (2020) found that preferential flow (estimated by conductivity based unmixing) was weakly correlated with P concentration, although the relationship between P and preferential flow was positive suggesting that preferential flow was a significant driver of P transport to tiles (Grant et al, 2019;Pluer et al, 2020). Given the relatively low cost of specific conductance, flow, and temperature sensors, widespread application of pathway-connectivity frameworks across environmental and management gradients has significant potential for advancing our understanding of contaminant transport in tiledrainage.…”