Abstract
Agricultural drainage networks increase hydrological connectivity from the field to the receiving environments. The response to the issue of surface water quality therefore implies an understanding of the hydrological processes related to drainage, particularly at the field scale. Drainage by tile drains and drainage ditch are the two most studied types at the plot scale. They can be complemented by temporary surface drains to improve the removal of surface runoff. The hydrological processes and functioning of tile-drained fields have been extensively studied at the event scale. However, few studies have been conducted over a full hydrological year and the description of water pathways in the soil generally relies on either exogenous tracer monitoring or irrigation experiments. In addition, only a few studies have been conducted on fields combining tile drainage and temporary surface drainage. In this study, high temporal resolution quantification of runoff from surface and subsurface drainage was conducted for a full year to establish one of the first water balances for a surface and subsurface drained field. Soil water pathways were studied under dry and saturated soil conditions tracing water by measuring stable isotope concentrations (18O and 2H) on rainwater, soil water, and surface and subsurface runoff. Runoff quantifications showed that surface drainage and subsurface drainage respectively evacuate 41% and 32% of the annual cumulated effective rainfall. The water balance highlights the importance of infiltrations to the deep horizons: 46% of the water transferred to the soil is not captured by the subsurface drains. Water tracing showed that rainwater was directly transferred to subsurface drains on dry soil, likely through macropores. On saturated soil, soil water present before the rain remains the main source of water to the subsurface drains, but event-rainwater also reaches the subsurface drains and can constitute up to 25% of the subsurface runoff volume.
Key words: tile drain, surface drainage, water balance, water tracing, soil water pathways, percolation.
1. INTRODUCTION
Subsurface drainage constitutes an artificial water pathway that directly links subsoil to surface waters. Since the 1950s, production-oriented agricultural policies have led to draining hydromorphic soils to increase the area of cultivated land and its yield (Musgrave, 1994). According to the last data of the International Commission on Irrigation and Drainage (2018), 11% of the arable lands of the world are drained. Two types of drainage are commonly used: drainage by digging ditches at the edge of the field and subsurface drainage by laying buried pipes.
At the watershed scale, one of the first drainage impact studied is the impact on flooding (Skaggs et al., 1994). In particular, drainage by tile drains reduces the intensity of floods that have a low return period, typically, less than 2 years (Nedelec, 2005). This is due to an increased transit time of water percolating through the soil to the tile drains in comparison to a direct transfer of water on the soil surface. Studies have also focused on the ecological and biological impact of drainage systems on receiving environments (Blann et al., 2009; Gilliam & Skaggs, 1986). Reducing subsurface drainage impact at the watershed scale implies to understand drainage impact at the field scale.
At the field scale, in temperate climates, the functioning of tile drains is generally highly seasonal (Arlot, 1999; Gramlich et al., 2018; Hirt et al., 2011; K.W. King et al., 2014). The decrease in evapotranspiration during the fall leads to the formation of a saturated zone in the soil. The saturated zone, or perched water table, is usually present from fall to spring. Zimmer, (1988) defines this period as a period of intense drainage during which runoff coefficients are at their maximum. During the rest of the year, the drains only flow during intense storm events. In addition, subsurface drainage has been shown to reduce surface runoff most of the time. For example, Arlot (1999),comparing drained and no-drained fields with Cambisol on altered shale, showed subsurface drainage represents 90% of the total runoff and decreases by a factor 10 to 20 the surface runoff. Grazhdani et al., (1996), in fields with clay loam, showed subsurface drainage increases the water yield by 34% but reduces surface runoff by 40%. In this case however, subsurface runoff contribution to the total runoff varied from 47% to 69%. The impact of drainage on surface runoff depends on the type of soil but also on the design of the drainage system. In lowland agricultural regions such as the Région Centre-Val-de-Loire in France, subsurface drainage is complemented by a third type of drainage: surface drainage by digging surface drainage rills (SDR). Contrary to tile drainage and ditches, SDRs are temporary: they are installed after seeding and destroyed by the first tillage following the harvest. This type of drainage is used to improve the evacuation of excess surface water and to direct surface runoff. It provides an additional water transfer pathway. Drainage by SDR is therefore likely to modify the hydrological functioning of the field. However, to our knowledge, no work have been carried out on fields or watersheds with this type of drainage. Therefore, no water balance has been established for a surface and subsurface drained field.
Studies at the scale of the soil profile in drained contexts have shown that macropores are likely to be an important pathway for dissolved (King et al., 2015) and solid transfers (Michaud et al., 2019; Øygarden et al., 1997). Understanding the pathways of water flow in drained soils is essential to reduce the negative impacts of drainage. Water flow in soils can be categorized in two types of flow: one through the soil matrix (Skopp, 1981) and the other through macropores (Jarvis, 2007; McDonnell, 1990; Richard & Steenhuis, 1988). Matrix flow is generally slower than macropore flow, which is qualified as preferential flow. Macropores can have biological or structural origin and are distinguished from the rest of the porosity by the heterogeneity of their distribution (vertical and lateral), their large diameter and their strong connectivity. Geochemical tracing experiments and observation of water flow paths through the use of brilliant blue has highlighted the role of macropores in the hydrological functioning of drains (Stamm et al., 2002). This study showed that macropores enhance the connectivity between the surface and the drains. Using bromide tracing, Everts & Kanwar (1990) measured, during two irrigation experiments, that 29% and 20% of the total volume flowing out of the drain was from a preferential flow. Stone & Wilson (2006), using chloride tracing, measured a preferential flow contribution of 11% and 51% during two rainfall events. After two years of isotopic monitoring, Leaney et al. (1993) estimated that the share of preferential flow at the drain outlet was at least more than 80%. These studies underline the difficulty that persists in predicting the share of preferential flow reaching the drain during a rainfall event. A few studies have therefore looked for factors influencing the functioning of macropores. For example, Grant et al. (2019), by observing with brilliant blue the water pathways taken in two types of soils, have highlighted the influence of soil type on the flow of water through macropores: the preferential flow in a clayey soil is greater than in a sandy-silt soil. In addition to the soil properties, the functioning of macropores seems to be linked to the hydric state of the soil. Following blue-glow tracing experiments combined with high temporal resolution soil moisture measurements, Weiler & Naef (2003) showed that water circulation in macropores was dependent on the water content of the different horizons. The authors explained that the reason for this difference is due to exchanges between matrix and macropores. Moreover, they showed that the circulation of water in the macropores could either start at the level of a saturated horizon or from the soil surface when the rainfall intensity exceeded the infiltration capacity of the soil. Smith & Capel (2018), by monitoring the specific conductance of water at the drain outlet, have shown that even a light rainfall (< 5 mm) can lead to a preferential flow of rainwater through the macropores.
However, most of the studies concerning water flow in a drained context have focused on a few rainfall events without accounting for the seasonal variability of drain operation. In order to improve agricultural practices to reduce dissolved and solid exports from drains, it is preferable to determine the type of water flow in drained soils throughout the year, particularly with regard to variations in the hydric state of the soil.
Studies dealing with the different aspects of the hydrological functioning of subsurface drains reveal that:
In order to answer the questions raised by these gaps, the objectives of this study are:
To meet these objectives, we propose to quantify the surface and subsurface drainage flows of an agricultural field by measuring at high temporal resolution the flow rates at the outlet of drains over a whole hydrological year. In parallel, we also propose to monitor the soil hydric state by tensiometric and piezometric measurements with high temporal resolution. Finally, the study of water pathways in the soil according to the hydric state of the soil will be addressed through isotopic tracing of water during a winter runoff event and a spring runoff event.