CONTAMINATION OF TILE-DRAINAGE Water DUE TO Outwintering OF Beef and Dairy Cows ON SACRIFICE FIELDS

M B McGechan, A Barnes

SRUC, West Mains Road, EdinburghEH9 3JG

SUMMARY

Pollution of water bodies by ammonium and particulate phosphorus arising due to outwintering beef and dairy cows was investigated using a weather-driven simulation model. This model demonstrated the influence of saturated soil conditions on rapid transport of these excreted contaminants through soil layers to the tile drains. Saturated conditions inevitably arise around the fringes of areas of heavily damaged (‘poached’) soil in the vicinity of feeders, so high losses are unavoidable where outwintered livestock are fed in a field. However, additional high losses can be reduced by avoiding outwintering livestock in fields with ponding, as this is an indication of saturated conditions in parts of the field away from the feeder.

Introduction

This work concerns the pollution of water bodies arising due to outwintering beef and dairy cows. Such pollution arises when excreted urine and faeces contaminate rainwater transported through the soil layers and tile drains to the river system. Pollutant transport processes have been studied by the method of simulation modelling, supported by observations at two of the field sites where outwintering was carried out.

Background and models

Contaminants and transport processes

Of particular interest is the process of ‘macropore flow’ or ‘by-pass flow’where excreted contaminants pass rapidly through the soil layers to reach the tile drains at a very short interval following deposition. During the winter period, this form of transport occurs in saturated soils, either at the fringes of ‘poached’ ground, orwhere ponding occurs due to the water table rising to the soil surface. In saturated soils, spaces around aggregates are water-filled, whereas in drier conditions water can only move through smaller pores within aggregates (McGechan, 2002a). The two contaminants considered here are ammonium derived from urine, and inorganic phosphorus attached to particulate faecal material. Water-borne ammonium passes rapidly to field drains by macropore flow, but if it enters the soil matrix it becomes sorbed and eventually converted to nitrate by nitrification. Particulate phosphorus from faecal material passes rapidly by macropore flow but is trapped if it enters the soil matrix.

Modelling tools

Modelling of contaminant flows to tile drains was carried out using the Swedish model MACRO. This is a dual-porosity model which has separate representation of water and contaminant dynamics in the macropore domain and in the soil matrix (micropore) domain. MACRO is a weather-driven model which is run using historical data for a number of weather variables, including precipitation and variables used for estimating evapotranspiration and soil evaporation. The model can be run for periods over which experimental measurements of contaminant flows have been made, or for longer periods for scenario testing. In previous studies MACRO has been calibrated using soil hydrological parameters, and tested using data describing water flows, ammonium and particulate phosphorus losses following slurry spreading and during grazing (McGechan, 2002b, 2003a, 2003b, McGechan et al., 2002).

Experimental sites

Two field sites were set up to observe the effects of overwintering of cows. At the first site with a sandy loam soil located at Easter Howgate farm on Bush Estate near Edinburgh, Scotland, a group of beef cows were outwintered during 2009-10 and 2010-11, with silage bales placed in a static feeder. At the second site with a silty clay loam soil located at Trawscoed, near Aberystwyth, Wales, a group of dairy cows were outwintered during the same two periods. However, at this site unwrapped silage bales were placed in a row across the field at the beginning of the winter, then unwrapped one at a time at an interval of two or three days with the feeder placed over the newly unwrapped bale. Weather data had been measured over long periods including the experimental periods at both Bush and Trawscoed. Soil hydraulic parameters needed for calibration of the MACRO model had previously been measured for these two soil types (McGechan et al., 2002).

Deposition of contaminants by excretion

Assumptions about quantities and concentrations of liquid, nitrogen and phosphorus deposited by excretion for a single cow were based on various literature values, including those for slurry as presented in RB209 (2010). However, it was assumed that ammonium concentrations in urine excreted from grazing cows would be higher than those measured in slurry since a large proportion of the urine-derived ammonium from housed cows volatilises before it reaches the slurry storage facility. Daily quantities of ammonium and inorganic phosphorus deposited at the experimental sites during the outwintering periods were then estimated taking account of the stocking rates of around 5 lu/ha.

Representation of poached areas around feeder

The situation in an outwintering field was represented by subdividing the field into three ‘zones’ with a different simulation procedure for each (Fig.1). Zone 1 is the poached area around the feeder with extensive soil damage due to trampling and compaction by animal hoofs. In such a zone, soil infiltration can be considered to be almost zero, with a permanent surface pool of contaminated water partially retained by the hoof-prints. The only possible movement of water is by surface runoff out of the zone to the adjacent field area. There will therefore also be a ring surrounding such a damaged area (Zone 2) where there is rapid infiltration by heavily contaminated runoff rainwater, which creates a saturated zone with potentially high losses of excreted contaminants. The area of this saturated zone will vary according to weather conditions, eventually dropping to zero if no rain occurs for several days (despite contaminated water remaining in hoof-prints in Zone 1). Zone 3 is the main field area away from the feeder where cows go to chew and lie down. In a real outwintering field, there are likely to be a number of small Zone 1 areas, each surrounded by a Zone 2 ring, but for

Figure 1: Sub-division of field into zones around feeder

simplicity a single Zone 1 surrounded by a single Zone 2 has been considered. It was also assumed that a cow would subdivide its time roughly half-and-half between Zones 1 and 3, with a stocking density much higher in Zone 1 than in the much larger Zone 3. The zone subdivision was further modified to represent the Trawscoed site where the feeder was moved at short intervals. In this case it was assumed that the feeder would be at five locations each for one fifth of the outwintering period, requiring five each of Zones 1 and 2, but the high stocking density in each Zone 1 for one fifth of the period only. Losses from the whole field are calculated as a weighted mean of those taking place in each zone, using the estimate of how much time animals remain in (and excrete in) each zone.

Contaminant transport associated with ponding

The second potential high contaminant loss situation arises in the main grazing field area(Zone 3),if the soil becomes saturated and the water-table rises to the surface following periods of high cumulative rainfall in poorly drained soils. This can be in soils with a high clay content, or if the tile drainage system is inadequate, typically if it is old with drainage tiles partially blocked, with periodic occurrence of ponding and saturation surface runoff.

Results of simulations

Representation of scenarios at experimental sites

In the main part of the field (Zone 3), losses of both ammonium and phosphorus were generally very low or zero (Fig. 2a), except in the case of the second winter at the Trawscoed site where there was one extended rainfall event which caused some surface runoff and very high losses of both contaminants (Fig. 2b). In this case, cumulative losses of ammonium over the outwintering period exceeded 10 kgN/ha for ammonium, and were around 0.3 kg/ha for inorganic phosphorus, with the losses occurring almost entirely during a single event lasting about 5 days.

With the three-zone model, losses of both contaminants were always high from Zone 2, except during extended dry periods when the area of this zone dropped to zero. However, when the

Figure 2: Simulated cumulative losses of ammonium and inorganic phosphorus over overwintering periods

a). First overwintering period at Easter Howgate, Scotland site

b). Second overwintering period at Trawscoed, Wales site

whole field was considered to give the weighted mean loss, the overall loss was lower than in Zone 2 but still generally much higher than in Zone 3 (Table 1). This represents a pollution level which should be a cause of concern.

Testing scenario of moving feeder compared to a static feeder

The effect of the feeder being moved at intervals was tested for the Easter Howgate site, by setting up for this site the moving feeder scenario similar to that at the Trawscoed site, and comparing simulated results with those for the actual static feeder scenario. However, the overall weighted mean loss was found to be almost the same regardless of whether the feeder was moved at intervals or remained static (Table 1).

Table 1: Cumulative pollutant losses for whole field over outwintering period (mid-November – end February) at two experimental sites, kg/ha

Site / Trawscoed, Wales / Easter Howgate, Scotland
Pollutant Winter / 2009-2010 / 2010-2011 / 2009-2010 / 2010-2011
Ammonium / 22.5 / 48.8* / 21.3 (20.7**) / 45.9
Inorganic phosphorus / 0.24 / 0.48* / 0.47 / 0.70
  • * Overall losses include high losses from Zone 3
  • ** Alternative scenario with feeder moved to five different locations during different periods throughout winter

Scenario testing using long periods of weather

A further investigation was carried out of the situation where high losses occurred in the main part of the field (Zone 3) due to saturated conditions with a high water table and ponding, as found at the Trawscoed site during the second experimental period. For this, simulations were carried out using historic weather data covering 10 winters at both the Scottish and the Welsh sites, and for the two soil types as found at the two sites. The influence of the efficiency of the field drainage system was also investigated. In fact, the field sites for which the model had originally been calibrated had newly installed, efficient tile drains at a spacing of around 7m. In practice, many grazed grassland fields have old, partially blocked ineffective drainage systems. Such an inefficient drainage system was represented in the model by specifying a drain spacing of 20m, a procedure adopted in a previous study.

Figure 3: Losses of ammonium in scenario simulations over ten years with two soil types and two drainage spacings, Trawscoed, Wales weather

Simulated results for ammonium loss showed that the weather had the greatest effect, with negligible losses for all options with the weather record for the Scottish site. With the Welsh weather with higher average winter rainfall, losses were particularly high with the inefficient drainage system (Fig. 3). Losses were somewhat higher with the silty clay loam soil than with the sandy loam soil.

Conclusions

This study has shown that outwintering of beef and dairy cows will lead to significant levels of water body pollution by ammonium and particulate phosphorus. Such pollution arises due to rapid transport of components of deposited excreta to tile drains through macropores in saturated soil during or after rainfall. Saturated soil conditions arisearound the periphery of any field areas which have become compacted due to trampling by animal hooves, and this situation is almost inevitable during outwintering. Saturated soil can also arise in a second situation after prolonged rainfall if the tile drainage system is inadequate so in effect the water table rises to the surface. In this situation, there is a significant additional level of pollution further to that arising from soil compaction due to trampling. It is very obvious when this second situation arises as it is associated with surface runoff and ponding.

Two clear conclusions arise in terms of recommending Best Management Practices to minimise water pollution during outwintering. Firstly, soil compaction due to trampling leads to a significant base level of pollution which is unavoidable. There is no benefit at all in moving the feeder to different locations periodically over the winter. Secondly, outwintering should not take place in any fieldwhich is subject to frequent ponding or surface runoff due to the inadequacy or degradation of the tile drainage system.

acknowledgements

This research was supported by the DEFRA project ‘Identification and mitigation of the environmental impacts of out-wintering beef and dairy cattle on sacrifice areas’ (contract no. SFFSD 0702

references

McGechan, M. B., Jarvis, N. J.,Hooda, P. S., & Vinten, A. J. A. (2002). Parameterisation of the MACRO model to represent leaching of colloidally attached inorganic phosphorus following slurry spreading. Soil Use and Management, 18, 61-67.

McGechan, M. B. (2002a). Water pollution due to macropore flow when wastes are spread on wet soil. Proceedings of SAC/SEPA Conference ‘Agriculture, wastes and the environment’, Edinburgh, 26-28 March 2002, pp 121-128.

McGechan, M. B. (2002b). Effects of timing of slurry spreading on leaching of particulate and soluble phosphorus explored using the MACRO model. Biosystems Engineering, 83, 237-252.

McGechan, M. B. (2003a). Modelling contamination of field drainage water by ammonium following slurry spreading. Biosystems Engineering, 85, 111-120.

McGechan, M. B. (2003b). Modelling phosphorus leaching to watercourses from extended autumn grazing by cattle. Grass and Forage Science, 58, 151-159.

RB209 (2010). Fertliser Manual (RB209), 8th Edition, TSO, Norwich.