Impact of horse grazing and feeding on phosphorus concentrations in soil and drainage water

Abstract

The number of horses in Sweden has increased, from 77 300 in 1970 to 283 000 in 2003 (ca. 250%). These horses are kept on 300 000 ha, which represents 10% of total agricultural land in Sweden. Maximum recommended livestock density in Sweden is 2.5 units/ha for grazed pasture, but no limits have yet been set for outdoor keeping and feeding areas (paddocks) for horses. This study characterized the potential risk of phosphorus (P) losses from a horse paddock established on a heavy clay soil with a stocking rate of 3.75 livestock units/ha compared with nearby arable land. The horse paddock received 15 kg P/ha/yr and 75 kg N/ha/yr through horse excreta, while annual input of P and N to the adjacent arable land was 13 and 112 kg/ha, respectively. There was no significant difference in water‐soluble P (WSP) in fresh and dried soil samples between the horse paddock (mean values: 0.62 and 0.43 mg/100 g soil; n = 15) and the arable field (mean values: 0.52 and 0.37 mg/100 g soil; n = 5). In contrast, phosphorus extractable in ammonium acetate lactate (extractable P) in the topsoil of the horse paddock (mean: 15 mg/100 g soil) was significantly higher (P = 0.03; n = 15) than in the arable land, whereas total P extracted with nitric acid (total P) showed no statistically significant differences. Furthermore, there was no significant difference in lactate‐extractable iron and aluminium (extractable Fe and Al), organic carbon (C), total nitrogen (N) or phosphorus sorption index between the two parcels of land. However, the degree of P saturation in soil was significantly higher (P = 0.02; n = 15) in the horse paddock. Extractable Al and Fe were highly correlated to extractable P (P < 0.001; n = 69), the correlation being negative for Al. No relationship was found with calcium, but soil C content was found to be correlated with extractable P (P < 0.001; n = 69). Over the past 8 yr, high P concentrations (up to 1.5 mg/L), mainly in dissolved reactive form, have been recorded in drainage water from the grazed catchment. We concluded that horse grazing at high stocking rates (>2.5 livestock units/ha) may pose a risk of high P losses to nearby water bodies.

Are horse paddocks threatening water quality through excess loading of nutrients?

“…However, they clearly indicate that some areas within paddocks can act as potential hotspots for elevated nutrient losses. Such hotspots may account for the higher nutrient losses from a paddock catchment than an arable catchment reported by Parvage et al (2011). According to Sharpley (1995), unless nutrients are removed with the harvested crop, high nutrient inputs to fields over long periods can exceed the nutrient retention capacity of the soil and cause subsequent losses.…”

“…After only one year of paddock management, Airaksinen et al (2007) measured an eighteen-fold increase of P in surface runoff from non-cleaned feeding areas. Eight years of field observation of drainage water, Parvage et al (2011) measured three-fold higher P leaching losses from paddocks than the adjacent arable field with similar annual P input. However, a general assessment of soil P (and other elements) status of the paddocks could not well explain such high losses.…”

Comparing different types of rural landowners: Implications for conservation practice adoption

“…For instance, studies have shown high nutrient levels in runoff from horse paddocks and pastures grazed at high stocking rates (Airaksinen et al 2001;Parvage et al 2011). Moreover, livestock (cattle, horse, and sheep) manure often contains high levels of E. coli (Escherichia coli) (Weaver et al 2005), and livestock grazing, in general, causes nutrient runoff and the sedimentation of waterways (Webber et al 2010).…”

Corrigendum to “A survey of soil phosphorus (P) and nitrogen (N) in Swedish horse paddocks” [Agric. Ecosyst. Environ. 178, 1–9]

“…Besides development of mitigation measures, introducing guidelines for new paddock establishment and its proper management is obligatory to avoid any further development of nutrient enriched zone. In order to reduce potential P and N leaching losses, four simple but probably rather effective measures can be proposed for horse paddocks from this and a previous study (Parvage et al, 2011). These are: (1) regular removal of excreta from paddocks (intervals depending on horse density); (2) covering of the soil within the feeding area or provision of a feed trough to avoid losses to soil; (3) limiting the horse density within paddocks to a maximum of 2.5 LSU ha À1 to keep P loading below 22 kg ha À1 year À1 ; and (4) shift of paddock area to other land use after 10-20 years.…”

“…The current recommended maximum horse density at farm level on Swedish pastures is 3 horses or 2.4 livestock unit (LSU) ha À1 (1 horse = 0.8 LSU (European Commission, 2013)) which is based on the amount of nutrients can be added (especially P, 22 kg ha À1 year À1 ) through manure to soil (Swedish Board of Agriculture, 2012). However, the recommended density is often exceeded when horses are kept in comparatively small, fenced-off paddocks near stables (Parvage et al, 2011). Furthermore, as feed not eaten by the horses remains on the soil and decomposes, and as horses do not excrete randomly but prefer an excretion zone mainly in the corners of the paddock and/or along fences, the potential exists for much larger stocks of P, N and other nutrients to build up within the paddocks (especially in feeding and excretion areas) than in other agricultural areas.…”