**1. Introduction**

Improving availability of phosphorus (P) and at the same time reducing its loss to streams can contribute to sustainability of grazed pastures. Phosphorus is a vital nutrient for plants and animals, but it has been long identified as a major contributor to eutrophication. The underlying problem of the existing P-cycle is the failure to recover and reuse P in human waste, livestock manure, and food waste [1]. Grazing animals urinate and defecate around 81% of the P eaten [2]. Retention of P in animal waste, within the soils of grazing systems, will improve the efficiency of P use which could contribute to solving this problem. Management practices aimed toward vertical and horizontal distribution of P for better retention and utilization could reduce overall P losses from pastures. Distribution of P should focus on utilization of P for plant production while at the same time maintain soil surface properties that maximize infiltration, minimize over-land flow [3] and P improved water quality.

Internal inputs of P in pastures from weathered parent materials and external inputs from mineral fertilizers and dust deposited from the atmosphere are internally cycled through dung deposited by cattle, plant residues, and animal matter [4].

Surface application of phosphatic fertilizers exceeding the agronomic requirements of forages to kaolinitic soils has been shown to transfer readily soluble forms of inorganic phosphorus to lower depths in soil profile [5]. However, a study by McClaren et al. [6] reported total concentration of organic and inorganic forms of soil P in the 10–20 cm soil layer to be only 42% of the concentration measured in the 0–10 cm soil layer. Dung and plant residues are organic sources of P in soil that can reduce the need for external inputs of P, however, nutrient management strategies seldom consider both inorganic and organic forms of P [7]. Dung deposited by cattle during in-situ grazing can have cumulative benefits of improved pH and decreased P sorption, thus improving the efficiency of P cycle in the long term [8]. Adding P to the soil cannot be the sole strategy as research has shown increased losses of P with increased accumulation of P in soil. Phosphorus is relatively immobile in soil [9] until it reaches relatively high levels of soil saturation. Studies have shown that around 80% of manure P incorporated into soil by rain remains in the top 2 cm [10]. Surface soil P (2 cm) is a major contributor of P in runoff due to the desorption by runoff water [10], and this effect is exacerbated when vegetative cover is minimal [11]. Hence, P management should encompass strategies aimed to uniformly distribute available P to ensure better production of grass and forages and at the same time prevent addition of P in vulnerable areas with high transport and export potential to aquatic systems.

Managing P in pastures is particularly challenging given the diversity of landscapes under pastures and the complexity of P cycling, which is usually site specific. Dung produced by animals in pastures is deposited in patches that correlate with animal loci due to microtopography [12]. Dung returned to pasture soils is spatially heterogenous with greater dung P concentration near resting places (waterers, feeding stations, shades) [13,14].

Inherent vulnerability of sites based on proximity to streams and vulnerability to runoff and erosion should be considered [15]. The incidence of extreme events is expected to increase in the coming decades and research has reported increased rainfall intensity [16,17] and runoff [18,19] can increase P loss. Where P is deposited by cattle may greatly influence its retention within the pastures. In their study on grazed pastures, under contrasting grazing management and fertilizer applications, Bilotto et al. [20] reported greater mean annual changes in soil P on low slopes in comparison to high and medium slopes attributing the difference to movement of phosphorus in animal dung from higher slopes to the lower slopes. Nellesen et al. [21] indicated greater loss of P from pastures with unrestricted stream access compared to pastures with restricted stream access. Grazed grasslands are a significant source of P inputs to surface waters [16]. Additionally, Kurz et al. [22] suggested higher loss potential of organic P in pastures due to manure from grazing animals. In pastures under in-situ grazing, cattle dung stabilized P and increased soil pH [8]. Thus, effective P management strategies should involve use of techniques to reduce continuous treading of soil and excessive inputs of manure at vulnerable sites and maintain continuous vegetative soil cover to lower the losses as both particulate and dissolved P forms.

It is therefore important to generate strategic management systems that can uniformly distribute P inputs (dung and mineral fertilizers) in the pastures such that surface accumulation of P in erosion-prone regions of pastures is minimized. Areas with free access to cattle had 57–83% lower soil macroporosity and 8–17% greater bulk density when compared to areas where cattle were excluded [22]. Use of off-stream watering points (OSWPs) can have a potential benefit of reducing time spent by cattle in riparian areas, but, inclusions with shades, consideration of slope, size of paddock, and good grazing management practices could influence the effectiveness of OSWPs [23]. Rotational grazing in combination with a fenced riparian buffer can be effective in reducing runoff and erosion from pasture soils [24]. Management practices such as stocking rate and methods to manipulate distribution of shade structures, supplement feeding stations, waterers, fertilizers, and forage species diversity can affect efficiency of nutrient cycling in pastures [25].

Grasslands managed under different management practices have been extensively studied for nutrient cycling that encompasses nutrient composition of soils, input of nutrients, and management strategies to minimize losses and associated risks. Research on the relationships between soil test P and runoff P at the field- or pasture-scale is lacking. Most of the previous studies in grazing management have been small-scale plot studies and runoff simulation studies. Understanding the extent to which grazing management can vertically and horizontally retain and redistribute soil P concentrations and reduce P in runoff water at the pasture scale is needed. Strategies such as rotational grazing, exclusion of areas vulnerable to loss of P, and lure management of cattle have been identified as best management practices with the potential to improve P distribution, recycling, and retention in grazed pastures. This study compares two combinations of best management practices, strategic grazing (STR) or continuous-grazing-with-hay-distribution (CHD), to determine their impacts on distribution of soil P (Mehlich-1 P, M1P) and P loss as dissolved reactive P (DRP) and total Kjeldahl P (TKP) in runoff water, over a period of three years.
