Grazing Management Affects Soil Phosphorus and Potassium Levels

J.R. Gerrish, P.R. Peterson, and J.R. Brown1
Soil fertility can be affected by pasture utilization and manure distribution patterns of grazing cattle. Management to keep manure evenly distributed in a pasture to maintain soil fertility should be a producer goal. This research examined factors affecting soil fertility changes in grazed pastures. Grazing systems with 3, 12, or 24 paddocks and either having water available in each paddock or using a lane to provide water access were compared. Use of a lane to provide access to water for livestock resulted in declining P1 soil test values over a two year period. A trend (p=.07) existed for decline in exchangeable potassium soil tests in pastures using lane access to water. Phosphorus and potassium soil tests generally increased nearer to water in larger pastures grazed at relatively low stock density. Fertility gradients toward water source were the least in smaller paddocks grazed at a higher stock density. Providing water in every paddock and placing water to keep cattle travel distance to water under 800 ft should result in less soil nutrient transport off the grazable pasture area.

Introduction: Distribution of soil nutrients in a pasture varies across the landscape and is affected by many factors. Parent material, pre-settlement plant community, slope and aspect all contribute to the fertility status of a pasture. Human factors include both past and current management practices. Tillage, hay removal, and fertility application are examples of human effects. The grazing animals themselves will also affect soil nutrient availability and distribution. Differences in grazing distribution patterns, preferred loafing sites, and watering sites can all create nutrient gradients in pasture (Borrow, 1967; West et al., 1989). Mathews et al. (1994) reported greater concentration of exchangeable potassium in the front one-third of both continuously and rotationally grazed pastures. The research reported by both West et al. and Mathews et al. were both conducted on very small pastures with limited number of animals. Such research, while well designed and interesting, frequently yields information that may be of little consequence in commercial settings.

The challenge to a producer is to maintain soil nutrients at or near optimum levels for as great a part of the pasture as is economically feasible. Choice of forage-livestock management system can have a profound impact on the efficiency of nutrient return to grazing lands by the grazing animal. In this research project we have sought to determine some of the critical factors affecting soil nutrient status as it relates to grazing management.

Materials and Methods: A 3 year grazing study was conducted at the University of Missouri-Forage Systems Research Center located in the deep loess and drift region of north Missouri. Six grazing systems were established on pastures consisting of well established, diverse cool season grass-legume mixtures. Two cells each were subdivided into 3, 12, or 24 paddocks. One paddock system within each level of subdivision had stock water available in each paddock while the other system used lanes to provide access to water. Pastures ranged in size from 31 to 42 acres with individual paddocks ranging from 1.3 acres for the 24 paddock cell to 14 acres for the 3 paddock cell. Each system was grazed by both cow-calf pairs and yearling steers. Stocking rate was 2.0, 1.6, and 1.25 acres/cow-calf equivalent from mid-April through mid-July and 3.3, 2.7, and 2.1 acres/cow-calf equivalent from mid-July through mid-November for the 3-, 12-, and 24- paddock systems, respectively.

Detailed soil sampling was conducted on a total of 14 paddocks, three each from the 12- and 24-paddock systems and one from each of the 3-paddock systems. Each of the 14 monitor paddocks was sub-sampled on a grid basis with each paddock being divided into 60 to 80 sampling blocks. Soil samples were taken in the fall of 1992, 1993, and 1994 from 12 blocks in each paddock. Block sizes were approximately 30 X 30 ft, 40 X 40 ft, and 80 X 80 ft for the 24, 12, and 3 paddock systems, respectively. Sampled blocks were selected to provide a cross section of landscape positions in each paddock and provide a transect that could be spatially related to the water source.

Changes in soil P and K levels due to level of paddock subdivision or water access were analyzed using analysis of variance procedure. Spatially dependent changes in P and K levels were determined through linear regression procedures.

Results and Discussion: Lane effects: Both level of subdivision and water access had a significant effect on soil P levels (Table 1). The interaction term was not significant. Potassium levels were not significantly different due to either factor although a strong trend (p>F=.07) existed due to water access but not level of subdivision. The data in Table 1 does not include the sample block where water was located. In most cases, soil nutrients tend to accumulate in the watering area (Gerrish et al, 1993) but does not contribute to the productivity of the pasture. The greatest apparent P loss occurred in the 24 paddock unit with lane access to water. Cattle travelled the greatest distance to water in this system and were observed to spend a good deal of time in the lane. Manure distribution measurements (Peterson and Gerrish, 1995) indicate that approximately 13% of the dung was deposited in the lane.

The higher apparent loss of P in the 24 paddock system is related to the sampling procedure used and the more dense manure concentration in the 24 paddock system (Peterson and Gerrish, 1995). The student collecting soil samples had been instructed to avoid sampling in dung pats but in the 24 paddock system such a high proportion of the total area was visibly affected by dung that microsite areas of P concentration would have been avoided in the sampling process. On an annual basis, mean manure concentration per 500 ft2 in the 24 paddock system was 2 to 3 times greater than in the 3 paddock system. The mean manure concentration per 500 ft2 and apparent P loss for the 12 paddock system was intermediate between the 3 and 24 paddock systems. Saunders (1984) described differences in soil nutrients from sites either affected or unaffected by dung and urine. The impact of dung pats on pasture growth is more long term than is the impact of urine. This effect is due to two factors. Grazing animals avoid dung sites longer than urine sites and the subsequent growth tends to become more mature around a dung pat. The second factor is the relative nutrient content of dung versus urine. Most P is excreted in dung and the stability of P in both organic compounds and the soil results in a more long term effect on soil and plant growth. Urine is the primary excretory path for K and soluble N, thus the effect of urine is more short term.

The lane effect was not significant in the 3-paddock system. This is probably due to a confounding factor of natural water being available through much of the 3 paddock lane system during most of the grazing season. All stock tanks were equipped with water meters to monitor daily water intake. Very little water was actually drunk from the stock tank in the lane. Through most of the season, water would have been available within a few hundred feet of any point in the paddock.

The changes measured over the course of this study would not warrant P and K maintenance fertilization on an annual basis. Pastures using lanes could effectively be fertilized on a three year frequency based on our results. Grazing systems supplying water in every pasture and having optimal placement of that water, would apparently require very infrequent fertilization. We would recommend soil testing on a 3 to 4 year frequency to monitor nutrient status and plan fertilization strategies.

Even though the absolute change in soil tests for a particular pasture may not appear to be significant, the movement of nutrients within the system may be a point of concern. While average fertility may not decline at a biologically significant rate, some areas of the pasture may be declining in fertility at a greater rate as other areas are increasing in fertility level.

Spatial effects: Soil P and K distribution patterns in grazing cells of different configuration were examined. A 3-paddock cell with individual paddock size of approximately 10 acres and a 24-paddock cell with individual paddocks of approximately 1.3 acres were compared. Changes in both Bray P1 soil test and exchangeable K levels from 1992 to 1994 were spatially dependent in the 3-paddock grazing cell but not in the 24-paddock cell (Figures 1 and 2). Distance traveled to water was approximately 500 ft maximum for the 24-paddock cell while maximum travel distance was about 1250 ft in the 3 paddock cell. Grazing distribution would likely be different in these two situations with a higher degree of utilization gradient developing toward water in the larger paddocks (Gerrish et al, 1995). Greater grazing pressure in the front of the pasture resulted in a significant manure deposition gradient toward water in the 3 paddock cell while the manure deposition gradient was less extensive in the 24-paddock cell (Peterson and Gerrish, 1995).

A factor which also explains why greater nutrient gradients are likely to develop in systems having fewer paddock subdivisions is a time function. In the 3-paddock cell, the livestock have the opportunity to deposit manure around a particular watering site for 33% of the grazing season. In the 24-paddock cell, only about 4% of the grazing season is spent around a particular watering site, assuming water is made available in each paddock. Observation of the herd behavior in these two cells also provides some answers to the difference in fertility patterns. The cattle in the 3-paddock cell function very much as a herd. Rarely would there be animals in all parts of the 10 acre paddock simultaneously, rather they would remain congregated. In the 24-paddock system, the animals could frequently be seen scattered over the entire area.

Herd behavior is highly dependent upon the ability of individuals within the herd to maintain visual contact with their herdmates. In the smaller paddock, visual contact could be maintained at all times within almost the entire paddock area. Water placement was at midpoint on a slope and facilitated a good field of vision even in what is generally a rolling landscape. In the 3-paddock cell, water was at a lower point than most of the pasture and two swales cut the 10 acre pasture area making it impossible for visual contact to be maintained among the herd on any more than about one-quarter of the paddock. The herd in the 3-paddock system consisted of 9 cow-calf pairs and 13 yearling steers.

In larger herds and pastures, the animals can spread over several folds in the landscape and still maintain visual contact among the herd. Another mechanism that operates in larger herds is the breakup of the total herd into smaller social units or grazing herds The cow group in this research was not large enough to divide into grazing sub-herds which may have resulted in more uniform grazing distribution across the landscape.

Several practical soil fertility management recommendations which can be made based on these results. One of the most obvious is not to apply fertilizer within 100 to 200 feet of watering sites in a pasture. Shade areas where the animals are observed to spend a good deal of time should also be avoided. While the nutrient gradient does not seem to extend as far into the pasture area around shade compared to the gradient toward water, the nutrient level near to the shade is frequently even higher than the concentration around water. When designing rotational grazing systems, placement of water in each individual paddocks will result in much more uniform manure distribution and maintenance of fertility. If the net loss of phosphorus and potassium to lanes is converted to economic value, it is clear than just savings in fertilizer cost will pay for the water reticulation system in 3 to 7 years.

Literature cited:

Borrow, N.J., 1967. Some aspects on the effects of grazing on the nutrition of pastures. J. Aust. Inst. Agric. Sci. 33:254-262.

Gerrish, J.R., J.R. Brown, and P.R. Peterson. 1993. Impact of grazing cattle on distribution of soil minerals. p.66-70. In American Forage and Grassland Council Proc. Des Moines, IA, 29-31 March, 1993.

Gerrish, J.R., P.R. Peterson, and R.E. Morrow. 1995. Distance cattle travel to water affects pasture utilization rate. American Forage and Grassland Council Proc. Lexington KY, 12-16 March, 1995.

Mathews, B.W., L.E. Sollenberger, P Nkedi-Kizza, L.A. Gaston, and H.D. Hornsby. 1994. Soil sampling procedures for monitoring potassium distribution in grazed pastures. Agron. J. 86:121-126.

Peterson, P.R. and J.R. Gerrish. 1995. Grazing management affects manure distribution by beef cattle. In American Forage and Grassland Council Proc. Lexington, KY, 12-16 March, 1995.

Saunders, W.M.H., 1984. Mineral composition of soil and pasture from areas of grazed paddocks, affected and unaffected by dung and urine. N.Z. J. Agr. Res. 27:405-412.

West, C.P., A.P. Mallarino, W.F. Wedin, and D.B. Marx. 1989. Spatial variability of soil chemical properties in grazed pastures. Soil Sci. Soc. Am. J. 53:784-789.


Table 1. Change in soil phosphorus and potassium levels in grazing systems differing in level of subdivision and water access.

                            P1 soil test                  K soil test    
             Water      --------------------         ---------------------
System       Access     1992   1994   Change         1992    1994   Change
                        ---- (lb/A) ----               ---- (lb/A) ----     
3-paddock    Paddock     20     21     1              226     223     -3
             Lane        27     26    -1              229     199     -30

12-paddock   Paddock     19     20     1              224     212     -12 
             Lane        24     20    -4              252     228     -24

24-paddock   Paddock     23     19    -4              246     224     -22
             Lane        27     19    -8              227     200     -27

                               LSD = 2.5                              n.s.

Figure 1. Change in Bray P1 soil test in two grazing systems each having water available in each paddock.


Figure 2. Change in K soil test in two grazing systems each having water available in each paddock.


1Research Assistant Professor, University of Missouri-Forage Systems Research Center, RR1 Box 80, Linneus MO 64653; Assistant Prefessor, Plant Science Department, McGill University, Montreal, Quebec, Canada; Professor Soil Science, School of Natural Resources, University of Missouri.