Research Accomplishment Reports 2009

Spatial and Temporal Characteristics of Grassland Agroecosystems

C. T. Dougherty
Department of Plant and Soil Sciences

 

Non-Technical Summary

Grazing system research is often limited in application because we have not been able to determine where, when, and how livestock are grazing in a grassland landscape that ranges from less than an acre to many thousands of acres, over as little as a few hours to many years. This problem may be addressed by monitoring where the animals are and what they are doing. It may also be addressed by careful sampling of pasture and soil transects and the application of advanced statistical techniques to introduce the dimensions of space and time to the defining data of grassland landscapes. This information will help determine the variables and sampling procedures needed to define the spatial and temporal properties of grassland landscapes.

Spatial statistics can be used to minimize the number of variables needed to define grassland landscapes in space and time. Application of spatial analysis should lead to better experimental designs and reduce the cost of grazing system research compared with those operating under classical designs. These procedures will reinforce research and models necessary to ensure the continued sustainability and profitability of grassland livestock enterprises as well as address present day and future environmental issues.

2009 Project Description

Quantitative analysis of grazing systems is limited at a landscape scale when conventional experimental designs and statistical analysis are used. Spatial and temporal statistical techniques have attributes that will advance the analysis of landscapes managed as grazing systems.

We recently completed analysis of a linear transect across a bermudagrass pasture grazed by horses in a six paddock system stocked with horses at 2.3 mares per hectare and a stock density of 13.75 mares per hectare. Soil and pasture variables were measured at 0.5 m intervals six weeks after the end of the grazing season. Spatial analysis detected effects on equine grazing not revealed by conventional statistical analysis. Spectral analyses indicated that the mares preferentially grazed herbage growing in hummocks that were about 1.5 to 2 meters from urination sites in the previous grazing cycle. This confirms observations that mares stopped grazing and urinated 1.5 to 2 meters from the area being grazed. This approximates the distance between the mouth and the landfall of urine of adult mares. In contrast to their urination behavior, mares defecated as they grazed. Spatial analysis, confirmed by observation, and contrary to convention wisdom, reveals that mares did not stop grazing and move to areas of the paddocks dedicated to dunging and urination.

Spatial arrangement of hummocks was initially determined by location of urine deposits from the first grazing cycle thereafter establishing the site of urine deposits in subsequent grazing cycles. Mares grazed down hummocks, invariably at one mare per hummock, although they grazed in close proximity to their herdmates. It was also noted that all of the mares in the herd tended to graze in the same compass orientation as has been reported in the literature. Mares urinated about three times each day and with mare grazing days established the density of hummocks.

Analysis of soil and herbage indicates that the area covered by urine deposits was enriched by up 2,800 kg per hectare of nitrogen and potassium per urination event, which accounts for the herbage mass of hummocks approaching 10,000 kg per hectare. Grazing mares consumed most of the herbage in hummocks during the next week long grazing phase and consequently these hummocks did not reform over the first generation urine deposit in the next recovery phase of the six-paddock rotation. Thus the urinary nitrogen and potassium was consumed in second generation hummock growth.

We concluded that hummocks were short lived if conditions were conducive to bermudagrass growth. Hummocks persisted longer if conditions were not conducive to grass growth. Spectral analysis also indicated repeating patterns in some soil and pasture variables occurring at 10-15 meters possible related to overlap of nitrogen fertilizer applications. Urinary excretion of nitrogen and potassium at such high rates indicates the possibility of these nutrients leach through macropores and slower pathways to the groundwater.

2009 Impact

This research has contradicted the conventional view that grazing horses graze in one area of a pasture and defecate and urinate in another area. It establishes that horses graze and defecate while in motion but stop grazing momentarily to urinate. It also shows that bermudagrass monocultures provide the energy and nutrient needs of grazing adult mares without supplementation and loss of body condition.

Grazing horses adhere to foraging theory preferentially grazing patches with high energy density, high biomass and high herbage nutritive yield as indicated by hummocks growing over urine deposits. The spatial grazing behavior of horses was shown to be complex with horse selecting herbage growing in hummocks over urine spots, while distant from but in close proximity to herd mates in a similar compass orientation. It also refutes the commonly-held view that horses do not graze in proximity to dung and urine.

We also determined that hummocks growing over urine deposits are short-lived if conditions favor grass growth, however they may persist over winter or in drought when growth is minimal. Further there is no indication that stocking rate or stock density moderate equine grazing behavior at relatively high stocking rates. Analysis of ingestion and excretion of herbage and soil nutrients indicate that urination behavior of horse may lead to enhanced nitrogen and potassium in ground water.