37 Figure 4.6 Spatial pattern of spots (white) and non-spots (grey) on five treatments on Kroomie. 41 Figure 4.8 The ratio between the average size of all patches and all non-patches for each treatment.
LIST OF APPENDICES
It turned out that there are two different grass communities in Kroomie, and these are associated with shoulder height (Le. patches and non-patches). Application of CCA demonstrated a significant difference between patch and non-patch species composition.
ACKNOWLEDGEMENTS
OVERVIEW
- INTRODUCTION
- OVERVIEW OF PATCH DYNAMICS
- THE IMPACTS OF SOME L1iRGE HERBIVORES ON GRASSES
- ANIMAL SELECTIVITY
- GRASS GROWTH HABIT IN RESPONSE TO GRAZING
- CHANGES IN SPECIES COMPOSITION
- SUMMARY.AND CONCLUSION
In the agricultural sense, this is considered the most favorable shoulder structure for animal production; or. For example, O'Connor (1994) found that species abundance was more responsive to rainfall variability than grazing, while O'Connor and Roux (1995) found community variation in the karoo.
DESCRIPTION OF THE EXPERIMENT
- INTRODUCTION
- EXPERIMENTAL SITE
- LOCATION, TOPOGRAPHY AND SOILS
- VEGETATION
- DESCRIPTION OF THE KROOMIE TRIALS
- MANAGEMENT VARIABLES TESTED
- DESCRIPTION OF THE TREATMENTS
- STUDY PERIOD
However, it should be noted that the long-term occupancy rates of the CR and NR treatments are similar. This study is possible as a result of the long-term trials, but does not directly report on the effects of animal type, stocking rate or grazing system on pasture dynamics or secondary production.
- GENERAL INTRODUCTION AND AIM
- TREATMENTS STUDIED
- STUDY 1: SEPARATION OF PATCHES AND NON-PATCHES USING MAXIMUM LIKELIHOOD ESTIMATION
- INTRODUCTION AND AIM
- METHODS
- RESULTS AND DISCUSSION
- STUDY 2: DESCRIPTION OF PATCHES BASED ON SPECIES COMPOSITION AND SWARD HEIGHT
- INTRODUCTION AND AIM
- METHODS
- RESULTS AND DISCUSSION
- SUMMARY.AND CONCLUSION
The purpose of this section was to determine the grass patch structure of the five treatments using a disc meter, and then analyze the distributions using MLE to determine the height at which patches were distinguished from non-patches. It was expected that the grass structure of all treatments would be bimodally distributed, and the height of the patch/non-patch interface would be the same for all treatments. a Patches b I - Combined I. The two structural components of the turf (i.e. patches and non-patches) have an interface at approximately 3.5 cm (vertical arrow).
An attempt was made to determine the height of the patch/non-patch interface using MLE. The aim of this research was to correlate the species composition of the vegetation with the grass height, to determine whether patches can be separated from non-patches at a certain height based on the species composition. In each subquad, vegetation height was measured using a disc meter (single reading).
The species composition of each subquadrat was then described by estimating the basal cover of each of the three most common species (see Appendix 2 for details). By combining the data from the four sub-quadrats, we get a more accurate estimate of the species composition of the 1 x 1 m square (where up to nine species can be included). For four out of five treatments, he predicted the patch/non-patch interface to occur between 5 and 6 cm.
VARIATION IN SWARD GEOMETRY ACROSS TREATMENTS
- INTRODUCTION AND AIM
- METHODS
- RESULTS AND DISCUSSION
- EXTENT OF PATCHES ACROSS TREATMENTS
- VARIATION IN THE SIZE OF PATCHES ACROSS TREATMENTS
- LAYOUT OF PATCHES AND NON-PATCHES ACROSS TREATMENTS This section serves to
In other words, a 42% increase in stocking rate resulted in a 100% increase in patch-grazed area. Therefore, under conservative stocking rates, the total area grazed appears to be correlated with the metabolic mass of animals present, regardless of animal species. If the stocking rate of sheep was increased, I would expect a smaller increase in the patch grazed area.
Therefore, data showing the frequency of patches of different sizes were converted to reflect the relative proportion of the total area grazed in patches of different sizes. The result is the total 'length' of patches in each size class, which is a linear index of the total grazed area (Figure 4.4). The results show that, although there is a predominance of small patches, a significant part of the total grazed area is found within large areas (Figure 4.4).
However, the total area covered by CH is approximately twice that of the CR treatment. However, in the CR treatment, the total area grazed is significantly greater under continuous than under rotation. Both distributions show positive kurtosis and strong positive skewness: again, the total grazed area is similar.
SUMMARY AND CONCLUSION
The grazed area is influenced by the grazing system and stocking rate, with continuous or heavy stocking resulting in an increase in the grazed area. The effect of stocking density is consistent with results from other studies, while the effect of the grazing system can provide insight into why animal performance is higher with continuous than with rotational stocking. The lack of effect of animal type may be due to the low stocking rate or density at Kroomie, but this conclusion is tentative.
All treatments showed a high predominance of small spots and their frequency decreased exponentially, described by a decline curve, as spot size increased. However, the distribution of the total patch-grazed area among patches of different sizes differed between treatments. Animals that were regularly rotated at conservative rates tended to graze within small areas, while animals that were large-scale or continuously grazed over a larger size range.
Analysis of peakedness and skewness of these distributions indicated that distributions were always positively skewed, particularly in the CR and SR treatments. Correspondingly, the CR and SR treatments were strongly peaked, while the other treatments were slightly flattened. I suspect that the distributions would be negatively skewed - in a heavily grazed system, one would predict that a relatively small proportion of the total area grazed would fall within small patches, and a larger proportion within large patches.
THE RELATION BETWEEN SWARD STRUCTURE AND SPECIES COMPOSITION
- INTRODUCTION AND AIM
- SPECIES COMPOSITION OF PATCHES AND NON-PATCHES
- INTRODUCTION
- METHODS
- RESULTS AND DISCUSSION
- DEGRADATION OF PATCHES
- INTRODUCTION
- METHODS
- RESULTS AND DISCUSSION
- SUMMARY AND CONCLUSION
Furthermore, it was demonstrated that the species composition of these patches and non-patches differed. In other words, species composition varies more between patches and non-patches than between treatments. Therefore, thematic plant densities in patches and non-patches were measured and used as an index of degradation.
During April and May 1999, the density of theme plants on patches and non-spots was estimated (by this time of year all the theme plants were established and no young seedlings were found). In all treatments, the density of themes on patches and on non-patches differed significantly. There was very little apparent effect of animal type on the density of themda on patches (compare SC and SR with CC and CR treatments).
Using themed density as a measure of decomposition (low densities equate to decomposition), experiments at Kroomie showed that continuous stocking systems (for cattle and sheep) outperformed rotational systems (compare SC and CC treatments with SR and CR treatments). There is an interesting relationship between temeda plant density and patch and non-patch size. The relationship between the density of the subject and the size of the corrections should also be taken into account.
MOVEMENT OF PATCHES
- INTRODUCTION
- CAN PATCHES MOVE?
- ANECDOTAL EVIDENCE ON THE MOVEMENT OF PATCHES
- OLD FENCELINE CONTRAST
- RESIDUES FROM A PREVIOUS STUDY
- ACACIA TREES THAT HAVE FALLEN OVER
- GRAZING OF NON-PATCHES AFTER A PERIOD OF LOW RAINFALL
- CONCLUSION
The fence was then removed and the sheep were able to graze freely throughout the pasture. After seven years, the original boundary line was still clearly visible: the patch/non-patch interface extends in a straight line for approximately fifty meters (Figure 6.1). In 1996, a study was initiated to study, among other things, the life history of grass plants on adjacent plots and non-plots (Martens 1996, unpublished data).
In almost all cases, there was a group of such rings in one patch, and a similar group nearby (less than a meter away) in a patch. This evidence tentatively suggests that, as found with the fence contrast, the spots are relatively stationary during the medium term. In all the cases I observed, the grass grew in the protected area, and usually quite quickly (several months).
This indicates that the grass in the patch, when relieved of grazing pressure, regains vigor and grows tall. Furthermore, it suggests that themeda can be present in the seed bank and germinate and develop in the absence of grazing. As for patch movement, this anecdotal evidence suggests that non-patches can form within patches.
CONCLUSIONS
- GENERAL REVIEW AND CONCLUSIONS
- SEPARATING PATCHES AND NON-PATCHES
- SWARD STRUCTURE
- SPECIES COMPOSITION
- MOVEMENT OF PATCHES
- OVERALL CONCLUSIONS, AND COMMENTS ON FUTURE RESEARCH 1. The formation of patches is a ubiquitous consequence of grazing at Kroomie
This is at odds with other studies, but may be due to the low levels of animal stocking at Kroomie. Due to the nature of the trials, the effect of sheep at high stocking levels could not be tested. Similar trends were evident when considering the size of spots formed in the five treatments.
Cattle and sheep grazed mainly in small patches, if grazed continuously at low livestock numbers, and in small and large patches if continuously or at high livestock numbers. As expected, there was a general inverse relationship between the size of patches and of non-patches. The species compositions of patches were generally similar across treatments - this was also the case for non-patches.
The research reported here provided an objective method for separating patches and non-patches in Kroomie. The technique is sensitive and consistent in fields grazed by both smallmouth (sheep) and largemouth (cattle) species, under conditions of rotational and continuous grazing, and at high and low stocking rates. A critical question is whether this is consistent over several seasons, and what role the existence of patches and non-patches plays in plant dynamics in relation to plant species richness and diversity.
In: Proceedings of the Prestige Grazing Symposium held in Adelaide on 10 November 2000 (ed. du Toit JCO). Influence of veld condition and stocking intensity on species selection patterns in cattle in the southern high grasslands of Natal. TWINS PAN - FORTRAN program for multivariate data in an ordered two-way table with classification of individuals and attributes.
The influence of urine and dung deposition on spotted grazing patterns of cattle and sheep in the southern high range. Influence of rainfall and grazing on the compositional change of the herbaceous layer of a sandveld savanna. Vegetation changes (1949–71) in a semi-arid, grassy dwarf bush landscape of the Karoo, South Africa: influence of rainfall variability and sheep grazing.
Sequence and process of cattle species selection in relation to optimal foraging theory on an old land in the Natal Sour Sandveld. The effect of field condition on the quality of the diet selected by cattle grazing the Natal Sour Sandveld. An evaluation of the empirical basis for grazing management recommendations for grasslands in southern Africa.
