Abstract.The effects of different whole-farm management systems were explored in a farmlet trial on the Northern Tablelands of New South Wales, Australia, between July 2000 and December 2006. The three systems examined were first, a moderate input farmlet with flexible grazing on eight paddocks considered 'typical' of the region (farmlet B), a second, also with flexible grazing on eight paddocks but with a high level of pasture renovation and increased soil fertility (farmlet A) and a third with the same moderate level of inputs as farmlet B but which practised intensive rotational grazing on 37 paddocks (farmlet C).The changes in herbage mass, herbage quality and pasture growth followed a seasonal pattern typical of the Northern Tablelands with generally higher levels recorded over spring-summer and lower levels in autumn-winter but with substantial differences between years due to the variable climate experienced. Over the first 18 months of the trial there were no significant differences between farmlets in total herbage mass. Although the climate was generally drier than average, the differences between farmlets in pasture herbage mass and quality became more evident over the duration of the experiment. After the farmlet treatments started to take effect, the levels of total and dead herbage mass became significantly lower on farmlet A compared with farmlets B and C. In contrast, the levels of green herbage were similar for all farmlets.Throughout most of the study period, pastures on farmlet A with its higher levels of pasture renovation and soil fertility, had significantly higher DM digestibility for both green and dead herbage components compared with pastures on either of the moderate input systems (B and C). Thus, when green herbage mass and quality were combined, farmlet A tended to have higher levels of green digestible herbage than either of the other farmlets, which had similar levels, suggesting that pasture renovation and soil fertility had more effect on the supply of quality pasture than did grazing management. This difference was observed in spite of the higher stocking rate supported by farmlet A after treatments took effect.Levels of legume herbage mass, while generally low due to the dry conditions, were significantly higher on farmlet A compared with the other two farmlets. While ground cover on farmlet A was found to be less than the other farmlets, this was largely associated with the higher level of pasture renovation. Generally, all three farmlets had ground cover levels well above 70% for the duration of the experiment, thus being above levels considered critical for prevention of erosion.A multivariate analysis showed that the main explanatory factors significantly linked (P < 0.01) with the supply of high quality herbage were, in decreasing order of importance, those related to season and weather, pasture renovation, grazing management and soil fertility. Measurements of net pasture growth conducted using a limited number of grazing exclosure cages on three paddocks per farmlet reve...

Abstract. As part of the Cicerone Project's farmlet experiment, conducted on the Northern Tablelands of New South Wales, Australia, between July 2000 and December 2006, this study assessed the effects of varying soil fertility, pasture species and grazing management on the botanical composition of three 53-ha farmlets subjected to different management strategies. Starting with the same initial conditions, the farmlets were managed to reach different target levels of soil phosphorus (P) and sulfur (S); Farmlet A aimed at 60 mg/kg of Colwell P and 10 mg/kg S (KCl 40 ) whereas Farmlets B and C both aimed at 20 and 6.5 mg/kg of P and S, respectively. Pastures were renovated on six out of eight paddocks on Farmlet A, but only one paddock of each of Farmlets B (typical management) and C (intensive rotational grazing) was renovated. Flexible rotational grazing was employed on Farmlets A and B (each of eight paddocks) while Farmlet C used intensive rotational grazing over its 17 major paddocks, which were further subdivided into 37 subpaddocks. This paper focuses on the botanical composition dynamics observed across all three farmlets and the explanatory variables associated with those changes. Eight assessments of botanical composition were carried out at approximately annual intervals across each of the 37 major paddocks distributed across the farmlets and the results for each of 49 species were aggregated into seven functional groups for analysis. The strongest correlation found was a negative curvilinear relationship between sown perennial grasses (SPG) and warm-season grasses (WSG). The most significant factors affecting the functional group changes were soil P, sowing phase, paddock and date. These factors led to significant increases in SPG and correspondingly lower levels of WSG on Farmlet A compared with Farmlet B. Farmlets B and C experienced similar, declining levels of SPG, and increasing levels of WSG suggesting that intensive rotational grazing did not lead to substantial changes in botanical composition, compared with flexible rotational grazing, in spite of the fact that intensive rotational grazing had much longer grazing rests and shorter graze periods than the other two farmlets. Soil P levels were also significantly associated with levels of cool-season annual grasses, legumes and herbs, especially on Farmlet A. In general, the largest differences in botanical composition were between Farmlet A and the other two farmlets; these differences were most closely associated with those plants categorised as sown, introduced, C 3 pasture species. The levels of legume were generally low on all farmlets, due largely to the dry seasons experienced over most of the trial. Efforts to increase the legume composition on all farmlets were more successful on Farmlet A than on the other two farmlets due, presumably, to higher soil fertility on Farmlet A. Farmlet C, with its long rest periods and short graze periods, had a small proportion of legumes, due to the competitive effects of the accumulated tall grass herbage...

Abstract. Farming systems research conducted under dryland conditions is subject to the vagaries of the climate during the experimental period. Whether such an experiment experiences a representative series of climatic years must be examined in relation to the longer term climatic record. The Cicerone Project's farmlet experiment was conducted on the Northern Tablelands of New South Wales, Australia, to investigate the profitability and sustainability of three different management systems: one managed under typical, moderate-input conditions (farmlet B); a second which employed a higher level of pasture inputs and soil fertility (farmlet A); and a third which focussed on the use of moderate inputs and intensive rotational grazing (farmlet C).The climate experienced during the 6.5-year experimental period was compared with the 118-year climatic record, using a biophysical simulation model of grazed systems. The model utilised the long-term daily climate data as inputs and provided outputs that allowed comparison of parameters known to affect grazed pastures. Modelled soil-available water, the number of soil moisture stress days (SMSDs) limiting pasture growth, and growth indices over the experimental period (2000-06) were compared with data over the climatic record from 1890 to 2007. SMSDs were defined as when the modelled available soil moisture to a depth of 300 mm was <17% of water-holding capacity. In addition, minimum temperatures and, in particular, the frequency of frosts, were compared with medium-term (1981-2011) temperature records.Wavelet transforms of rainfall and modelled available soil water data were used to separate profile features of these parameters from the noise components of the data. Over the experimental period, both rainfall and available soil water were more commonly significantly below than above the 95% confidence intervals of both parameters. In addition, there was an increased frequency of severe frosting during the dry winters experienced over the 6.5-year period. These dry and cold conditions were likely to have limited the responses to the pasture and grazing management treatments imposed on the three farmlets. In particular, lower than average levels of available soil water were likely to have constrained pasture production, threatened pasture persistence, and reduced the response of the pasture to available soil nutrients and, as a consequence, livestock production and economic outcomes.Ideally, dryland field experimentation should be conducted over a representative range of climatic conditions, including soil moisture conditions both drier and wetter than average. The drier than average conditions, combined with a higher than normal frequency of severe frosts, mean that the results from the Cicerone Project's farmlet experiment need to be viewed in the context of the climate experienced over this 6.5-year period.

Abstract. This paper reports changes in livestock weights recorded in a whole-farmlet experiment, which aimed to examine the profitability and sustainability of three different pasture and grazing management strategies. The assessment of liveweights was considered a key component of measuring the integrated effects of the farmlet-scale treatments. The three farmlets comprised a typical management regime, which employed flexible rotational grazing over eight paddocks with moderate soil fertility (farmlet B), a system based on the same grazing management and paddock number but with higher levels of sown pasture and soil fertility (farmlet A) and a farmlet with moderate soil fertility and intensive rotational grazing over 37 paddocks (farmlet C).Early in the experimental period, there were no significant differences between farmlets in the liveweight of any class of livestock. However, from the second year onwards, as the pasture renovation, soil fertility and grazing management treatments took effect, differences in liveweight between farmlets became more apparent and significant. The stocking rate, which was treated as an emergent property of each farmlet, reached a maximum annual average value after 5 years of 12.6, 8.5 and 7.7 dry sheep equivalents (dse)/ha on farmlets A, B and C representing 84, 113 and 51% of their respective target stocking rates which were 15, 7.5 and 15 dse/ha.The liveweights of ewes, both before joining and during pregnancy, varied with year and farmlet with those on farmlets A and B tending to be significantly heavier than those on farmlet C. From 2003 to 2006, liveweights were significantly (P < 0.001) affected by a wide array of factors and their interactions including: date, ewe age, green digestible herbage, legume herbage mass, proportion of farmlet grazed, stocking rate and level of supplementary feeding.The weights of lambs/weaners/hoggets, both pre-and post-weaning, were at times also higher on farmlets A and B compared with those on farmlet C and were affected by a similar range of factors to those which affected ewe weights. Similar relative differences also applied to the liveweights of the other livestock run on the farmlets, namely wethers and non-reproductive cattle.The results suggest that stocking rate was able to be increased towards the higher target of farmlet A due to the higher level of pasture renovation and soil fertility on that farmlet, which led to high liveweights per head as well as the higher stocking rate. However, as the stocking rate increased on farmlet A, the differences between farmlets in liveweight per head diminished and the need for supplementary feeding increased. In contrast, the intensive rotational grazing practised on farmlet C did not allow the farmlet to increase its stocking rate towards its higher target. It appears that the higher proportion of each of farmlets A and B grazed at any one time allowed all classes of livestock to reach higher liveweights per head than on farmlet C, due presumably to the greater proportion of those two farmlet...

Abstract. Managing infections of sheep with gastrointestinal nematode parasites (worms) and problems of resistance to anthelmintic treatments continue to be major challenges for graziers on the Northern Tablelands of New South Wales, Australia. The whole-farmlet study of grazing enterprises undertaken by the Cicerone Project tested the broad hypotheses that compared with typical management (farmlet B), internal parasites can be more effectively managed with improved nutrition (farmlet A) or by intensive rotational grazing (farmlet C). Further aims were to identify the major sources of variation in faecal worm egg count (WEC) over the 6-year period and to examine the efficacy of the various anthelmintic treatments used during the experiment. This paper describes the management of sheep worms at the whole-farmlet level during the experiment, and analyses data from the routine WEC monitoring (5644 records) and larval differentiation tests (322 records) carried out on behalf of the Cicerone Management Board and by a doctoral candidate. It complements more detailed investigations published elsewhere.Over the period from July 2000 to December 2006, worm infections in ewes, lambs, hoggets and wethers were, with some exceptions, successfully controlled on the farmlets through a combination of regular monitoring of WEC, treatment with a wide array of anthelmintics and grazing management. Farmlet C had lower mean WEC (444 epg) and annual anthelmintic treatment frequency (3.1 treatments/year) over the whole experimental period than farmlets B (1122 epg, 4.3 treatments/year) or A (1374 epg, 4.7 treatments/year). The main factors influencing WEC were the time since the last anthelmintic treatment, and the anthelmintic used at that treatment. The magnitude of these effects dwarfed those of climatic and management factors that might be expected to influence the epidemiology of gastrointestinal nematode infections via environmental or host-mediated mechanisms. Nevertheless management factors associated with stocking rate and grazed proportion (proportion of each farmlet grazed at any one time), and climatic indicators of both temperature and moisture availability had significant effects on WEC.The results show that, in a region with Haemonchus contortus as the major sheep nematode, improved host nutrition in a higher input system (farmlet A) did not provide more effective control of gastrointestinal nematodes than typical management (farmlet B); however, it was observed that gastrointestinal nematode control was no worse on farmlet A than on farmlet B in spite of farmlet A supporting a 48% higher stocking rate by later in the trial period (2005). The study provided strong support for the proposition that intensive rotational grazing (farmlet C) provides more effective control of gastrointestinal nematodes than typical management (farmlet B) as evidenced by significantly lower WEC counts and anthelmintic treatment frequency. Tactical worm control based on routine monitoring of WEC provided adequate control of worms on all three ...

Abstract. Data from the Cicerone farmlet study were used to quantify the balance between pasture feed supply and the demand from grazing livestock, in terms of metabolisable energy (ME), on three differently managed farmlets (each of 53 ha) on the Northern Tablelands of New South Wales, Australia. Farmlet A had a high level of pasture renovation and higher soil fertility than the other two farmlets and employed flexible grazing management over eight paddocks. Farmlet B was designed to represent management 'typical' of the region and had the same grazing management and number of paddocks as farmlet A but moderate levels of pasture renovation and soil fertility. The third farmlet (C) had the same level of inputs as farmlet B but practised intensive rotational grazing over 37 paddocks.Regular measurements of the feed supply, namely herbage mass and quality, pasture growth and supplement fed and of feed demand were assembled to provide monthly estimates of the balance between feed supply and animal demand of all classes of livestock run on the experiment over its duration of 6.5 years. The significantly greater stocking rate, liveweight and reproductive rate of sheep reached on the higher input system (farmlet A) meant higher levels of ME were required to satisfy the nutritional demands of these animals. As only limited measurements were taken of animal intake, it was assumed that the supply of ME was derived from pasture growth and supplement fed. Using key livestock management dates and measurements of liveweights, the changes in the energy requirements of each class of animal were calculated and aggregated to provide an estimate of overall livestock energy demand over time. Subtracting the energy demand from the estimated energy supply provided a partial net energy balance.Measurements of the rates of change of green herbage during grazing events were found to be highly dependent on stock density with farmlets A, B and C recording rates of change of up to -50, -30 and -200 green DM/ha.day, respectively.Over a series of generally drier-than-average years, the ME supplied in pasture growth and through supplementation was at times inadequate to meet the energy demands of the livestock, resulting in periods during winter when the partial energy balance became negative. Similar feed deficits were observed for all three farmlets, suggesting that they were overstocked to a similar extent. In spite of the divergence in the stocking rate supported by each farmlet, the similarity of the ME balances between farmlets suggests that no farmlet was subjected to bias because of decisions relating to feed supply and demand. The analyses presented suggest there is considerable potential for practical paddock and grazing management to be improved if more timely and regular assessments can be made of changes in the feed energy supply using satellite images of normalised difference vegetation indices and feed energy demand using calculations of the ME required by grazing livestock.

Abstract. As part of the Cicerone Project's whole-farmlet experiment on the Northern Tablelands of New South Wales, Australia, the fat scores and reproductive performance of ewes were measured to assess the effect of different management systems on these important production parameters over time. The three farmlets (each of 53 ha) included one (farmlet B) subjected to 'typical' district management consisting of moderate levels of inputs and a target stocking rate of 7.5 dse/ha, with flexible grazing management across eight paddocks. A second farmlet (A) was managed in a similar fashion to farmlet B with respect to number of paddocks and grazing management, but modified by high rates of pasture renovation and higher levels of soil fertility, with a target stocking rate of 15 dse/ha. The third farmlet (C) was managed at the same level of moderate inputs as farmlet B but employed intensive rotational grazing over 37 paddocks and also had a high target stocking rate of 15 dse/ha. The experiment was conducted over 6.5 years from July 2000 to December 2006. In spite of the fact that target levels of stocking rate were chosen at the beginning of the experiment, stocking rate, together with fat scores and reproduction were treated as emergent properties of each farmlet system. Joining took place in April-May and lambing occurred in September-October of each year. Over the first 2 years of the experiment, there were few differences among farmlets in ewe fat scores or reproductive performance. From 2003 onwards, while the percentage of ewes pregnant was similar between farmlets, the average proportion of multiple births (ewes scanned in late July, with twins) was 30%, 16% and 12%, respectively, on farmlets A-C. However, lamb losses were greater on farmlet A, with average lamb mortalities recorded on farmlets A-C of 29%, 10% and 19%, respectively. Over the duration of the experiment, ewes on farmlets A and B were more often above a fat score level of 3, and less often below 2.5, than were ewes from farmlet C. Differences among farmlet ewes in fat score were found to be significant in 7 of the total of 13 assessments over the duration of the experiment. A generalised additive model applied to whole-farmlet data showed that green digestible herbage, legume herbage, stocking rate, the amount of supplement fed and especially the proportion of each farmlet grazed at any one time all influenced fat scores of ewes. While fat scores and conception rates tended to be highest on farmlet A, farmlet B had slightly better reproductive outcomes due to less lambing losses, whereas ewes on farmlet C tended to have somewhat lower fat scores and levels of reproduction. These farmlet-scale findings highlighted the importance for livestock managers to focus not only on grazing management, stocking rate and stock density during lambing, but also on the availability of sufficient green, and especially legume herbage, and the difficulty of overcoming a deficit in quality herbage with supplementation.

Abstract. The Cicerone Project conducted a grazed farmlet experiment on the Northern Tablelands of New South Wales, Australia, from July 2000 to December 2006, to address questions raised by local graziers concerning how they might improve the profitability and sustainability of their grazing enterprises. This unreplicated experiment examined three management systems at a whole-farmlet scale. The control farmlet (farmlet B) represented typical management for the region, with flexible rotational grazing and moderate inputs. A second farmlet (farmlet A) also used flexible rotational grazing but had a higher level of pasture renovation and soil fertility, while the third farmlet (farmlet C) had the same moderate inputs as farmlet B but employed intensive rotational grazing. The present paper provides an integrated overview of the results collated from component papers and discusses the inferences that can be drawn from what was a complex, agroecosystem experiment. The measurements recorded both early and late in the experiment were tabulated for each of the farmlets and compared with each other as relative proportions, allowing visual presentation on a common, indexed scale. Because of equivalent starting conditions, there was little difference between farmlets early in the experimental period (2000-01) across a wide array of measured parameters, including herbage mass, potential pasture growth rate, liveweight, wool production per head, stocking rate, gross margin and equity. Although the experiment experienced drier-than-average conditions, marked differences emerged among farmlets over time, due to the effects of treatments. During the latter half of the experimental period (2003-06), farmlet A showed numerous positive and a few negative consequences of the higher rate of pasture renovation and increased soil fertility compared with the other two farmlets. While intensive rotational grazing resulted in superior control of gastrointestinal nematodes and slightly finer wool, this system had few effects on pastures and no positive effects on sheep liveweights, wool production or stocking rate. Whereas farmlet A showed higher gross margins, it had a negative and lower short-term cash position than did farmlets B and C, due largely to the artificially high rate of pasture renovation undertaken on this farmlet during the experiment. Although farmlet B had the highest cash position at the end of the experiment, this came at a cost of the declining quality of its pastures. Modelling of the farmlet systems allowed the results to be considered over the longer timeframes needed to assess sustainability. Thus, returns on investment were compared over realistic amortisation periods and produced outcomes based on long-term climatic expectations which were compared with those that arose under the drier-than-average conditions experienced during the experimental period. The main factors responsible for lifting the productivity of farmlet A were the sowing of temperate species and increased soil fertility, which enhanced the amo...