A Comparison of Methods Used to Determine Biomass on Naturalized Swards

An experiment was conducted in 2000 to compare simple visual estimate, sward height and rising plate meter (RPM) methods for determining forage biomass in mixed-species, naturalized, rotationally grazed dairy and beef pastures. Measurements were taken pre- and post-grazing on 10 sampling dates at the dairy pasture, and post-grazing at 13 sampling dates at the beef pasture. For each sampling date, the effectiveness of each method for estimating the actual biomass from a quadrat was evaluated using regression analysis. The results for the visual estimate method were not consistent, with non-linear relationships occurring early and late in the season. While the meter stick was most effective in the dairy pasture, the RPM was most effective in the beef pasture. Species composition and structural characteristics of the stand were important factors affecting accuracy of biomass estimation. Equations developed for each method and site using data from all dates had low R²-adjusted values, and were unreliable predictors of biomass. The results from individual sites and dates were extremely variable, with no single method effective in all circumstances. To estimate forage biomass in mixed-species, naturalized pastures, standard quadrat harvesting remains the most reliable method, provided that enough quadrats are clipped to adequately represent a given area.     

Sward Height and a Capacitance Probe for Estimating Herbage Mass

Estimation of herbage mass is essential in pasture research and in making practical grazing management decisions. Sward height and a capacitance probe were evaluated as non-destructive measures of herbage mass at seven dates from April 16 to July 30 on continuously grazed predominantly Kentucky bluegrass ( Poa pratensis L.) pasture. Paddocks were 1 ha in size and were stocked at 3, 4 or 5 steers ha⁻¹ (average initial weight 312 kg). Four to six, sampling points were located randomly in each paddock at each date. Sward height was measured with a ruler, the vegetation in a 0.2 nr area visually designated "high" or "low" according to legume percentage, ten readings using a single capacitance probe were made and the herbage clipped. Paired data with clipped dry weight as dependent variable and height and/or the average corrected probe reading as covariates were analyzed statistically. The relationship between clipped dry weight and nondestructive measures of herbage mass was found to be linear. There was a small effect of sampling date in Spring on the relationship between destructive and non-destructive measures of herbage mass, but each of the measures, including herbage mass predicted from probe and height measurements, resulted in the same inferences being made about differences between treatments. Neither stocking rate nor small differences in percentage legume affected the relationship. Thus, sward height and capacitance probe show promise as nondestructive measures of pasture herbage mass.     

Productivity of Bahiagrass Pastures in South-western Japan: Synthesis of Data from Grazing Trials

This study examined the rate of herbage production and herbage quality of bahiagrass (Paspalum notatum Flügge) using data from five grazing trials in the low‐altitude region of Kyushu, south‐western Japan, in an effort to (a) evaluate productivity of bahiagrass pastures, (b) analyse its relationship to meteorological, vegetational and managerial variables, and (c) obtain implications for better management of bahiagrass pastures. The rate of herbage production, ranging from ?56 to 213 kg DM ha^?1 day^?1, tended to increase from spring (April–May) to mid‐summer (July) and decrease thereafter. The rate was expressed by a multiple regression equation where nitrogen fertilizer rate, air temperature, rainfall and herbage mass had positive effects. Dry matter digestibility (DMD) and crude protein (CP) concentration of herbage were in the range of 471–727 and 84–161 g kg^?1 DM respectively. DMD was expressed by a regression equation where the day number from 1 April and herbage mass had negative effects, and the sampling height and nitrogen rate had positive effects. CP concentration was expressed by an equation showing a positive effect of nitrogen rate and a negative effect of herbage mass. The results indicate that management of bahiagrass pastures should aim at maintaining herbage mass closely above the critical level below which intake by grazing animals is restricted, in order to increase quality and ensure quantity. This is particularly important when nitrogen fertilizer is applied.     

Pasture restoration improvement methods for temperate degraded pastures and consequences of the climatic seasonality on soil–pasture complex

Short persistence of renovated pastures is a major problem for pastoral production systems. Pasture species establishment, species performance and persistency and their relationships with soil–water attributes following pasture improvement methods were evaluated over two years. Fertilised naturalised pasture (FNP), non-fertilised naturalised pasture (NFNP), Lolium perenne–Trifolium repens cultivated pasture (CP), L. perenne–T. repens direct-drilled pasture (DP), and Bromus valdivianus, Holcus lanatus, Dactylis glomerata, L. perenne and T. repens direct-drilled pasture (DDP) were established as a randomised complete block design. Pastures were limed, fertilised and sheep-grazed. Plant growth compensatory mechanisms, by decreasing tiller number and increasing tiller size, equalised pasture yield during the establishment. In both years, the improved pastures yield was greater than that of NFNP. In the second year, DDP and FNP reached the higher pasture growth rate, with a seasonality diminishment. During the second summer, soil matrix potential at 0–20 cm soil profile exceeded the permanent wilting point, resulting in the pasture growth rate of NFNP and the sown pastures ranging from 0.0 to 3.6 kg DM/ha per day, while for FNP, it was 17.8 kg DM/ha per day. Sown species, fast-growing species spontaneous colonisation and growth were all stimulated in the improved pastures over species that tolerate stress, improving agronomical and ecological pasture features.     

Quantifying the effect of nitrogen on productivity of cocksfoot (Dactylis glomerata L.) pastures

Inorganic nitrogen fertilisers are commonly applied to crops and pastures to increase or maintain productivity. The benefits of N application must be balanced with the potential for environmental damage. At Lincoln University, Canterbury, New Zealand a split plot experiment with two irrigation levels (irrigated and dryland) and two N fertiliser application rates (0 and 800 kg N/ha/y (2003/2004) or 1600 kg N/ha/y (2004/2005) was established on a 9-year-old cocksfoot (Dactylis glomerata L.) dominant pasture to validate the N dilution curve for temperate grass species. The extent of N deficiency of pastures which had suboptimal N was quantified by calculation of a ratio between measured N% of herbage and optimum N% from the N dilution curve. The N dilution curve had the form N% = 4.8 DM^?0.35 (R² = 0.65) and the NNI ranged from a maximum of 1.2, which indicated luxury uptake, to a minimum of 0.2, which has been proposed as the minimum NNI required to result in net growth. When moisture was adequate for growth, the main cause of yield differences between +N and ?N pastures was radiation use efficiency with ?N pastures producing 0.54 g DM/MJ PAR in spring or less than half that of +N pastures (1.16 g DM/MJ PAR). The intrinsic link between water availability and N uptake in dryland and irrigated pastures was explained (R² = 0.88) by the relationship between the water use efficiency and N uptake ratio. Periods of low N uptake (N deficiency) were associated with low water use efficiency.     

Management targets for continuously stocked mixed oat × annual ryegrass pasture in a no-till integrated crop–livestock system

Feeding livestock with cover crops can improve the efficiency and sustainability of integrated crop–livestock systems under no-till. However, no-till systems are based on permanent soil cover by organic material, so grazing livestock can compete for soil cover. Hence, managing stocking rates during the grazing period of the cultivated forage species is a key factor to assure enough herbage mass for maintaining long-term sustainable no-till systems. In this context, the objective of this study was to determine sward management targets for a continuously stocked mixed oat (Avena strigosa) × annual ryegrass (Lolium multiflorum) pasture in rotation with soybean in a no-till integrated crop–livestock system to determine the optimum balance between animal production and herbage mass for soil cover. The effects of sward height management on animal performance and herbage mass covers were evaluated. Treatments corresponded to four sward heights: 10, 20, 30 and 40 cm, maintained throughout the experimental period through continuous stocking and variable stocking rate, plus a no-grazing control area. Treatments were arranged in a randomized complete block design with three replications. Herbage mass and animal performance increased linearly with sward height, but weight gain per hectare decreased. Grazing efficiency fitted to a quadratic regression and conversion efficiency a logarithmic model. Equilibrium between grazing and conversion efficiencies was reached on swards managed at 20 cm, indicating that this sward height provided enough herbage mass to allow both animal performance and no-till crop demand for soil cover.     

Italian ryegrass establishment by self-seeding in integrated crop–livestock systems: Effects of grazing management and crop rotation strategies

We evaluated the re-establishment of an Italian ryegrass pasture by self-seeding on a no-till integrated crop–livestock systems (ICLS) in the southern region of Brazil. This work is part of a long-term experimental protocol initiated in 2003. We tested the effects of various management practices, such as summer crop systems (soybean vs. maize–soybean rotation), stocking methods (continuous vs. rotational) and grazing intensities (low vs. moderate), on Italian ryegrass pasture establishment. In addition, we tested resilience of the system by testing pasture's ability to re-establish following a year without seed head production. The experiment consisted in the rotation, on the same area, of Italian ryegrass pasture grazed by sheep during the winter and up to the end of the grass production cycle, and soybean or soybean–maize grain crops rotation cultivated during the summer. The pasture established itself by self-seeding since 2005. Data were collected in 2011 and 2012 stocking season. The soybean summer crop, continuous stocking and low grazing intensity, all positively affected the production of reproductive tillers in 2011. Grazing intensity in 2011 strongly influenced early vegetative tiller densities (before crop harvest) in 2012. However, none of the grazing intensity or the stocking method treatments affected herbage mass at the end of pasture establishment in 2011 or 2012. On the other hand, the soybean summer crop positively affected pasture establishment, both in term of tiller densities and herbage mass at the end of pasture establishment. The removal of all seed heads in 2011 (preventing seed production) resulted in the total failure of pasture establishment in 2012. Overall, Italian ryegrass establishment by self-seeding relies on the annual replacement of the soil seed bank. This experiment demonstrated that under various stocking methods, moderate grazing intensity and maize or soybean summer crop, Italian ryegrass pasture establishment by self-seeding remains successful even when the stocking periods extended up to the end of the grass production cycle. Self-seeding with moderate grazing intensity ensures successful pasture establishment, reduces labour and costs and allows to increase the stocking period and so animal live weight gain over the grazing season.     

Italian ryegrass establishment by self-seeding in integrated crop-livestock systems: Effects of grazing management and crop rotation strategies

We evaluated the re-establishment of an Italian ryegrass pasture by self-seeding on a no-till integrated crop-livestock systems (ICLS) in the southern region of Brazil. This work is part of a long-term experimental protocol initiated in 2003. We tested the effects of various management practices, such as summer crop systems (soybean vs. maize–soybean rotation), stocking methods (continuous vs. rotational) and grazing intensities (low vs. moderate), on Italian ryegrass pasture establishment. In addition, we tested resilience of the system by testing pasture's ability to re-establish following a year without seed head production. The experiment consisted in the rotation, on the same area, of Italian ryegrass pasture grazed by sheep during the winter and up to the end of the grass production cycle, and soybean or soybean-maize grain crops rotation cultivated during the summer. The pasture established itself by self-seeding since 2005. Data were collected in 2011 and 2012 stocking season. The soybean summer crop, continuous stocking and low grazing intensity, all positively affected the production of reproductive tillers in 2011. Grazing intensity in 2011 strongly influenced early vegetative tiller densities (before crop harvest) in 2012. However, none of the grazing intensity or the stocking method treatments affected herbage mass at the end of pasture establishment in 2011 or 2012. On the other hand, the soybean summer crop positively affected pasture establishment, both in term of tiller densities and herbage mass at the end of pasture establishment. The removal of all seed heads in 2011 (preventing seed production) resulted in the total failure of pasture establishment in 2012. Overall, Italian ryegrass establishment by self-seeding relies on the annual replacement of the soil seed bank. This experiment demonstrated that under various stocking methods, moderate grazing intensity and maize or soybean summer crop, Italian ryegrass pasture establishment by self-seeding remains successful even when the stocking periods extended up to the end of the grass production cycle. Self-seeding with moderate grazing intensity ensures successful pasture establishment, reduces labour and costs and allows to increase the stocking period and so animal live weight gain over the grazing season.     

Persistence of Legumes in Hill Land Pasture Following Sod Seeding

Unimproved hill land pastures are often characterized by low production and poor forage quality. The introduction and persistence of a legume component in the sward may increase both forage quality and quantity. Persistence of red clover ( Trifolium pratense L.), birdsfoot trefoil ( Lotus corniculatus L.) and white clover ( T. repens L.) sod-seeded into a rotationally grazed hill land pasture was examined. Legumes were introduced by broadcasting by hand, sod-seeding with a zip seeder, and sod-seeding with a Power-till seeder. The effects of paraquat (1, 1-dimethyl 4–4'-bipyridinium ion), lime and P on legume persistence also were examined. The pasture was mostly Kentucky bluegrass ( Poa pratensis L.) with 5 % white clover initially. Percentage legume in the pasture followed the same trend over time irrespective of whether sod-seeded or not, or method of sod-seeding. There was 30 % legume in control plots the year after sod-seeding and 15 % at the end of 4 years. Application of lime and P resulted in an increase in legume content of at least 5 % through the 4 year period. Paraquat, legume species, and sod-seeding method had no effect on legume persistence. It is concluded that management, soil fertility and time are the most important determinants of legume persistence in hill land pastures.     

Biological nitrogen fixation in a grazed perennial grass/clover ley and correlation with herbage and soil variables

In a grazed grass/clover pasture, biological nitrogen fixation (BNF) is a source of nitrogen (N). The determination of this N input and how it is influenced by N, phosphorus (P) and potassium (K) in the soil, is complicated by grazing animals, which cause variation in defoliation and soil mineral content in both space and time. In a perennial grass/clover ley on a commercial Danish organic farm, exclusively grazed by dairy cows, we investigated BNF, clover proportion of the herbage, accumulated dry matter of clover and N, P and K concentrations in separates of clover and grass in relation to inorganic concentrations of N, P and K in the soil. The ¹⁵N natural abundance technique was used to determine the proportion of clover N derived from the atmosphere (pNdfa) using grass as a non-fixing reference. In September 1997, soil and biomass sampling was carried out on 81 randomly selected plots. The value of pNdfa was on average 0.6, ranging from 0.12 to 0.96 and BNF exhibited a significant but weak correlation (r=0.432) with accumulation of dry matter in clover. For the significant correlations, r varied between −0.403 and 0.648. The proportion of clover in the sampled herbage averaged 48%, ranging between 18 and 78%, but varied independently of BNF. Due to a high concentration of P and moderate to high concentration of K in the soil, clover and grass were well supplied with P and K, which was confirmed by the concentrations of K and P in shoot material.     

Genetic variability and heritability estimates for hardseededness and flowering in balansa clover (Trifolium michelianum Savi) populations

Suitability of annual pasture legume species like balansa clover (Trifolium michelianum Savi) in southern Australian farming systems depends on their hardseededness and time to flowering. Late maturing varieties with increased rate of hardseed breakdown, particularly in the late summer—early autumn period are desirable to ensure reliable regeneration in permanent pastures. Two half-sib family balansa clover populations were used to estimate heritability for the rate of breakdown of hardseededness and time to flowering. High narrow-sense heritability estimates (90.5–96.1%, rate of hardseed breakdown; 85.9–94.5%, time to flowering) were obtained, and were mainly attributed to additive gene effects. There was no relationship between rate of hardseed breakdown and time to flowering in Group1 half-sib family, which indicated that selection for both traits could be undertaken independently. In half-sib family Group 2 the relationship between the two traits suggested the possibility of selecting late maturing cultivars with increased rate of hardseed breakdown. Significant differences in time to flowering and rate of hardseed breakdown both within and between the two half-sib family groups suggested that phenotypic recurrent selection would be effective in improving the traits in these populations.     

Harvest interval affects lucerne (Medicago sativa L.) taproot total yield,starch, nitrogen and water-soluble carbohydrates

Lucerne (Medicago sativa L.) has a large taproot to store and release starch, carbohydrates and nutrients during the plant's growth. Recommended management of the lucerne crop aims to keep the taproot stable subject to the demands for feed provision and stand longevity. Field experiments were conducted in Victoria, Australia, to examine the effects of recovery period on taproot mass and nutritive status. Both experiments used established SARDI Seven lucerne crops and were either cut every 21 days (short recovery SR) or every 42 days (long recovery LR). At each defoliation, taproots were extracted for determination of DM yield and starch, water-soluble carbohydrate (WSC) and nitrogen (N) concentration and DM yield. At both sites, WSC and N DM yields along with total taproot DM yield did not change greatly as the experiment progressed. Starch was responsive to herbage accumulation with both concentration and DM yield, increasing when herbage accumulation rates were high and decreasing when herbage accumulation rates were low. At both sites, LR taproots had starch levels equal to or higher than those of SR taproots. We conclude that short intervals between defoliations disrupt the energy cycling between shoots and roots. This is likely to reduce lucerne productivity, particularly during periods of rapid growth.     

Accumulation of Secondary Metabolites Formed by Field Fungi in Autumn-Saved Herbage

Taking into consideration the fact that in recent years weather conditions were sufficient to feed cattle on pasture until late autumn or early winter, an experiment was carried out during the years 2000 and 2003 on a low input pasture to show the effect of the pre‐utilization date (June, July, August) and the harvest date in the autumn/winter period (November, December, January) on the accumulation of the following secondary metabolites formed by fungi: ergosterol (ERG), ochratoxin A (OA) and zearalenone (ZEA) in the autumn‐saved herbage. The highest levels of concentration of these metabolites were 470.3, 1.63 and 34.9 ng g^?1 respectively. Lower ERG levels were influenced by later pre‐utilization, which is a logical consequence of the shorter period of time for field toxigenic fungi development. Pre‐utilization in August is strongly recommended as the concentration of ochratoxin A (OA) in the autumn‐saved herbage in January was the lowest (not much changed in comparison with earlier harvest). Later harvest was also affected by higher ZEA concentrations in the analysed samples, which indicates that, under conditions of prolonged pasture utilization, field toxigenic fungi (producing ZEA) increased their population and had proper weather conditions stimulating the biosynthesis of the toxin. The later suggests that the final harvest of autumn‐saved herbage should be made rather in December than in January. Higher levels of humidity influenced higher concentrations of ERG and ZEA in the autumn‐saved herbage.     

Co-adaptation between neighbours? A case study with Lolium perenne genotypes

Summary Co-adaptation between neighbouring genotypes of Lolium perenne was examined in a pot experiment over an 18 month period. Competitive interactions in pure and mixed stands were examined using 9 genotype pairs (established as neighbours from electrophoresis data), 7 from a 10 year-old, and 2 from a 40+ year-old pasture. Combinations of each pair of neighbours grown together, and each grown with 5 randomly chosen alien (non-neighbour) genotypes sampled from the same pastures were also assessed. Seven harvests of green matter produced above 20 mm were made, and 50–135 kg N ha^-1 was administered after each harvest. Data for dry weight at each harvest for pure and mixed stands, and Relative Yield Totals (RYT) were obtained for neighbours and aliens in each treatment.In 6 of the 9 pairs of neighbours grown as pure stands, the yields of neighbours did not differ significantly at the final harvest. In 8 of the 9 pairs grown as mixtures of pairs of neighbours, the yield of one of the neighbours was significantly reduced and one neighbour had been eliminated. Over the last 5 harvests, yields from (alien + neighbour) mixtures had the greatest dry matter yields. No suppression of one component by the other was recorded for this (alien + neighbour) combination.Based upon RYT estimates, neighbours did not yield more when grown together than when grown separately with aliens. There was no relationship between original genotype size in their fields of origin and dry matter yield in this experiment.This experiment provides no evidence for he evolution of co-adaptation between surviving neighbours in the two pastures from which these individuals were sampled. The genotypic architecture of pasture would seem to be the product of random and/or competitive exclusion of individual genotypes, rather than the evolution of co-adapted groups of individuals through time. The consequences of extreme depletion of genetic variation in the first two years after establishment, the very low rates of seedling establishment in established pastures, and the impact of the grazing animal would suggest that there is little likelihood of or potential for, co-adaptation evolving between neighbours.     

Agronomic Factors in Evaluating Forage Crops I. Predicting Quality Measures of Crude Protein and Digestibility from Crop Leafiness

Stage-to-stage changes in leaf percentage, crude protein concentration and in vitro true disgestibility of alfalfa herbage materials of the same age and common harvest dates across there different growing season were determined. Leafiness decreased with advancing crop maturity. Herbage leaf content was closely associated with the total herbage C P and IVTD. When leaf percentage was regressed on quality measures of C P and IVTD, the regression analysis provided highly significant linear relationship for CP(r²= 0.823), IVTD (r²= 0.854) for alfalfa crop diverse in stand and canopy ages, stages of maturity and across locations and seasons of the year. These relationships indicated practical applicability of the use of crop leafiness for predicting the herbage quality.     

不同秋眠类型苜蓿品种的生产性能研究

通过对国内外42个秋眠级为2~9级的紫花苜蓿品种4个生产年的研究, 探讨苜蓿秋眠级与生产性能的关系。结果表明: (1)不同秋眠级苜蓿的干物质产量均呈第4年≈第3年＞第1年＞第2年的趋势, 其中第2生长年(2003年)天气温度较低和降雨过量是其低产的直接原因; (2)在郑州地区, 4、2秋眠级苜蓿干物质产量显著高于非秋眠品种, 其他秋眠或半秋眠苜蓿产量也程度不同地高于非秋眠苜蓿; (3)不同秋眠级的苜蓿品种第1茬产量最高, 其次为第2、3、4茬, 第5茬产量最低; (4)苜蓿的秋眠级与年干物质产量、粗蛋白质(CP)含量呈负相关, 与粗纤维(CF)呈正相关, 认为郑州地区不宜选用非秋眠品种。     

安塞黄土丘陵区人工草地生产力与土壤水分特征

摘 要 人工草地建设是半干旱黄土丘陵区生态建设与农业生产的重要组成部分,比较不同类型人工草地生产力、土壤水分差异,可为草地建设中草地类型的选择与管理提供依据。通过比较川地与山地不同类型草地生产力、土壤水分年季变化差异,表现为以下特征：川地人工草地0-100cm土壤含水量受季节降雨量影响变化较为活跃,为活跃层。100-350cm土壤含水量与根系利用有关,为次活跃层。350-500cm季节变化不大,为相对稳定层。柳枝稷与苜蓿草地350-500cm土层干燥化明显。在2004-2005年,各草地4月到9月0-200cm土壤储水量减少,9月到11月0-200cm储水量增加,200cm以下各草地间差异明显,但季节变化量接近。山坡不同类型草地0-100cm土壤含水量表现为秋末高于春季,100-500cm不同季节均未显示明显的水分亏缺现象。     

Functional contribution of two perennial grasses to enhance pasture production and drought resistance under a leaf regrowth stage defoliation criterion

Lolium perenne L. (Lp) is the main pasture species in New Zealand, but climate change increases the likelihood of drought during summer and consequently lowers its growth rates. Bromus valdivianus Phil. (Bv) tolerates better the summer soil water restriction, but its competitiveness in a diverse pasture relates to the defoliation management and soil moisture levels. The performance of both species in terms of production, persistence and feed quality across seasons under different defoliation regimens was evaluated when the species were mixed and in monoculture. The treatments were applied in a randomised complete block design with three blocks. Bromus valdivianus and Lp monocultures and the 50/50 mixture (Mx) were grazed at low and high defoliation frequencies (i.e. based on Bv and Lp optimal leaf regrowth stage [LS], respectively). Herbage mass production was 15% higher in the Mx, mainly due to an increase in production in spring and summer. In spring, there was complementarity growth between the species, while during summer/early autumn, the Mx production relayed in the higher participation of Bv. Root biomass at depth (31–70 cm) was 38% greater for Bv monoculture and the Mx compared with Lp monoculture, which demonstrated the contribution of Bv to a better production of the mixture during and following the dry periods. In addition, the maintenance of the Bv tiller population and the higher growth rate during dry periods demonstrated its better fitness to low soil water availabilities than Lp. Although herbage mass was not modified by the defoliation frequency, forage quality of Bv showed values at low defoliation frequency that could compromise animal production during spring. However, there was an increase in root biomass under low defoliation frequencies, which suggested that a flexible grazing management across seasons would increase the niche utilisation between species in the mixture. The defoliation frequency of the mixture could be based on Lp LS during spring and winter, and slowed down, following Bv criteria, during summer and autumn.     

Effects of elevated CO2 and cutting frequency on the productivity and herbage quality of a semi-natural grassland

Monoliths of a fertile, although N limited, C₃ grassland community were subjected (or not) to an atmospheric CO₂ enrichment (600 μmol mol⁻¹), owing to the Mini-FACE system from August 1998 to June 2001, at two contrasting cutting frequencies (3 and 6 cuts per year). The present study reports the effects of elevated CO₂ on the above-ground productivity and on the herbage quality. Elevated CO₂ did not affect the dry matter (DM) yield of the swards in 1999. In 2000, the second year, there was a positive CO₂ effect (+26%) both on the DM and on the nitrogen yields (+30%). With the frequently cut monoliths, the DM of the legume component of the sward was strongly increased by elevated CO₂. This effect became also significant in July 2000 for the low cutting frequency treatment. These results are in good agreement with the concept of an increased legume development and symbiotic N₂ fixation triggered by an increased ecosystem scale demand of N under elevated CO₂. At a low cutting frequency, the DM of the forbs was strongly increased in elevated compared with ambient CO₂. This increased development of the forbs apparently led to a competitive decline of the grasses. Therefore, the total DM yield response to CO₂ was smaller at a low (+15%) compared with a high (+36%) cutting frequency in 2000. An increase in the water soluble sugar content of the bulk forage under elevated CO₂ and a corresponding decline in cell wall contents (NDF) were observed. In June 1999, the decline in NDF was correlated with an increased in-vitro DM digestibility. The forage quality was also indirectly affected by elevated CO₂ through changes in leaf:stem ratio and in botanical composition. At a low cutting frequency, the increased forb content favoured the herbage quality because of a higher digestibility of the forb shoots and, indirectly, through the reduction in the mass of the grass stems. These results emphasise the role of species dynamics for elevated CO₂ impacts on semi-natural grassland productivity and herbage quality.     

基于气象适宜度指数的若尔盖湿地天然牧草产量估算模型

气象条件极大地制约着若尔盖湿地天然牧草的生长状况,为了从气候条件适宜角度评估牧草长势年景的优劣,利用若尔盖湿地天然草场牧草1983—2013年近31年(2002年缺测)牧草产量观测资料,得到能够反映光、温、水对牧草生长影响程度的气象适宜度指数,分析了若尔盖湿地天然牧草产量与气象因子之间的关系,分别建立了5月、6月和7月牧草产量气象适宜度线性和生长曲线估算模型。结果表明:所建立的牧草产量气象适宜度估算模型能够较好地模拟牧草产量的演变趋势;各月气象适宜度产量线性模型的精度普遍高于生长曲线模型;与干重模型相比,鲜重模型精度较差。受牧草产量观测资料精度的限制,6、7月牧草产量估算模型还需要进一步改进。