Journal of New Zealand Grasslands

47 years of plant breeding

Alan Stewart — Sun, 25 Oct 2026 00:00:00 +0000

It is an honour to be recognised by my peers for this prestigious award and to have this opportunity to give the Levy Oration. I have now completed 47 years of my plant breeding career in pastures. I was among the first wave of breeders employed in New Zealand after the Plant Varieties Rights Act was passed into law 1973. I was employed in Pyne
Gould Guiness (PGG), a small regional company but fortunately with the advice of Dr Corkill a former breeder and Director of Grasslands DSIR. Lionel Corkill was the first forage breeder employed by DSIR Grasslands in 1931 and breeder of many widely used cultivars.

The Pyramid: a typical Marlborough family farm revisited

Sun, 25 Oct 2026 00:00:00 +0000

The Pyramid, a family farm in Marlborough, illustrates how long-term adaptation can sustain pastoral production while responding to climate, environmental and market pressures. This paper revisits the farm 18 years after an earlier review, using farmer narratives, farm records, reports and greenhouse gas modelling to examine changes in production, tree planting and diversification. Since 2008, the farm has intensified and diversified through improved sheep performance,
flexible cattle policies, indoor lambing, expanded forestry and native planting, and development of a vineyard enterprise. Production increased from 148.9 to 216.6 kg carcass weight per hectare by 2020, lambing percentage rose from 119% to 158%, and profitability improved from $306 to $457 EBITDA per hectare. Although biological greenhouse gas emissions increased with higher output, emission intensity declined by 12%, and forestry sequestration was estimated to exceed whole-farm emissions. Vineyard development added a complementary land use, supported succession, and retained integration with livestock grazing. The findings show that resilience at The Pyramid has come from matching land use to
land capability, investing steadily in infrastructure and soil fertility, and maintaining enterprise flexibility. This case highlights practical pathways for combining productivity, environmental stewardship and family business continuity in dryland hill country farming. Keywords: dryland resilience, farm

Dry matter yield, botanical composition and nutritive value of diverse and standard pastures under contemporary and regenerative dairy management systems

Sun, 25 Oct 2026 00:00:00 +0000

This study investigated the potential of diverse pastures within regenerative dairy systems that aim to emulate the structure and function of grassland ecosystems through increased species diversity and adaptive grazing dynamics, compared with contemporary temperate dairy systems. Over three consecutive production years, three farmlet systems were compared: standard contemporary, diverse contemporary, and diverse regenerative. Annual pasture production differed significantly (P < 0.01), with diverse regenerative systems achieving the highest mean yield [dry mass] (~11 t). Seasonal growth was broadly consistent across systems, with the highest growth rates in spring (53-55 kg DM/ha/day) and minimal growth in winter. Diverse pastures in both systems demonstrated enhanced yield and greater resilience under variable climatic conditions. Botanical composition analysis indicated that diverse pastures maintained more balanced proportions relative to simpler pastures. However, the proportion of legumes and herbs declined over time across all systems. Nutritive values were largely comparable among systems, although diverse pasture systems
exhibited significantly lower fibre concentration and elevated mineral content. Overall, diverse pasture systems, particularly under regenerative management, enhanced both pasture productivity and mineral composition without compromising feed quality over a three-year period. These findings highlight the potential of diversified pastures and altered management to improve the resilience and sustainability of low nitrogen pasture-based dairy systems.

Graze less, grow more? Leaf stage defoliation management of diverse and ryegrass-based pastures in Golden Bay, New Zealand

Lucella Jordan — Sun, 25 Oct 2026 00:00:00 +0000

With climate change, and rising fertiliser costs, lowcost pasture growth strategies are needed for New Zealand farmers. This study compared functionally diverse pastures, (six grasses, four legumes, two herbs) with standard perennial ryegrass-white clover under high-rainfall, moderate fertility conditions in Golden Bay. Over three years, a 2×2 factorial study applied two defoliation intervals across two pasture types: cut when the standard pasture reached the 2.5-leaf stage (2.5LS) and when the diverse pastures reached the 3.0-leaf stage (3.0LS). In 2024, diverse pastures matched standard
pasture in dry matter produced. Yield differences between treatments occurred seasonally in 2024, with standard pasture cut at the 2.5LS growing 12% more biomass than diverse pastures during autumn. In spring, both pasture types cut at the 3LS produced 20% more than pastures cut at 2.5LS. In summer 2024–25, defoliating at 3LS defoliation grew 9% more than
the 2.5LS. Sown species declined from 85% (year 1) of total dry matter to 50% by year three. Seasonal rainfall was average, though late autumn sowing may have impacted subsequent persistence. These findings show that defoliation interval has a greater impact on increasing seasonal yield than the pasture mixtures compared. However, well-managed diverse pastures can match or in some seasons exceed ryegrass-white clover systems.

Persistence of perennial ryegrass, tall fescue and cocksfoot following repeated annual sowings: influence of pasture age, grass species and cultivar in Canterbury

Thomas Maxwell, Emma Gordon, Kate Crawford, Belaraine Jordan, Gerald Cosgrove — Sun, 25 Oct 2026 00:00:00 +0000

Persistence is an important determinant of sown perennial pasture-grass productivity. Defining traits that affect persistence is essential for improving pasture longevity through plant breeding and identifying criteria for inclusion in cultivar ranking indices (Forage Value Index). A repeated sowings experiment commenced in 2015 under continuous sheep grazing in Canterbury. Eight cultivars of perennial ryegrass representing different ploidy, heading date, and decade of cultivar release, and one cultivar each of tall fescue and cocksfoot were sown in autumn in each year for 10 years. This
paper compares persistence trait data from a single sampling point (autumn 2025). Pasture age influenced production and composition more than grass cultivar. Peak yields occurred in 4- and 8-year-old pastures (~3430 kg DM/ha) and the lowest in 10-year old pasture (2494 kg DM/ha). Sown grass proportion dropped from 65% at establishment to around 50% in older stands. Persistence was strongly species dependent. Cocksfoot (cv. Savvy) showed the strongest persistence traits: highest sown grass proportion (61%), lowest unsown species (9%) and highest persistence score (8.2). Perennial ryegrass cultivars generally showed moderate and similar persistence, while tall fescue (cv. Hummer) performed poorly (27% sown grass, lowest tiller density and low persistence score). Persistence is an outcome driven by species morphology, establishment success, compatibility with regional soil and climatic conditions and regime of grazing management. This study reinforces the suitability of cocksfoot for long-term persistence on stony Canterbury soils, demonstrates autumn-sowing and relatively hard continuous grazing is detrimental to tall fescue persistence, while perennial ryegrass cultivars generally showed moderate and similar persistence.

Sheep liveweight and dry matter production from Years 3 and 4 of the Regenerative Agriculture Dryland Experiment

Luke Robb, Obaka Ikani, Alistair Black, Anna Mills, Derrick Moot — Sun, 25 Oct 2026 00:00:00 +0000

In Year 4 of Lincoln University’s Regenerative Agriculture Dryland Experiment, the Regenerative sheep production system of multispecies forages and short-duration, high-density rotational grazing produced 130 kg/ha less liveweight than the Conventional system of simpler forages and longer-duration, lower-density grazing. Both systems produced the same amount of total dry matter (9.93 t/ha), but the Regenerative system produced less legume dry matter (3.43 versus 5.09 t/ha), crude protein (1.85 versus 2.24 t/ha) and metabolisable energy (111 versus 119 GJ/ha). The two levels of soil Olsen P (10–15 and 20–25 mg/kg) had no influence on the farm system effect. These results demonstrate that the greater legume content for the Conventional system was the most important factor to influence liveweight gain in this dryland environment.

Changes in forage yield and botanical composition in a species diversity by harvest height by frequency experiment over 3 years in Hawke’s Bay

Beverley Thomson, Noel Smith, Kay Ward, Paul Muir — Sun, 25 Oct 2026 00:00:00 +0000

There has been considerable interest in regenerative agriculture and the potential impact of diverse forage mixtures and grazing to higher residuals on pasture production and soil quality. This study evaluated the species mixtures at three defoliation heights and two defoliation frequencies. The experiment was carried out over 3 years at a summer-dry site in Hawke’s Bay. The treatments were: mixtures of six cultivars of four species (ryegrass/clover) and 18 cultivars of 16 species (diverse), defoliation heights of 5, 10 and 15 cm to replicate normal, lax and very lax grazing, and defoliation frequencies of 4 and 8 weeks, over the 3 years. Total dry matter production was 23,910 kg/ha for the diverse mixture compared with 19,060 kg/ha for ryegrass/clover and decreased from 29,410 kg/ha for the 5 cm defoliation height to 21,200 kg/ha for 10 cm and 13,840 kg/ha for 15 cm, with no effect of defoliation frequency. As defoliation height increased, so did the amount of dead matter, and thus feed quality would have declined. The greater dry matter production in the diverse mixture was likely to be at least partially due to the presence of chicory and plantain.

Simulating farm system implications of incorporating plantain (Plantago lanceolata L.) into dairy pastures across four regions

Lydia Farrell, Pierre Beukes, Mike Dodd — Sun, 25 Oct 2026 00:00:00 +0000

Understanding the farm system impacts of narrowleafed plantain (Plantago lanceolata L.) is important for supporting farmer adoption to reduce nitrogen (N) leaching from pastoral farming. This modelling used observed herbage accumulation rates and nutritional composition data from pastures across New Zealand comprised of perennial ryegrass
(Lolium perenne L.) and white clover (Trifolium repens L.) without plantain or with plantain at 11% to 32% (averaging 25%) of sward DM. Regional dairy farm systems were simulated using the Whole Farm Model and Urine Patch Framework to investigate the implications of plantain mixed pastures for pasture yield, milk production, profitability, N leaching,
and greenhouse gas emissions. Inclusion of plantain did not impact milk production and maintained or slightly increased pasture yield. Operating profit averaged $11/ha (1%) lower with plantain due o changes in the seasonal pasture supply curve, requiring additional conservation of surplus pasture and additional supplementary feeding during feed deficits. Inclusion of plantain reduced N leaching by 11% to 19%. The reductions in N leaching without significant negative trade-offs make plantain an attractive option for wide-scale adoption. Farmers should consider the potential for changes in seasonal pasture supply and the need for regular re-sowing of plantain due to low persistence in most perennial ryegrass-white clover swards.

Sowing strategies to increase the plantain (Plantago lanceolata L.) content of Manawatu dairy pastures

Sun, 25 Oct 2026 00:00:00 +0000

pastures to reduce nitrogen loss, but poor establishment and persistence are commonplace. Four sowing strategies, involving the substitution of ryegrass with increased plantain sowing rates and repeated re-seeding of plantain and other species, were compared on No. 4 Dairy Farm at Massey University. Treatments were: 20, 15, 10, or 5 kg ryegrass with 3 kg white clover and 0, 4, 7, or 10 kg plantain/ha in autumn 2019 (Nil–High), re-seeded with 3, 6, or 9 kg plantain/ha sown in autumn 2020 (Low–High), 3 kg ryegrass, 3 kg white clover and 10 kg plantain/ha after grass herbicide in autumn 2021 (High), 4, 6, or 8 kg plantain/ha in spring 2021 (Low–High) and 4 or 6 kg plantain/ha (Low–Medium) or 12 kg Italian ryegrass/ha (High) in autumn 2023. On average, pasture growth and weed content were not different across treatments, but plantain content increased from 4% to 24%, 32% and 38% for Nil to High, which maintained the highest plantain content over 5 years. These results show that the substitution of ryegrass with increased plantain sowing rates, and
repeated re-seeding of plantain can increase the plantain content of dairy pastures.

Accumulated yield of clovers grown in a summer dry environment after three years

Sonya Olykan, Richard Lucas, Derrick Moot — Sun, 25 Oct 2026 00:00:00 +0000

A 2019 experiment quantified the productivity of clover monocultures in a summer dry environment. ‘Taipan’
and ‘Viper’ balansa clovers were the most productive in their first early spring (see Olykan et al. 2021). They set seed at the end of Year 1 but there was poor establishment in Year 2. The unexpected reestablishment of high balansa populations in Year 3 prompted the experiment’s continuation to study balansa growth alongside ‘Woogenellup’ sub and ‘Relish’ red clovers that were still persistent and productive. Over three years, the total yield during the annual clover growing season (~late winter to early summer) was ~20.0 t DM/ha for sub and red clover, which was higher than the ~13.3 t for the balansa cultivars due to its re-establishment failure. In Year 3, the final accumulated yield from the best plot areas of all clovers, including balansa, was ~3.6 t DM/ha. Supported by sufficient early summer rainfall (i.e., 50+ mm), red clover provided an additional summer yield of 1.7 t in Year 2 and 7.2 t DM/ha in Year 3. The results confirmed that the growth rate and yield patterns of sub clover (late winter to late spring) and red clover (Late spring to mid-summer) are complementary. The possible reasons for the failure of balansa clover to re-establish in Year 2 relate to hot and dry conditions in 2020. Under these conditions, management options for balansa may require the inclusion of other pasture
species.

Temperature requirements for the germination of alternative forage species for northern New Zealand climates and a prominent weed

Zachary Ngow, Deborah L. Hackell, Christopher E. Buddenhagen — Sun, 25 Oct 2026 00:00:00 +0000

Ryegrass-clover pastures are increasingly facing challenges with reduced persistence and weed ingress, especially in northern New Zealand dairying regions. Seed germination is the first, and arguably most important step, to determine the successful establishment of pasture (or weed) species. Soil temperature and soil moisture are important factors for germination and therefore determining species specific preferences for these parameters can indicate the environmental conditions facultative to their establishment. This study aimed to determine the cardinal germination temperatures of four forage cultivars: Haven tall fescue (Lolium arundinaceum(Schreb.) Darbysh.), Oakdon meadow fescue (Lolium pratense (Huds.) Darbysh.), berseem clover (Trifolium alexandrinum L.) and Fulkerson kikuyu (Cenchrus clandestinus (Hochst. ex Chiov.) Morrone) and one prominent weed: Madagascar ragwort (Senecio madagascariensis Poir.) relevant to northern pastures. Seeds were germinated across a range of temperatures for 28 days and cardinal temperatures were determined
using broken-stick models. Cardinal temperatures and thermal time requirements are reported in the text. Kikuyu has a high base temperature for germination, which suggests a slow germination rate in spring compared to temperate forages, limiting its use in regions below the northern North Island. Madagascar ragwort has similar temperature requirements to ryegrasses and fescues. Further study of the interaction between soil temperature and moisture (hydrothermal constraints) would increase understanding of the germination behaviour of these species.

The effect of perennial ryegrass sowing rate and companion grass on the establishment of plantain in mixed pastures

Sarah McKenzie, Luke Visser, Kylie Dobbs, Jessica Dunbar, Mike Dodd — Sun, 25 Oct 2026 00:00:00 +0000

Managing the content of plantain in grass-based swards has become an important consideration for farmers intending to use this species to support seasonal forage supply and nitrogen loss mitigation. Changes in companion perennial ryegrass sowing rate and companion grass species were investigated to understand their impact on initial sward yield and the success of plantain establishment in grass-based pastures. In 2021, an agronomic experiment involving
six large-plot trial sites was established in Northland, the Waikato, Manawatū, Canterbury and Southland. This paper reports the sward yield and botanical composition results of the establishment phase, defined as the first two harvests at each site after sowing. Adjusting the diploid perennial ryegrass sowing rate from 8 to 20 kg/ha increased establishment phase yield but reduced mean plantain content across sites from 25% down to 20%. Fast establishing grass species (Italian and Hybrid ryegrass), which achieved the highest early yields, also had the lowest plantain botanical composition at the second harvest (mean across sites 8-13% compared to 21-23% for perennial grasses). Regional and soil type
differences in the success of plantain establishment were evident, with the best results on the Waikato Allophanic soil and the poorest on the Canterbury Pallic soil. Competition during the establishment phase, whether from different grass species or varying sowing rates, can affect plantain percentage. Farmers and advisors need to consider this when establishing multispecies pastures.

Feed conversion efficiency of regenerative compared with conventional dryland sheep production systems

Obaka Ikani, Luke Robb, Breanna Holt, Jim Gibbs, Alistair Black, Derrick Moot — Sun, 25 Oct 2026 00:00:00 +0000

The feed conversion efficiency of a regenerative dryland sheep production system was compared with a conventional system at two soil Olsen P levels (10–15 and 20–25 mg/kg) in Years 3 (2023–24) and 4 (2024–25) of an on-going experiment at Lincoln University. There were four 1.936-ha farmlets, each stocked with mixed-age Bohepe ewes and their progeny. The two systems were not different in dry matter (DM) intake (average 7.30 t/ha per year) and there was no effect of
soil P fertility. However, the feed conversion efficiency for the Regenerative system was lower by 29 kg of liveweight gain/t of DM intake in Year 3 and by 20 kg of liveweight gain/t of DM intake in Year 4. In Year 3, the DM intake for the Regenerative system was composed of 0.24 t/ha less crude protein (CP), 0.76 t/ha more neutral detergent fibre (NDF) and a 5.3:1 lower CP-to-metabolisable energy (ME) ratio. In Year 4, it comprised 1.31 t/ha less legume, 0.26 t/ha less CP and 2.6:1 less CP:ME. These results show that the Regenerative system produced less sheep liveweight gain from a similar quantity of lower quality feed consumed than the Conventional system.

Producing more without getting more productive

Mubashir Qasim, Mario Fernandez , David Silva-Villacorta, Ben Marmont — Sun, 25 Oct 2026 00:00:00 +0000

In the 2023/24 season New Zealand dairy farms are producing more milksolids per cow and per farm than ever before, but sector total factor productivity has been broadly flat since 2012/13.
• The difference reflects input intensity. Since 2012/13, additional output has increasingly been supported by increasing reliance on purchased feed, fertilisers and labour resulting in flat productivity trend.
• Flat productivity is better read as robustness through a difficult decade, but future competitiveness will depend on lifting pasture utilisation and input efficiency rather than simply spending more to produce more.

Pasture potential estimation for dairy farms: A hybrid modelled approach using integrated industry and spatial data

Josiah Cleland — Sun, 25 Oct 2026 00:00:00 +0000

Pasture eaten (t DM/ha) is known to be a primary driver of profitability in New Zealand pastoral dairy farm systems. However, locally relevant estimates of achievable home-grown feed eaten (the “pasture potential”) are difficult to derive at industry scale. This study used a multi-stage hybrid methodology to estimate farm-specific pasture potential across a large industry dataset. Each farm was spatially enriched with environmental covariates to assign it to a farm typology
system. Within each farm type, a local nitrogen response slope was estimated based on pasture eaten values for nearby farms within an expanding radius (5-120 km); local pasture eaten was then estimated as a nominal zero-N figure, to remove the confounding impact of fertiliser usage. The pasture potential value was defined as the 90th percentile of home-grown feed eaten within farms of the same type, estimated via quantile regression. A machine learning model (XGBoost) was then developed to impute values for sparsely populated typologies. The analysis identified an average distance from pasture potential of 2.5-3.5 t DM/ha across the different regions. These findings highlight the potential
benefit to farms having land-specific insights into their pasture potential and support the identification of pathways for improving farm performance in terms of profitability and emissions footprint.

Impact of diverse pastures and reduced nitrogen inputs on dairy pasture yields, botanical composition and soil health across three regions

Sun, 25 Oct 2026 00:00:00 +0000

Pastoral farmers can enhance productivity and minimise environmental impacts by utilising a range of management practices on farm. Different pasture management options being explored for their potential benefit to the farm and wider ecosystem include the use of diverse pasture species and targeted fertiliser nitrogen (N) inputs. On each of nine dairy farms across Waikato, Canterbury and Otago, one paddock was sown with a pasture mix containing perennial ryegrass/white clover/plantain (simple) and an adjacent paddock sown with an additional 12 species (diverse). Half of each paddock received either standard or low N fertiliser inputs. Herbage yields were determined for three years, with pasture species composition assessed seasonally. Soil health was assessed annually using measures of soil fertility, organic matter, physical condition and biological abundance. Total annual pasture yields were no different between the simple and diverse pastures. Pasture growth tended to be reduced in response to lower N fertiliser inputs, although results were not significant. The most consistent pasture responses to N were seen in Canterbury under irrigation. Under low N inputs there was a small increase in white clover (>1%), and although non-significant still created the potential for
increased biological N2 fixation. Plantain declined during the study from 20±2.7% to 2±0.8% in both pasture types. Grass contributed over half of the pasture sward. Although ryegrass was the dominant species in both pasture types, diversity remained higher within the diverse sward in the final year of study, with other grass species contributing up to 20±1.8% of the pasture sward. There was no effect of pasture type, N rate, or their interaction on any measured soil health property.

Do baleage wintering systems improve cow comfort compared with fodder beet wintering? Soil conditions and behavioural responses of non-lactating dairy cows in Southland, New Zealand

Dawn Dalley — Sun, 25 Oct 2026 00:00:00 +0000

Winter grazing systems in southern New Zealand can create wet, disturbed soil conditions that may restrict lying opportunities for non-lactating dairy cows. This observational study examined whether wintering practice influenced soil conditions and cow behaviour, with a focus on area allocation, baleage allocation frequency, and feed type (pasture baleage vs fodder beet). It was hypothesised that a larger area or reduced frequency of baleage allocation would
reduce soil damage and improve lying times. Six groups of approximately 73 pregnant cows each were managed on six paddocks over 30 days (12 Jun –11 Jul, 2024). Five groups received pasture and baleage, under differing baleage allocation frequencies (daily or 3-day) or area allocation (6.7, 7.5 and 8.6 m2/cow/day), while one group grazed fodder beet supplemented with baleage. Soil conditions (dry, wet, sodden) and behavioural measures (lying and rumination time)
were recorded. There were no significant differences between treatments in soil wetness or lying time, with cows averaging 9.4–10.3 h/day lying. However, cows grazing fodder beet had 26–30% lower rumination than all other treatments viz, 420 vs 572-600 min/day. These findings suggest that, under the conditions studied, baleage wintering did not provide superior lying conditions compared with fodder beet. Reduced rumination with fodder beet feeding may have
implications for animal health and subsequent milk production, warranting further investigation.

Evaluating pasture-based dairy cow wintering systems across soil and climatic conditions in southern New Zealand: implications for nutrition, welfare and environmental risk

Sun, 25 Oct 2026 00:00:00 +0000

Autumn-saved pasture offers a potentially attractive alternative for wintering livestock in the South Island, although variability in feed accumulation and uncertainty over suitability on poorly drained or structurally vulnerable soils constrain adoption. We evaluated pasture-based wintering using six case-study farms across Canterbury, Otago and Southland selected to represent a range of soil, landscape and rainfall conditions. Kale and fodder beet paddocks within a wintering system were also assessed on two farms. System performance was evaluated using proxy indicators including assessments of ground cover (drone-derived), soil condition, cow cleanliness, body condition score and diet adequacy. Overall, body condition score (BCS) outcomes from the pasture wintering case studies indicated adequate nutrition
despite variation in estimated ME supply. Where soils were suitable, pasture systems delivered improved cow cleanliness and reduced environmental risk compared with forage crops that were monitored on two of the farms. Successful outcomes were associated with permeable, structurally resilient soils and lower grazing pressures (≥15 m² cow-¹ day-¹) than typically applied to winter forage cropping systems.

Recovery of soil nitrogen by maize at various soil depths: a field study using 15N-labelled fertiliser nitrogen

Sun, 25 Oct 2026 00:00:00 +0000

Nitrogen (N) recovery by maize (Zea mays L.) in harvested grain is typically only 30–45% of applied fertiliser N, with the remainder susceptible to leaching and denitrification. In New Zealand, nitrate-loss regulation has increased reliance on nutrient budgeting models such as OverseerFM®, which until recently assumed a maximum effective maize rooting depth of 60 cm, potentially under-representing N uptake by deep-rooted crops, e.g., maize. Using 15N labelled fertiliser, maize uptake, timing and partitioning of absorbed N from soil depths of 60, 90, and 120 cm were assessed under New Zealand field conditions. A two year field experiment was conducted on a Waikato allophanic soil, with 15N applied at the V6 growth stage via subsurface injection beneath individual plants. Plant tissues were sampled through to maturity
to trace depth specific uptake and redistribution. Recovery of 15N in aboveground biomass mostly occurred post-silking, and was 65-69%. Results show that maize can access soil N from ≥120 cm depth, supporting improved representation of deep rooting and late-season N uptake in nutrient budgeting models.

Yield and phosphorus use efficiency of irrigated lucerne grown in a low Olsen P soil

Lauren Jones, Annamarie Mills, Derrick Moot — Sun, 25 Oct 2026 00:00:00 +0000

Phosphorus (P) is a key nutrient recommended for optimal nitrogen fixation by legumes in New Zealand pastoral systems. However, few field studies have examined the response of legume monocultures to soil P levels. This research aimed to quantify how phosphorus affects lucerne growth and development in an irrigated cut and carry system with a starting Olsen P of 9 mg/L. A randomized complete block design with five superphosphate treatments (0, 50, 100, 200, 400 kg/ha/y) and four replicates was established within a stand of ‘Kaituna’ lucerne. All herbage was cut-and-carried to ensure no nutrient transfer or return from livestock. Lucerne stem height and annual yields (12000–20000 kg DM/ha) were not different among treatments over four years even when the Olsen P in the Control declined to ~3 mg/L in the top 150 mm. Herbage phosphorus use efficiency was consistently highest in the 50 kg superphosphate/ha treatment with eight times the quantity of applied phosphorus extracted from the soil profile. The 400 kg superphosphate/ha treatment was the only P application rate that was sufficient to fully replace P removed in herbage. These findings suggest lucerne growth was not limited at Olsen P levels below current recommendations, which suggests lucerne was acquiring P below the measured depth and/or accessing P not measured by the Olsen P test.

Application of spent mushroom compost to a riverine silt loam at the time of pasture renewal and its effect on soil infiltration rate

Nicole L Schon, Seth Laurenson, Ray Moss — Sun, 25 Oct 2026 00:00:00 +0000

Pasture renewal is an important strategy for maintaining pasture production. While cultivation during pasture renewal can improve soil macropores in the short term, these benefits do not necessarily persist. Evidence from previous studies indicates that compost applied to agricultural soils can improve soil physical properties and increase earthworm numbers. We assessed whether adding mushroom compost at the time of pasture renewal would have further benefits on soil infiltration. We applied commercial mushroom compost to a riverine silt soil that had low soil infiltration and regularly ponded temporarily during irrigation events. The experimental treatments were assigned to replicated field plots (5m x 6m) and included: A) no pasture renewal, no cultivation, B) pasture renewal, no cultivation, C) pasture renewal, cultivated, D) pasture renewal, cultivated, 50t/ha compost applied, E) pasture renewal, cultivated, 100t/ha compost applied, and F)
pasture renewal, cultivated, 150t/ha compost applied. Soil infiltration rates measured 2 months after initiating
treatments indicated no difference between plots that were cultivated and amended with compost compared with those plots left under the original pasture (and without cultivation). However, for pastures that were renewed via direct drill (i.e., without cultivation), soil infiltration rates were significantly (P<0.05) lower than all other treatments and likely reflected surface soil damage that was caused by heavy rainfall soon after drilling new seed rather than the absence of compost.
After 14 months none of the management approaches for renewing pastures that we assessed were found to have a lasting effect on soil infiltration rate. Earthworm numbers and total earthworm biomass measured after 7 months were not significantly different across treatments across all compost rates evaluated. However, earthworm biomass tended to be lower in the cultivated soils compared with the non-cultivated treatments (P<0.10). This experiment indicates that a
single application of mushroom compost during the renewal of pasture had limited benefit in improving soil infiltration rates or increasing earthworm numbers within 14 months after pasture renewal.

Improving crop nitrogen uptake prediction in OverseerFM through iterative soil-crop coupling and recalibration

Sun, 25 Oct 2026 00:00:00 +0000

OverseerFM is widely used for farm nutrient budgeting in New Zealand. Earlier versions under-predicted crop
nitrogen (N) uptake for some crops under moderate to high soil N supply, leaving more residual N in the simulated soil profile and limiting the model’s ability to represent the environmental benefits of catch crops and other cropping systems. To address this, field data from 37 experiments across a range of crop types and regions were used to recalibrate crop parameters. Two updates were introduced: a luxury N uptake factor to represent additional uptake when N supply exceeded baseline requirements, and an iterative monthly coupling between crop uptake and soil mineral N supply. Together, these changes reduced under-prediction of above-ground N uptake and improved agreement with
observations across diverse crops, including reduced prediction error in an independent validation dataset. Soil-mineral-N prediction error also decreased in one rotation dataset, although substantial discrepancies remained. The revised model therefore provides a stronger basis for representing N capture by catch crops, while winter-period uptake and nitrate leaching require further validation.

A comparison of three contrasting farming systems on forage production and soil characteristics over three years

Beverley Thomson, Noel Smith, Kay Ward, Paul David Muir — Sun, 25 Oct 2026 00:00:00 +0000

There has been considerable interest in regenerative agriculture and the potential impact of diverse forage mixes and grazing to higher residuals on pasture production and soil quality. This study aimed to evaluate the forage productivity, botanical composition, and soil characteristics of three different forage mixtures managed as per best practise for each forage type over three years in a summer-dry environment. The three treatments were A: ryegrass/clover (4 species) rotationally grazed by lambs in winter/spring and yearling cattle in late spring/summer; B: a hyperdiverse mix (23 species) grazed by cattle with daily shifts; and C: plantain/clover/chicory (7 species) rotationally grazed by weaned lambs in winter, and ewes and lambs in spring. Over three years the hyperdiverse mix produced significantly less dry matter than the ryegrass and plantain mixtures (29246, 36709 and 32743 kg DM/ha respectively P<0.001). Thirty months
after sowing the number of species present had declined to nine, three and four in the hyper-diverse, ryegrass and plantain pastures respectively. There was no effect of farming system on soil carbon, earth worm number or organic matter disappearance. Water infiltration rate was significantly slower on the hyper-diverse treatment, potentially due to the effect of cattle treading. Three years after sowing, the use of a hyper-diverse forage system provided no advantage for forage production or soil health over sowing a plantain or ryegrass mix.

Nitrous oxide emissions – can they be abated by forages in ruminant agricultural systems?

John Caradus, Cecile De Klein — Sun, 25 Oct 2026 00:00:00 +0000

Nitrous oxide (N2O) emissions account for approximately 10% of global greenhouse gas emissions, with about 90% of these emissions derived from agricultural practices. In New Zealand, the single largest source (60%) of agricultural N2O is animal excreta (urine and dung) deposited during grazing while nitrogen fertiliser use currently directly contributes only 2.7% of agricultural emissions. Currently available options to reduce N2O emissions (e.g., dietary interventions or using enhanced fertilisers), are more suited to more-intensive animal production while fewer options are available for more extensive grazing systems. The aim here is to examine possible forage solutions that might reduce N2O emissions from grazed pasture systems. Plant-based options that could lead to reduced N2O emissions include using plants with
chemistry that divert excess nitrogen into faeces rather than urine and/or by influencing soil N transformation
processes. Faeces degrade more slowly than urine and as a result the slow mineralisation of organic N results through a slow supply of ammonium results in a reduced rate of N2O emissions. The challenge will be to improve the agronomic characteristics of these plants which currently limits their usefulness. This could be approached through plant breeding and/or use of forb-based pastures. This thought piece on how New Zealand grassland farmers can reduce N2O emissions
without having to disrupt current management systems will seek to identify practical opportunities.

National and regional economic impacts of climate-driven pasture decline

Ryan Mills, Ben Marmont, Mark Neal — Sun, 25 Oct 2026 00:00:00 +0000

Declining pasture and crop harvests is increasingly evident across several New Zealand regions. Milk production per cow has continued to increase over the past two decades, driven largely through increased use of supplementary feed. This may mask the true effect of declining pasture and crop harvested, which remains fundamental to profitable dairy production. Pasture is the cheapest form of feed for a dairy farm, and relying on other feeds will erode New Zealand’s
competitive cost advantage. This study quantifies the national and regional economic costs of declining pasture and crop harvested and estimates the potential value of strategic investment in forage and farmsystem solutions. Using the Dairy Sector Pathways model, parameterised with DairyBase and Dairy Statistics data, four scenarios were simulated to 2064/65: a baseline holding pasture and crop harvested constant; a business‑as‑usual scenario with declining pasture and
crop harvested; a strategic mitigation scenario slowing the decline; and an improvement scenario increasing pasture and crop harvested. Outcomes were assessed using discounted net present values of operating profit for dairy farms. Nationally, pasture decline imposes an estimated net present value cost of $10.5 billion to the 2064/65 season, relative to a Baseline net present value of $98.6 billion. Strategic investment generates positive economic returns relative to business-as-usual, valued at $0.5 billion under mitigation and $2.6 billion under improvement. Benefits depended strongly on the timing and extent of technology uptake, highlighting the importance of early and widespread investment to
sustain pasturebased dairy systems.

Towards a more accurate assessment of the vegetation carbon stocks in New Zealand grasslands

Sun, 25 Oct 2026 00:00:00 +0000

Carbon (C) markets are well established for aboveground woody vegetation domestically, and for soil organic C overseas. By reviewing published literature, this project aims to provide a scientifically defensible framework and dataset for estimating the vegetation C stocks in New Zealand grassland ecosystems. New Zealand currently uses IPCC Tier 1 values for biomass and root-to-shoot ratios derived from global vegetation types that do not align well with our intensively managed grassland systems. Shifting the High and Low producing grassland types used in the Land Cover Database to a Tier 2 approach requires inclusion of New Zealand-specific values of aboveground and belowground biomass. From the literature search of field measurements, the overall mean root-to-shoot ratio was 1.35 ± 1.86 and 3.64 ± 2.91 for High and Low producing grassland, respectively. These differ substantially from those root-o- shoot ratios (4.0 and 2.8) currently used in National Inventory calculations. There is good reason to challenge the Tier 1 assumptions, as a higher proportion
of the C fixed by the plant is allocated belowground in environments where nutrients are scarce, such as in Low producing grassland, compared with High producing grassland. The review also identified an overall lower C concentration in plant tissue than that used in the inventory. This might be contributing to an overestimation of vegetation C stocks in grassland. We suggest the replacement of Tier 1 with Tier 2 values. Doing so provides a more accurate assessment of the
C stocks in New Zealand grassland for the purpose of inventory reporting.

Pasture productivity under pressure despite continued improvement in sheep and beef production efficiency

Ronaldo Eduardo Vibart, Jane Campbell, Alec Mackay, Mike Dodd — Sun, 25 Oct 2026 00:00:00 +0000

Pastoral hill country landscapes are the cornerstone of New Zealand’s sheep and beef sector. A growing body of evidence indicates that pasture productivity gains in New Zealand have stagnated or, in some cases, declined over recent decades. The use of methods to back-calculate animal intake from livestock energy requirements and energy concentration of pasture on offer, has proven effective in the assessment of home-grown feed for dairy, and to a limited extent, for sheep and beef. This paper explores the complexities of extracting similar information to that of Gobilik et al. (2021) to extend regional data (Class 4 lower North Island Hill Country) beyond 2011 to current times using data from the B+LNZ Sheep and Beef Farm Survey. Our analysis suggests that estimated feed demand has remained flat despite continued gains
in livestock reproductive performance indicators, consistent with heightened pressure on pasture supply. This divergence is critical for livestock managers where home-grown feed is so prominent in terms of forage supply. The present analysis relied on stock-unit based feed-demand proxies rather than a comprehensive set of animal production characteristics, and further work is needed to integrate detailed farm-level measurements and improved estimates of livestock liveweight and liveweight changes, to improve the assessment of pasture productivity trends.

Decadal trends in plantation Pinus radiata (D. Don) forest growth

Sun, 25 Oct 2026 00:00:00 +0000

Plantation Pinus radiata (D. Don) forests are longterm crops uniquely exposed to climate change; trees planted today and harvested after 2050 will experience multiple cycles of variable rainfall, along with an anticipated increase of another approximately 100 ppm in atmospheric CO2 (eCO2) and associated warming. While process-based modelling has suggested substantial increases in tree growth in higher-rainfall regions, empirical model outputs indicate the rate of
increases in productivity has slowed. We analysed three decades (1992–2022) of growth data across permanent
P. radiata plots in New Zealand to quantify recent trends in plantation productivity and to assess the implications of any decline for the sector, farm forestry and the wider economy. P. radiata growth increased by 25% from the 1990s to the 2000s but plateaued to a <5% gain in the 2010s, despite predicted benefits from rising eCO2, and continued adoption of improved germplasm, silvicultural practices and the practice of planting on landscapes with a pastoral history. The
age-class distribution became increasingly skewed towards recently established and younger plantations in later decades, which may partially mask underlyingchanges in productivity. These findings indicate that gains in forest productivity have slowed, highlighting the need for more targeted management interventions to maximise the benefits of eCO2 through efficient water and nutrient supply and utilisation.

Impact of pasture allowance during the transition period in ewes carrying singles, twins and triplets on lamb survival, growth and ewe efficiency to weaning with and without human intervention

Sun, 25 Oct 2026 00:00:00 +0000

Ewe and lamb mortality increases with fecundity, but practical on-farm solutions remain limited. This study evaluated the effects of pasture allowance during the transition period, from 100 days of gestation to 4 weeks of lactation, and its interaction with litter size on ewe and lamb performance with and without human intervention. The design was a 2 × 3 factorial, with two pasture allowances — High and Std, (achieved through pasture allocation regimes of 3000/2000 kg DM/ha and 2000/1400 kg DM/ha pre-/post-grazing for High and Standard respectively) — and three litter-size classes:
single, twin, and triplet. Allowances were calculated for twin-bearing ewes and applied to all ewes regardless of litter size. Ewe body condition score was similar between all groups until late gestation, decreased from late gestation to 4 weeks of lactation as litter size increased, but were stable thereafter. These results highlight the mismatch between feed consumed and the energetic demands of late pregnancy and early lactation as litter size increases. Increasing pasture allowance
reduced ewe mortality and morbidity and improved ewe efficiency by increasing survival of triplet-born lambs with intervention and twin- and triplet-born lambs without intervention. First-year data suggest current New Zealand pasture allowance recommendations may limit performance of highly fecund ewes, although further research is needed to confirm mechanisms, practicality, and economics.

Fecundity and ewe liveweight affect ewe wastage, productivity and greenhouse gas emissions

David Stevens — Sun, 25 Oct 2026 00:00:00 +0000

Ewe wastage, defined as ewe death and failure to rear lambs, reduces productivity, profitability and ecoefficiency in New Zealand sheep systems. National survey data identified ewe fecundity and ewe liveweight as significant drivers of wastage, but their whole-farm impacts are poorly quantified. This study modelled the effects of changes in ewe liveweight (−10 to +10 kg) and fecundity (−20% to +20% scanning percentage) on flock dynamics, production, gross margins and greenhouse gas (GHG) emissions, assuming a constant annual feed supply. National average performance was represented by a fecundity of 184%, ewe liveweight of 66.4 kg, ewe mortality of 5.5% and 14.2% of ewes rearing no lambs.
Increasing ewe liveweight from 56.4 to 76.4 kg reduced ewe mortality from 6.4% to 4.6% and failure to rear lambs from 18.6% to 9.8%, increasing weaning rate from 131% to 150%. Increasing fecundity from 164% to 204% reduced ewe mortality from 6.1% to 4.9%, reduced ewes rearing no lambs from 17.8% to 10.6% and increased weaning rate from 117% to 166%. Under constant feed supply, higher ewe liveweight resulted in little change in lambs sold (≤20 head), whereas higher fecundity increased lambs sold by approximately 480 head per farm. Gross margin per ewe increased from
$102 to $125 with increasing liveweight and from $86 to $145 with increasing fecundity. Total GHG emissions declined by 5–6%, with emissions intensity reduced by up to 18% per unit of product. Reducing ewe wastage through improved fecundity delivers larger system level gains than increasing liveweight alone, although both pathways improve productivity, profitability and emissions efficiency.

Impact of rearing practices on modelled economic outcomes for pasture-based dairy heifer and dairy-beef production systems

David Stevens, Susan A McCoard — Sun, 25 Oct 2026 00:00:00 +0000

The application of economic modelling to the effectsof high milk inputs and failure of passive transfer of immunity (FPT) in both dairy heifer and dairy-beef calf rearing systems demonstrated the greater effect of milk input on production parameters than FPT. However, high FPT had greater effects on mortality and morbidity, resulting in increased input costs and financial impacts. Increasing milk volumes reduced the overall cost of rearing dairy heifers by increasing first year milk production. Reducing the incidence of FPT had a similar positive effect on both dairy heifer and dairy-beef calf rearing enterprises, reducing total costs
by approximately 19%.

Dataset providers’ and users’ perceptions of data sharing: insights from AgYields New Zealand National Database

Carmen Teixeira, Derrick Moot — Sun, 25 Oct 2026 00:00:00 +0000

• Data sharing applications have become a functional necessity to advance smart farming, to assist routine farm management strategies and for environmental compliance.

• An analysis of the database status revealed that 61% of the uploaded datasets are from unpublished sources; of this subset, the majority was collected onfarm (37%), followed by technical reports (30%).

• AgYield users experience revealed that data providers from education and technical sectors were comfortable with data sharing when it is framed as contributing to mutual benefit, assisting innovation and communities and supporting the wider agri‑sector to thrive.

• Commercial partners and other data providers want data access and the efficiency of integrated systems; however, some remain cautious of losing competitive advantage, privacy, data control and data sovereignty.

• Strategies to foster sharing include the exclusion of metadata perceived as sensitive by companies’ interests, particularly cultivar details

From Science to Software: Developing a product governance approach for environmental models

Mark-John Bruwer, Jill Gower — Sun, 25 Oct 2026 00:00:00 +0000

Environmental models deployed as software at scale require governance across science, data, code, release management, user enablement, and security. This extends beyond core scientific rigour without replacing its essential relevance.
• A practical governance framework can be organised around governance domains, explicit trade-offs, and records that support accountability, learning and improvement.
• The Overseer case is presented as an illustrative account of how these ideas are being explored in Aotearoa New Zealand. It is not offered as evidence that the approach is complete or effective in every respect.
• An important corollary is that stewardship funding must extend beyond scientific research publication to include software development, quality assurance, documentation, support, and ongoing governance.

Managing technological change in a New Zealand rural and farm supply business: “A case study involving AI technology adoption”

Emily Christensen, Xiaomeng Lucock, Meike Rombach, David Dean — Sun, 25 Oct 2026 00:00:00 +0000

Rural and farm supply businesses are essential to New Zealand’s agrifood sector, serving over 45,000 farms. The present study is dedicated to advisors working in one of New Zealand’s businesses that provide supplies for pasture management. This qualitative case study explores how farm advisors' perceive changes in their current practices towards a new AI-based platform designed to aid them in supporting farmers and their decision-making. Further, the study is dedicated to factors that drive and hinder the adoption of this platform. Using a qualitative case study to represent the voices of this stakeholder group is suitable, as it highlights how these individuals perceive, interpret and react to the change situation. The case study builds on a variety of data sources including webpage materials, responses to a 45-minute in-depth interview with an individual leading the AI transition within the rural and farm supply business. A purposive sampling frame was chosen, and a total of 51 farm advisors responded to the questionnaire. Results show that advisors involved in the study had mixed feelings about the adoption and were critical towards their corporate management. They believed their daily pressures of fieldwork, the requirements for, and nuances within, their work were not fully understood and accommodated within the change process. Advisors' trust and willingness to adapt appeared somewhat limited, due to shortcomings of prior systems and change management experiences. They viewed the change as an added burden, requiring double-handling of data. Further, advisors believed that the top-down approach towards change management hindered success. The implementation of training and focusing on peer-to-peer support would have been preferred. Solving these internal issues is necessary to ensure that digital investments successfully support staff and farmers

AR153 - a new Epichloë fungal endophyte for ryegrass

Sun, 25 Oct 2026 00:00:00 +0000

Perennial ryegrass (Lolium perenne L.) persistence in many parts of New Zealand is reliant on its association with the mutualistic Epichloë fungal endophyte. This endophyte protects ryegrass from various insect pests and can improve tolerance to some abiotic stresses such as drought. Over the last 25 years several selected endophyte strains have been commercialised. These strains have improved ryegrass persistence and minimised the negative impacts on livestock of ryegrass staggers, heat stress and productivity losses associated with the standard endophyte strain. A new endophyte
strain is becoming available – AR153, as a replacement for AR1. This Epichloë strain from Italy has a similar known chemical profile to AR1, through expression of peramine, but it also produces chanoclavine. AR153 protects ryegrass against the same insect pests as AR1. It exhibits improved or similar animal safety, high transmission to seed, improved storage in seed and similar yield and persistence compared with AR1. No food safety concerns have been identified. These traits indicate that AR153 is an effective replacement for AR1, supporting greater ryegrass persistence than
endophyte-free pastures while reducing the incidence of ryegrass staggers.