• The short-term effects of management changes on watertable position and nutrients in shallow groundwater in a harvested peatland forest

      Finnegan, Joanne; Regan, J. T.; Fenton, Owen; Lanigan, Gary; Brennan, Raymond B.; Healy, Mark G.; Environmental Protection Agency; Department of Agriculture, Food and the Marine, Ireland (Elsevier, 08/05/2014)
      Management changes such as drainage, fertilisation, afforestation and harvesting (clearfelling) of forested peatlands influence watertable (WT) position and groundwater concentrations of nutrients. This study investigated the impact of clearfelling of a peatland forest on WT and nutrient concentrations. Three areas were examined: (1) a regenerated riparian peatland buffer (RB) clearfelled four years prior to the present study (2) a recently clearfelled coniferous forest (CF) and (3) a standing, mature coniferous forest (SF), on which no harvesting took place. The WT remained consistently below 0.3 m during the pre-clearfelling period. Results showed there was an almost immediate rise in the WT after clearfelling and a rise to 0.15 m below ground level (bgl) within 10 months of clearfelling. Clearfelling of the forest increased dissolved reactive phosphorus concentrations (from an average of 28–230 μg L−1) in the shallow groundwater, likely caused by leaching from degrading brash mats.
    • The significance of livestock as a contributor to global greenhouse gas emissions today and in the near future

      O'Mara, Frank P. (Elsevier Inc., 23/06/2011)
      Animal agriculture is responsible for 8–10.8% of global greenhouse gas (GHG) emissions as assessed by IPCC accounting and, on the basis of lifecycle analysis, the contribution of livestock is up to 18% of global emissions. Asia is the source of the most enteric CH4 emissions with Latin America, Africa, Western Europe and North America being significant sources. These emissions are dominated by emissions from cattle. When GHG emissions are related to food production, the four most efficient regions are Eastern and Western Europe, North America, and the non-EU former Soviet Union which produced 46.3% of ruminant meat and milk energy and only 25.5% of enteric CH4 emissions in 2005. In comparison, the three least efficient producers (Asia, Africa, Latin America) produced an equivalent amount (47.1%) of ruminant meat and milk energy, and almost 69% of enteric CH4 emissions in 2005. Livestock related emissions will increase as world population and food demand increases; enteric CH4 emissions are projected to grow by over 30% from 2000 to 2020. There are mitigations available now, but it is imperative to develop new mitigations and ways to implement existing technologies more cost effectively.
    • Soil bacterial community structure and functional responses across a long-term mineral phosphorus (Pi) fertilisation gradient differ in grazed and cut grasslands

      Randall, Kate; Brennan, Fiona P.; Clipson, Nicholas; Creamer, Rachel; Griffiths, Bryan; Storey, Sean; Doyle, Evelyn; Programme for Research in Third-Level Institutions; European Regional Development Fund (Elsevier, 2019-03-06)
      Grasslands form a significant proportion of land used across the globe and future management is important. The objective of this study was to compare the long-term impact of inorganic phosphorus (Pi) fertilisation rates (P0, P15 and P30 ha−1 yr−1) under two grass management trials (grazed vs. cut and removed) on soil physicochemical properties, microbial biomass, phosphomonoesterase activity, bacterial community structure and abundance of a phosphorus (P) mineralising gene (phoD). Under grazing, microbial biomass and soil phosphorus concentrations (total and Pi) generally increased with Pi fertilisation rate, accompanied by significant differences in bacterial community structure between unfertilised (P0) and P30 soil. At the cut and removed site, although Pi was significantly greater in P30 soil, P concentrations (total and Pi) did not increase to the same extent as for grazing, with microbial biomass and bacterial community structures unresponsive to Pi fertilisation. Despite differences in soil P concentrations (total and Pi) and microbial biomass between sites, the abundance of bacterial phoD increased with increasing soil Pi across both sites, while phosphomonoesterase activity decreased. Amplicon sequencing revealed Acidobacteria were the dominant bacterial phylum across both grasslands, but significant differences in relative abundances of bacterial genera were detected at the grazed site only. The bacterial genera Gp6 and Gp16 increased significantly with Pi fertilisation under grazing. Conversely, Bradyrhizobium as well as unclassified genus-type groups belonging to Actinobacteria and Acidimicrobiales significantly decreased with Pi fertilisation, suggesting potential roles in P mobilisation when soil Pi concentrations are low. This study highlights the importance of long-term Pi fertilisation rates and aboveground vegetation removal in shaping soil bacterial community structure and microbial biomass, which in turn may impact soil fertility and plant productivity within agricultural soils.
    • Soil biodiversity, biological indicators and soil ecosystem services - an overview of European approaches

      Pulleman, Mirjam; Creamer, Rachel E.; Hamer, Ute; Helder, Johannes; Pelosi, Celine; Peres, Guenola; Rutgers, Michiel (Elsevier, 2012-11)
      Soil biota are essential for many soil processes and functions, yet there is an increasing pressure on soil biodiversity and soil degradation remains a pertinent issue. Therefore, the sustainable management of soils requires soil monitoring, including biological indicators able to relate land use and management to soil functioning and ecosystem services. Since the 1990’s, biological soil parameters have been assessed in an increasing number of field trials and monitoring programmes across Europe. The development and effective use of meaningful and widely applicable bio-indicators however, continues to be a challenging tasks. This paper aims to provide an overview of current knowledge in relation to soil biodiversity characterization and assessment. Examples of European monitoring approaches and soil biodiversity indicators are presented, and the value of soil biodiversity databases for developing a better understanding of the relations between soil management and ecosystem functions and services is discussed. We conclude that integration of monitoring approaches and data sets, together with state-of-the art ecological expertise, offers good opportunities for advancing ecological theory as well as application of such knowledge by decision makers.
    • Spatial evaluation and trade‐off analysis of soil functions through Bayesian networks

      Vrebos, Dirk; Jones, Arwyn; Lugato, Emanuele; O’Sullivan, Lillian; Schulte, Rogier; Staes, Jan; Meire, Patrick; European Union; 635201 (Wiley, 2020-08-23)
      There is increasing recognition that soils fulfil many functions for society. Each soil can deliver a range of functions, but some soils are more effective at some functions than others due to their intrinsic properties. In this study we mapped four different soil functions on agricultural lands across the European Union. For each soil function, indicators were developed to evaluate their performance. To calculate the indicators and assess the interdependencies between the soil functions, data from continental long‐term simulation with the DayCent model were used to build crop‐specific Bayesian networks. These Bayesian Networks were then used to calculate the soil functions' performance and trade‐offs between the soil functions under current conditions. For each soil function the maximum potential was estimated across the European Union and changes in trade‐offs were assessed. By deriving current and potential soil function delivery from Bayesian networks a better understanding is gained of how different soil functions and their interdependencies can differ depending on soil, climate and management. Highlights When increasing a soil function, how do trade‐offs affect the other functions under different conditions? Bayesian networks evaluate trade‐offs between soil functions and estimate their maximal delivery. Maximizing a soil function has varied effects on other functions depending on soil, climate and management. Differences in trade‐offs make some locations more suitable for increasing a soil function then others.
    • Surface and Groundwater Interactions: Location of a sub-surface remediation trench

      Fenton, Owen (Teagasc, 01/12/2006)
      The Water Framework Directive aims to achieve at least “good status” of all surface and groundwater bodies by 2015. In 2009 programmes of measures to achieve this status must be implemented. In 2012 water quality response to these measures will be examined at river basin catchment level. The adoption of the Water Framework Directive from the 1st January 2007 restricts the amount of nutrients which can be applied to agricultural land. A nutrient discharge to a waterbody has a negative impact on the environment and may lead to eutrophication. A broad strategy exists at European level to minimise nutrient loss to a waterbody. This strategy examines the source/pressure, pathway and receptor approach for nutrient transport. Such nutrient management strategies try to minimise nutrient loss while maintaining productivity. Nitrogen usage is now associated with environmental degradation even at lower levels than the maximum allowable concentration (11.3 mg NO3-N L-1). A further strategy proposes that nutrient management and increased utilisation of nutrients alone will fail to recognise nutrient loss even at high levels of efficiency. This strategy attempts to use remediation (Nitrate) and control technologies (Phosphorous) to intercept nutrients before discharge. Another function would be to further reduce concentrations presently at allowable levels. This introduces an interceptor phase into the nutrient transfer model.
    • Sustainability indicators for improved assessment of the effects of agricultural policy across the EU: Is FADN the answer?

      Kelly, Edel; Latruffe, Laure; Desjeux, Yann; Ryan, Mary; Uthes, Sandra; Diazabakanab, Ambre; Dillon, Emma; Finn, John; European Union; 613800 (Elsevier, 2018-06)
      Policy reform of the CAP and society’s expectations of agriculture have resulted in a growing need for improved information on the effectiveness of policy in achieving high-level objectives for more sustainable practice in agriculture. This is a high priority given its importance for consumers, public policy and private industry. Data collection programmes will need to adapt their scope if their information is to adequately address new information needs about high-level objectives. Assessment of sustainability at the farm level is hindered by the lack of data with which to derive appropriate, meaningful, and relevant indicators. This is particularly problematic for assessment of agricultural sustainability across the European Union (EU). Various databases exist at the EU scale regarding agricultural data sources and we identify one of these, the EU Farm Accountancy Data Network (FADN), as having considerable potential to assess farm-level sustainability at EU level. We critique several examples of published work that has attempted to assess agricultural sustainability using: FADN data alone; FADN data in combination with data from supplementary surveys, and; FADN data in combination with data from other EU databases. We conclude that the FADN would need to broaden its scope of data collection if it is to address the new information needs of policy, and we discuss the challenges in expanding FADN with a view towards wider farm-level assessment of sustainability. These include careful selection of indicators based on various criteria, the representativeness of the FADN, and the need to include new themes to address environmental, social, and animal welfare effects of policy.
    • Sustainable grassland systems in Europe and the EU Water Framework Directive

      Tunney, Hubert; Watson, C. J.; Kronvang, B.; Stamm, C.; Vertes, F.; Richards, Karl G.; Gibson, Mark; Fenton, Owen; Schulte, Rogier P. (School of Agriculture, Food Science and Veterinary Medicine, University College Dublin in association with Teagasc, 2009)
      The International Conference on ‘Sustainable grassland systems in Europe and the EU Water Framework Directive’ took place at Teagasc, Johnstown Castle, Wexford from 12th to 14th November 2008. There were approximately 150 participants from Europe, the USA and New Zealand. Most of the invited papers are published in this conference issue. The main aim of the Conference was to identify the challenges that the Water Framework Directive (WFD) presents for grassland agriculture and to help guide the implementation of cost effective mitigation measures. The Conference focused on nutrient (mainly nitrogen and phosphorus) loss from grassland and the implications for sustainable production and water quality. This paper summarises the main points and outcomes of discussions and recommendations from the Conference. It was concluded that it is difficult to link the management practices on individual fields or farms with the effects on water quality and ecological conditions in surface waters at the catchment-scale. There is a need to identify areas of highest risk of nutrient loss from point and diffuse sources to a waterbody of vulnerable status and then to focus mitigation measures in the critical source areas, where there is the greatest risk to water quality. Participants agreed that there can be a substantial lag time between the implementation of measures and improvements in water quality. A participatory approach at local level, with personal contact, is considered more productive for securing a positive response to adopting measures. Concerns were expressed that maps and models may be misinterpreted. It was recommended that estimates of accuracy should always be shown when presenting map data and modelled results. Success stories in reducing nutrient loss to water were reported and examples from Denmark and Switzerland were outlined. There is no consensus about the most important mitigation options; they will vary for different situations. The effective implementation of the Nitrates and Urban Waste Water Directives should go a long way towards meeting farming obligations under the WFD. The need for adaptive integrated management was recognised. How mitigation measures can be compared across a wide range of agricultural systems in several EU states, has not yet been explored and to achieve this, further cooperation on the most appropriate options is needed. Similarities and differences between the situation in New Zealand and the USA compared to the EU were also presented and discussed.
    • Sward composition and soil moisture conditions affect nitrous oxide emissions and soil nitrogen dynamics following urea-nitrogen application

      Bracken, Conor J.; Lanigan, Gary J.; Richards, Karl G.; Müller, Christoph; Tracy, Saoirse R.; Grant, James; Krol, Dominika; Sheridan, Helen; Lynch, Mary Bridget; Grace, Cornelia; et al. (Elsevier BV, 2020-06)
      Increased emissions of N2O, a potent greenhouse gas (GHG), from agricultural soils is a major concern for the sustainability of grassland agriculture. Emissions of N2O are closely associated with the rates and forms of N fertilisers applied as well as prevailing weather and soil conditions. Evidence suggests that multispecies swards require less fertiliser N input, and may cycle N differently, thus reducing N loss to the environment. This study used a restricted simplex-centroid experimental design to investigate N2O emissions and soil N cycling following application of urea-N (40 kg N ha−1) to eight experimental swards (7.8 m2) with differing proportions of three plant functional groups (grass, legume, herb) represented by perennial ryegrass (PRG, Lolium perenne), white clover (WC, Trifolium repens) and ribwort plantain (PLAN, Plantago lanceolata), respectively. Swards were maintained under two contrasting soil moisture conditions to examine the balance between nitrification and denitrification. Two N2O peaks coincided with fertiliser application and heavy rainfall events; 13.4 and 17.7 g N2O-N ha−1 day−1 (ambient soil moisture) and 39.8 and 86.9 g N2O-N ha−1 day−1 (wet soil moisture). Overall, cumulative N2O emissions post-fertiliser application were higher under wet soil conditions. Increasing legume (WC) proportions from 0% to 60% in multispecies swards resulted in model predicted N2O emissions increasing from 22.3 to 96.2 g N2O-N ha−1 (ambient soil conditions) and from 59.0 to 219.3 g N2O-N ha−1 (wet soil conditions), after a uniform N application rate. Soil N dynamics support denitrification as the dominant source of N2O especially under wet soil conditions. Significant interactions of PRG or WC with PLAN on soil mineral N concentrations indicated that multispecies swards containing PLAN potentially inhibit nitrification and could be a useful mitigation strategy for N loss to the environment from grassland agriculture.
    • TaLAM: Mapping Land Cover in Lowlands and Uplands with Satellite Imagery

      Cawkwell, Fiona; Raab, Christoph; Barrett, Brian; green, stuart; Finn, John; Environmental Protection Agency (Environmental Protection Agency, 2018-07)
      The Towards Land Cover Accounting and Monitoring (TaLAM) project is part of Ireland’s response to creating a national land cover mapping programme. Its aims are to demonstrate how the new digital map of Ireland, Prime2, from Ordnance Survey Ireland (OSI), can be combined with satellite imagery to produce land cover maps.
    • Teagasc submission made in response to the Consultation Paper on Interim Review of Ireland’s Nitrates Derogation 2019

      Spink, John; Buckley, Cathal; Burgess, Edward; Daly, Karen M.; Dillon, Pat; Fenton, Owen; Horan, Brendan; Humphreys, James; Hyde, Tim; McCarthy, Brian; et al. (Teagasc, 2019-06-04)
      This submission was made in response to the consultation process run jointly by the Department of Housing, Planning, Community and Local Government (DHPCLG) and the Department of Agriculture, Food and the Marine (DAFM) inviting views and comments on proposals for the Interim Review of Ireland’s Nitrates Derogation Programme in 2019. It has been prepared by Teagasc’s Water Quality Working Group in consultation with the Gaseous Emissions Working Group. These working groups have members drawn from both the Knowledge Transfer and Research Directorates of Teagasc. It was prepared following consultation with colleagues across Teagasc using their collective knowledge and expertise in agri-environmental science and practice and the implementation of the Good Agricultural Practice (GAP) and Nitrates Derogation Regulations.
    • Technical Note: Field experiences using UV/VIS sensors for high-resolution monitoring of nitrate in groundwater

      Huebsch, Manuela; Grimmeisen, F.; Zemann, M.; Fenton, Owen; Richards, Karl G.; Jordan, Philip; Sawarieh, A.; Blum, P.; Goldscheider, N.; Teagasc Walsh Fellowship Programme; et al. (European Geosciences Union, 02/04/2015)
      Two different in situ spectrophotometers are compared that were used in the field to determine nitrate-nitrogen (NO3-N) concentrations at two distinct spring discharge sites. One sensor was a double wavelength spectrophotometer (DWS) and the other a multiple wavelength spectrophotometer (MWS). The objective of the study was to review the hardware options, determine ease of calibration, accuracy, influence of additional substances and to assess positive and negative aspects of the two sensors as well as troubleshooting and trade-offs. Both sensors are sufficient to monitor highly time-resolved NO3-N concentrations in emergent groundwater. However, the chosen path length of the sensors had a significant influence on the sensitivity and the range of detectable NO3-N. The accuracy of the calculated NO3-N concentrations of the sensors can be affected if the content of additional substances such as turbidity, organic matter, nitrite or hydrogen carbonate significantly varies after the sensors have been calibrated to a particular water matrix. The MWS offers more possibilities for calibration and error detection but requires more expertise compared with the DWS.
    • Temperate Grassland Yields and Nitrogen Uptake Are Influenced by Fertilizer Nitrogen Source

      Harty, Mary A.; Forrestal, Patrick J.; Carolan, Rachael; Watson, C. J.; Hennessy, Deirdre; Lanigan, Gary; Wall, David; Richards, Karl G.; Department of Agriculture, Food and the Marine, Ireland; Teagasc Walsh Fellowship Programme; et al. (American Society of Agronomy, 25/01/2017)
      In temperate grasslands, N source influences greenhouse gas emissions. Nitrification and urea hydrolysis inhibitors can reduce these losses. The objective of this study was to evaluate the impact of N source, urease inhibitors, and nitrification inhibitors on temperate grassland yields and N uptake. Experiments were conducted at three locations over 2 years (6 site-years) on the island of Ireland, covering a range of soils and climatic conditions. Results showed that calcium ammonium nitrate (CAN), urea+N-(n-butyl) thiophosphoric triamide (NBPT), urea+NBPT+dicyandiamide (DCD), and urea had equal annual dry matter yield. Urea+DCD had lower dry matter yield than CAN for 3 site-years. Calcium ammonium nitrate and urea+NBPT consistently had the same N uptake, urea+DCD had lower N uptake than CAN in 4 of 6 site-years, urea had lower N uptake than CAN in 2 site-years, and urea+NBPT+DCD had lower N uptake than CAN in 1 site-year. Urea+NBPT is a cost-effective alternative to CAN, which is consistently equal in terms of yield and N uptake in temperate grassland.
    • Time lag: a methodology for the estimation of vertical and horizontal travel and flushing timescales to nitrate threshold concentrations in Irish aquifers

      Fenton, Owen; Schulte, Rogier P.; Jordan, Philip; Lalor, Stanley T. J.; Richards, Karl G.; Department of Environment, Heritage and Local Government; RPS Group Ltd (Elsevier Science Ltd., 02/04/2011)
      The Water Framework Directive (WFD) in Europe aims, inter alia, to achieve at least “good” water quality status by 2015 by mitigating the causes of pollution. However, with the implementation of programmes of measures in 2012, many catchments may not achieve good water quality status within this timeframe due to the time lag of nutrient transport from source to receptor via hydrological and hydrogeological pathways. An appraisal of catchment time lag issues offers a more realistic scientifically based timescale for expected water quality improvements in response to mitigation measures implemented under the WFD. A simplified methodology for the calculation of nitrate time lag in a variety of Irish hydrogeological scenarios is presented, based on unsaturated vertical and aquifer flushing times required to reach environmental quality standards. Horizontal travel time is estimated for first occurrence of nutrients in a surface water body. The results show that achievement of good water quality status in the Republic of Ireland for some waterbodies may be too optimistic within the current timeframe of 2015 targets but improvements are predicted within subsequent 6 and 12 year cycles.
    • Unsaturated zone travel time to groundwater on a vulnerable site

      Richards, Karl G.; Coxon, Catherine E.; Ryan, Michael (Taylor & Francis, 2005)
      A bromide (Br) tracing experiment was conducted to ascertain unsaturated zone travel time to groundwater on a site with a karstified limestone aquifer overlain by a thin free-draining overburden. Br tracer was applied to areas surrounding two boreholes; soil solution and groundwater Br concentrations were monitored. Bromide was first detected after eight and 34 days in the soil solution and groundwater. The quick break-through of the applied Br in the soil solution and groundwater indicates the presence of preferential flow in the soil at this site. The time to maximum groundwater Br concentration supports a dominant matrix flow path through the overburden and then preferential flow through the unsaturated limestone bedrock. The results indicated that the transport of conservative contaminants, such as nitrate, can be expected to occur in a single recharge season. The occurrence of preferential flow raises concerns over rapid transport of non-conservative contaminants such as faecal coliforms and this merits further investigation.
    • Urine patch distribution under dairy grazing at three stocking rates in Ireland

      Dennis, S.J.; Moir, James L.; Cameron, K.C.; Di, H.J.; Hennessy, Deirdre; Richards, Karl G. (Teagasc (Agriculture and Food Development Authority), Ireland, 2011)
      Nitrate pollution of water is a serious global environmental issue. Grassland agriculture is a major source of diffuse nitrate pollution, with much of this nitrate originating from the urine patches of grazing animals. To study nitrate losses from grassland it is necessary to consider the areas of grassland that are affected by urine separately from the remainder of the pasture. Urine patches can be observed in the field as areas of vigorously growing pasture, however the pasture may continue to respond for several months, making it difficult to determine when the observed patch was actually deposited. A global positioning system was used to record the location of all urine and dung patches in a pasture at every second grazing on an Irish dairy farm during the grazing season. Any patches reappearing were removed from the data, allowing the fresh urine patches to be identified. Dairy cows deposited 0.359 urine patches per grazing hour, a value that may be used to predict the distribution of urine patches under any grazing regime. This equated to 14.1 to 20.7% of the soil surface being wet by urine annually at stocking rates of 2.0 to 2.94 cows per hectare, consistent with previous research. These values may be used in conjunction with values for nitrate loss from urine and non-urine areas to calculate nitrate losses from grazed pasture at a range of stocking rates.
    • Use of Ochre from an Abandoned Metal Mine in the South East of Ireland for Phosphorus Sequestration from Dairy Dirty Water

      Fenton, Owen; Healy, Mark G.; Rodgers, M. (Soil Science Society of America, 2009-05)
      Ochre found at coal mine drainage sites in the United Kingdom shows a high phosphorus (P) retention capacity with little mobilization of metals. This indicates that ochre has the potential to adsorb P from agricultural wastewaters for possible use as a fertilizer. Little research has focused on the ability of metal mine ochre to sequester P in an environmentally sustainable way. Untreated acid mine drainage from an abandoned coppersulfur mine in the Avoca-Avonmore catchment in the south east of Ireland results in extensive low-value ochre deposition. In this study, P-amended water (50 mL) was mixed with this ochre (2.5 g) in batch experiments, and a maximum P adsorption capacity, calculated from the Langmuir equation, of between 16 and 21 g P kg−1 was calculated. However, mobilization of heavy metals from Avoca ochre in distilled, surface, and dirty water batch experiments was observed. This mobilization may inhibit ochre’s use in P removal from wastewaters.
    • The use of Sulphur as a Fertilizer

      Brogan, J.C.; Murphy, Matthew D. (An Foras Taluntais, 1979)
    • Using a multi-dimensional approach for catchment scale herbicide pollution assessments

      Khan, Majid Ali; Costa, Fabiola Barros; Fenton, Owen; Jordan, Phil; Fennell, Chris; Mellander, Per-Erik; European Union; 727450 (Elsevier BV, 2020-12)
      Worldwide herbicide use in agriculture, whilst safeguarding yields also presents water quality issues. Controlling factors in agricultural catchments include both static and dynamic parameters. The present study investigated the occurrence of herbicides in streams and groundwater in two meso-scale catchments with contrasting flow controls and agricultural landuse (grassland and arable land). Using a multi-dimensional approach, streams were monitored from November 2018 to November 2019 using Chemcatcher® passive sampling devices and groundwater was sampled in 95 private drinking water wells. The concentrations of herbicides were larger in the stream of the Grassland catchment (8.9–472.6 ng L−1) dominated by poorly drained soils than in the Arable catchment (0.9–169.1 ng L−1) dominated by well-drained soils. Incidental losses of herbicides during time of application and low flows in summer caused concentrations of MCPA, Fluroxypyr, Trichlorpyr, Clopyralid and Mecoprop to exceeded the European Union (EU) drinking water standard due to a lack of dilution. Herbicides were present in the stream throughout the year and the total mass load was higher in winter flows, suggesting a persistence of primary chemical residues in soil and sub-surface environments and restricted degradation. Losses of herbicides to the streams were source limited and influenced by hydrological conditions. Herbicides were detected in 38% of surveyed drinking water wells. While most areas had concentrations below the EU drinking water standard some areas with well-drained soils in the Grassland catchment, had concentrations exceeding recommendations. Individual wells had concentrations of Clopyralid (619 ng L−1) and Trichlorpyr (650 ng L−1). Despite the study areas not usually associated with herbicide pollution, and annual mass loads being comparatively low, many herbicides were present in both surface and groundwater, sometimes above the recommendations for drinking water. This whole catchment assessment provides a basis to develop collaborative measures to mitigate pollution of water by herbicides.
    • Using machine learning to estimate herbage production and nutrient uptake on Irish dairy farms

      Nikoloski, Stevanche; Murphy, Philip; Kocev, Dragi; Džeroski, Sašo; Wall, David; Teagasc Walsh Fellowship Programme; European Union; 635201 (Elsevier for American Dairy Science Association, 2019-08-22)
      Nutrient management on grazed grasslands is of critical importance to maintain productivity levels, as grass is the cheapest feed for ruminants and underpins these meat and milk production systems. Many attempts have been made to model the relationships between controllable (crop and soil fertility management) and noncontrollable influencing factors (weather, soil drainage) and nutrient/productivity levels. However, to the best of our knowledge not much research has been performed on modeling the interconnections between the influencing factors on one hand and nutrient uptake/herbage production on the other hand, by using data-driven modeling techniques. Our paper proposes to use predictive clustering trees (PCT) learned for building models on data from dairy farms in the Republic of Ireland. The PCT models show good accuracy in estimating herbage production and nutrient uptake. They are also interpretable and are found to embody knowledge that is in accordance with existing theoretical understanding of the task at hand. Moreover, if we combine more PCT into an ensemble of PCT (random forest of PCT), we can achieve improved accuracy of the estimates. In practical terms, the number of grazings, which is related proportionally with soil drainage class, is one of the most important factors that moderates the herbage production potential and nutrient uptake. Furthermore, we found the nutrient (N, P, and K) uptake and herbage nutrient concentration to be conservative in fields that had medium yield potential (11 t of dry matter per hectare on average), whereas nutrient uptake was more variable and potentially limiting in fields that had higher and lower herbage production. Our models also show that phosphorus is the most limiting nutrient for herbage production across the fields on these Irish dairy farms, followed by nitrogen and potassium.