Jordon, M.W. (2022) Regenerative Agriculture practices and their influence on soil organic carbon and farm productivity in temperate regions. Ph.D. thesis. University of Oxford.

Abstract
Global food production is currently contributing to numerous environmental challenges including anthropogenic climate change, biodiversity loss and widespread declines in soil health. Regenerative Agriculture (RA) is an emerging paradigm which proposes that restoring soil health, by adopting a suite of management practices that align with so-called ‘regenerative’ principles, will not only improve agricultural productivity but also contribute to climate change mitigation through increased soil organic carbon (SOC). Despite a plethora of policies and suggested practices emerging over the past decade to support the introduction of RA across the world, there is still remarkably little scientific evidence to quantify potential improvements in soil organic carbon or agricultural productivity as a result of adopting ‘holistic’ RA systems. This represents a large knowledge gap that I set out to address in my DPhil thesis using currently available evidence for individual management practices.

Research for my thesis was divided into four parts. First, I used gold-standard systematic review methods to collate the state of current knowledge and assemble datasets from published studies of management practices forming part of a RA approach. My study area was temperate oceanic regions and I examined papers that considered the following RA approaches: reduced tillage intensity, cover cropping and ley-arable rotations in arable systems, and rotational grazing, multi-species or herbal leys, and agroforestry in pasture systems. Second, using the datasets obtained from these published studies, I conducted Bayesian meta-analyses of the impact of these practices on SOC and agricultural productivity (crop yields, herbage production, livestock growth). I found that adoption of rotational grazing and herbal leys can increase herbage dry matter production and livestock growth rates in pasture systems, but there is currently insufficient evidence to determine their impact on SOC. Conversely, my results identified clear potential for RA practices in arable systems to increase SOC but no evidence of a SOC-yield win-win. Third, using this finding I simulated adoption of these arable RA practices across Great Britain using the soil carbon model RothC, and found that in the UK there is the potential to mitigate 16-27% of current agricultural emissions through soil carbon sequestration. Finally, I conducted semi-structured interviews with industry representatives and livestock farmers in England to further understand advantages and disadvantages of implementing rotational grazing, herbal leys, ley-arable rotations and trees on farms. This final part of my research highlighted potential constraints on the ability to realise the theoretical advantages demonstrated from my earlier analyses.

Mmolotsi, R.M. (2004) Potential uses of red alder (Alnus rubra Bong) in silvopastoral systems. Ph.D. thesis. Bangor University.

Abstract
Studies on Alnus rubra, Bong. (Red alder) were carried out at Henfaes agroforestry experimental site, in North Wales to asses its potentials for use as a component of silvopastural systems. The specific objectives were: 1) to study the nitrogen fixing capability of red alder using nitrogen-15 natural abundance method, 2) to investigate the dynamics of fine roots and root nodules in red alder and their contribution to organic matter and nitrogen balance under silvopastoral systems, 3) to assess the timber quality of red alder wood in comparison with sycamore and finally, 4) to study the firewood and biomass energy value of red alder in comparison with sycamore. Literature reviewed showed that red alder, as a nitrogen fixing tree species, inputs large amount of nitrogen into natural ecosystems where it occurs. Reports also showed that it provides the much needed nitrogen when it is grown with other crops. The wood of red alder was reported to be good for furniture, studs, panelling and for making household utensils. There is, however, no information on the use of red alder in silvopasture. In the present study, using nitrogen-15 natural abundance method 815N values close to zero were recoded in red alder plant parts except in root nodules. This indicates that a large proportion of nitrogen in red alder was fixed from the atmosphere. The highest fixed nitrogen was measured in leaves (91 %) and the least (78%) in wood of red alder. Fixed nitrogen was higher during the summer and autumn (90% and 99%, respectively) than in winter and spring (85% and 64%, respectively). The 815N values in root nodules were enriched with 15N and this indicates that soil was the source of nitrogen for root nodules of red alder. Overall, it was estimated that 63.45 and 329.53 kg N ha-1 was fixed by red alder in the agroforestry and forestry systems, respectively. Live and dead fine root weight densities of red alder were 2700 and 5400 kg ha-1 and 360 and 790 kg ha-1 in agroforestry and forestry, respectively. Live and dead root nodule weight density of red alder yielded 880 and 520 kg ha-1 in agroforestry in 800 and 310 kg ha-' in forestry, respectively. The amount of organic matter potentially added to the soil due to senescent leaves and dead roots and root nodules was estimated at 4.0 and 9.1 t ha-' yf 1, in agroforestry and forestry, respectively. These results showed that red alder has a potential to improve and maintain soil fertility in silvopasture. Fine root length density was also found to differ significantly between agroforestry and forestry treatments, between seasons and between depths but there was no difference between distances from the tree base of red alder. Overall, the results of the distribution of fine root length density showed that red alder is compatible with pasture. Wood mechanical properties of red alder were found to be significantly different from that of sycamore. Sycamore yielded higher wood density (0.64 g cm-3), modulus of rupture (90.24 MPa) and compression strength (36.49 MPa) than red alder (0.49 g cm- , 73 .48 MPa and 32.13 MPa, respectively). However, modulus of elasticity was higher in red alder (7614.64 MPa) than in sycamore (7430.05 MPa), although it was not significantly different. Based on the results of wood properties of red alder it was concluded that red alder is medium strength tree species with potential for furniture manufacturing and for ordinary non-structural uses such as panelling and studs. Although the calorific value of sycamore was higher than red alder, red alder wood gave significantly higher fuel value index (1637.77) than sycamore (1480.60) due to very high ash content of sycamore. Thus, it was concluded that red alder has a potential to provide a better bioenergy than sycamore for heating homes and generating electricity. On the basis of the findings of the present study it was finally concluded that red alder is a suitable tree species for incorporation in silvopasture.

Nworji, J. (2017) Physical and bioeconomic analysis of ecosystem services from a silvopasture system. PhD. thesis. Bangor University.

Abstract
The aim of this study was to evaluate some of the physical and bioeconomic potentials of a silvopastoral agroforestry system with focus on the Henfaes Silvopastoral Systems Experimental Farm (SSEF) of Bangor University, North Wales.
The study reviewed research studies written on the SSEF from 1992 to 2012; assessed changes in pasture species composition and abundance since establishment; developed allometric equations for the estimation of aboveground biomass (AGB), carbon (C) stock and carbon dioxide (CO2) emission potentials of red alder (Alnus rubra Bong); studied the effect of tree/solar radiation on pasture productivity and quality; and conducted bioeconomic analysis to compare treeless pasture/livestock, forestry, and agroforestry scenarios.
Review of the research studies show that as far as can be determined 66 research studies were conducted on ecosystem services of the UK’s Silvopastoral National Network Experiment (SNNE) and temperate Europe during the period 1992 - 2012. These papers were sourced mainly from the Henfaes SSEF, the UK’s SNNE, other UK and, other European research sites. The studied ecosystem services dealt with provisioning services (40%), regulating services (13%), and supporting services (47%). The scientific domains addressed include timber or wood-fuel potential (20%), pasture/livestock management (20%), biodiversity (20%), carbon sequestration (13%), water management (15%), and soils (12%).
The response of pasture species to thinning varied. The percentage composition by weight of the sown species declined, while that of the grass weeds and the forb weeds increased slightly one year after thinning (2013 – 2014) compared to the adjacent open pastures. The change was not statistically significant. The understory pasture species composition, abundance and diversity changed significantly 20 years (1992 – 2012) after the establishment of the Silvopastoral National Network Experiment at Henfaes. Generally, pasture on the three red alder blocks was found to be largely grass weeds (46-48%) followed by forbs or broadleaf weeds while the sown species declined significantly.
In 2012, 20 years after field planting, the mean AGB were found to vary from 130 kg tree-1 (26 Mg ha-1) to 246 kg tree-1 (49 Mg ha-1) in poor form and good form red alder trees, respectively, based on a stocking density of 200 stems ha-1. Mean C stock was 65 kg C tree-1 (13 Mg C ha-1) in poor form trees and 123 kg C tree-1 (25 Mg C ha-1) in good form trees. Mean CO2 potential was 237 kg CO2 tree-1 (48 Mg CO2 ha-1) in poor form trees and 450 kg CO2 tree-1 (90 Mg CO2 ha-1) in good form trees.
Pasture productivity increased significantly with increasing solar transmission, and with increasing distance from each grazing exclusion cage to the nearest alder tree. Concentration and availability of CP, ADF, NDF and ME were greater in the with-leaves than in the without-leaves growing seasons in response to variation of photoperiod (the duration of sunshine/day length) in the United Kingdom.
The bioeconomic analysis considered three land-use plausible scenarios (‘forestry’, ‘pasture / livestock’ and ‘agroforestry’) and found that, in the absence of grants/subsidies, none were viable. However, application of grants/subsidies, at the baseline assumptions, revealed that forestry was the most viable option with the highest net present value and annual equivalent value, followed by pasture/livestock and agroforestry options.

With the global population approaching 8 billion, we face an ever-increasing challenge to make efficient use of land for production of resources such as food and timber, while conserving biodiversity and safeguarding the environment for future generations. The practice of farming with trees, or agroforestry, could be one strategy to help meet this challenge, on the premise that trees make use of the space above and below annual crops while benefitting biodiversity. This summary (based on Tom Staton's PhD thesis ‘Evaluating the effects of agroforestry versus arable systems on functional biodiversity and associated ecosystem services’ conducted via the University of Reading) aimed to test this. Data was collected from UK farms to explore the impact of in-field fruit and timber trees on biodiversity, services provided by biodiversity such as pollination and the natural regulation of pests, crop production, and farm income...

The FAO (2015) define agroforestry as a ‘collective name for land-use systems and technologies where woody perennials are deliberately used on the same land-management units as agricultural crops and/or animals’. The five main agroforestry practices that are recognised in the EU include riparian buffer strips, forest farming, home gardens and silvopastural and silvoarable systems (Mosqura-Losada et al. 2018). The objective of this thesis was to obtain a greater level of understanding of agroforestry and its role in improving marginal land, with particular focus on farmer awareness of this practice in the North-West region of Ireland. The thesis aimed to determine factors which may restrict the uptake of agroforestry, and identify actions that could be taken to increase involvement. Another objective was to compare land occupied by agroforestry with grassland. The survey distributed to farmers proved there is an appetite for agroforestry in the North-West of Ireland. However, work must be done by the government and the Department of Agriculture, Food and the Marine (DAFM) to incentivise this and make it attractive to those interested. More information must be made available as a lack of knowledge on the topic currently exists. Completing a soil sample analysis was very beneficial as it presented clear differences in the soils obtained from silvopasture and grassland sites. Silvopasture was superior in a number of areas including moisture content, porosity and earthworm populations. It was very evident that incorporating silvopasture into a grassland region can provide many benefits to the area. Inconclusive of all data obtained, it is apparent that the presence of silvopasture provides the area with many benefits in comparison to grassland swards. Farmers in the North-West are open to considering silvopasture as a method of land and productivity improvement but additional funding and incentives are required alongside information sources.

Agroforestry is a multifunctional land used system combining trees and shrubs with crops and/or animals on the same land management unit for optimising environmental, social, and productive benefits. The system is gaining recognition in the European Union (EU) as part of the solution to addressing the sustainability challenges of modern intensive agriculture and food production. This work conducts analysis of literature and expert interviews to provide a comparative perspective of the legal frameworks, economic incentives and legal barriers directed towards agroforestry in France, Ireland, Germany, and the United Kingdom (UK). Analysis of seventeen enabling policies shows that France is leading in promoting the mainstream application of agroforestry into farming practice. A recent momentum in economic support for agroforestry establishment and maintenance in France, Ireland, and parts of the UK is emerging. Despite the existence of enabling measures and subsidies in these countries, farmers wanting to implement agroforestry across the study region face various legal obstacles, mainly inadequate economic support, lack of administrative support, and poorly adapted and coordinated policies. To overcome these constraints, solutions will come from many actions such as improved national support for agroforestry research, better legal recognition, flexible incentive conditions based on improved economic evaluations of the public benefits of agroforestry systems, wider policy framing coordinated at the local, national, and international scales, with multi-actor involvement and a focus on agroforestry system interactions and connections rather than boundaries.

EC agricultural policy has changed focus from production maximisation to sustainable development and diversification during the last two decades. Ireland has an expansionist afforestation policy with an emphasis on increasing the area of broadleaf forest. However, annual planting targets set by the government in 1996 had not been realised by 2003. The majority of afforestation in Ireland is on private agricultural land. Irish farms predominantly rely on cattle-enterprises for their main source of income and reserve their most productive land for livestock. Standard practice in Ireland when afforesting with oak (Quercus robur L.) is to use bareroot planting stock. Planting stock produced using the Root Production Method ® (RPM) has greater root biomass than bareroot planting stock and is reported to have greater establishment success and growth. The EC Rural Development Regulation to be implemented in 2007 will support the establishment of agroforestry systems on agricultural land. To date, agroforestry has received little research interest in Ireland. A lowland silvopastoral alley system, comprising of wide-spaced Quercus robur L. and cattle grazing, was established in 2002 at the Teagasc Environment Research Centre, Co. Wexford, to compare the productivity and ecological consequences of such a system with pasture control; forestry with bareroot trees; forestry with RPM trees; silvopasture with bareroot trees; and silvopasture with RPM trees. Plots are 0.126 ha., 0.063 ha., and 0.315 ha. respectively for pasture, forestry and silvopasture treatments. Trees in silvopasture treatments are planted in an alley configuration, 2.5 m within the tree-rows and 10 m between tree-rows. The current study was conducted during the first three years after establishment. Height growth, stem diameter growth, and survival of bareroot and RPM oak planted in silvopasture and pure forestry were compared. Three years after planting, tree type (RPM v bareroot) and system (silvopasture v forestry) had no significant effect on tree height increment from planting or on stem diameter at the end of the experiment. The height:diameter ratio forestry trees was greater than trees in silvopasture. Bareroot and forestry trees had greater survival rates than RPM and silvopasture trees respectively. ii There was no significant difference in total tree biomass and volume between RPM and bareroot trees or between silvopasture and forestry trees. Root:shoot ratios were similar for RPM trees and bareroot trees and for silvopasture and forestry trees. RPM trees maintained their enhanced root systems three years after establishment with a greater volume and length of root branches than bareroot trees but showed no increase in stem biomass relative to the bareroot trees. In 2004, the pasture treatment had less total animal grazing days per hectare of pasture than either of the two silvopasture treatments (1423 v 1494 days per hectare of available pasture for pasture and silvopasture treatments respectively). There was no difference in pasture dry matter yield per hectare of available pasture, the mean number of animal grazing days per hectare of available pasture or mean animal liveweight gain per head between the three treatments in 2003 or 2004. Pasture had the greatest mean soil penetration resistance, followed by silvopasture and forestry. Grazed areas had greater mean penetration resistance than those where grazing was excluded. Penetration resistance at 10cm depth within silvopasture tree-rows was similar to forestry and that of the silvopasture alley was similar to pasture control. An edge effect of the silvopasture tree-rows appeared to influence penetration resistance immediately adjacent to the tree-rows within the silvopasture alleys, reducing penetration resistance relative to the centre of the alley. Similar results were found for soil moisture data. Carabid assemblages collected during this study generally agreed with known preferred habitats of many of the collected species. The greater total carabid species richness of the silvopastoral system relative to either the pastoral or forestry systems was not significant. There were no significant differences in carabid beetle species richness or diversity between grazed and ungrazed plots, although a positive trend with exclusion of grazing is intimated. A spreadsheet-based bio-economic model was constructed with which to assess the economic profitability of a lowland Quercus robur L./cattle alley silvopastoral system in Ireland relative to forestry and livestock grazing. Some fundamental data required to build the model was not available and had to be estimated. Therefore the iii results of the model are not definitive and conclusions related to silvopasture profitability relative to forestry and livestock grazing are not possible. However, assuming that trends predicted by the model are realised, the model has provided some useful results. To maximise profitability, future silvopastoral systems should have tree-rows more widely-spaced than the 10 m spacing of the system on which the model is based. The model also predicted that livestock grazing will be able to continue throughout the 150 year rotation of the silvopastoral system. Initial data regarding the performance of RPM Quercus robur relative to bareroot planting stock does not favour the Root Production Method ®. Bareroot trees have, generally, performed well. The use of RPM plants for afforestation projects on eutric cambisols and humic gleysols in lowland Ireland cannot be recommended with the current available knowledge. A silvopastoral system with cattle can be successfully managed in lowland Ireland for agricultural production. Soil compaction in silvopasture alleys is influenced by proximity of silvopasture tree-rows. The mechanism causing this effect is not fully understood. The Carabidae assemblage data that has been collected will be invaluable as a baseline dataset with which to compare future Carabidae diversity and abundance as the site matures. Assuming trends predicted by the model are realised, future silvopastoral systems should have tree-rows more widely spaced than the 10 m spacing of the Teagasc silvopastoral trial to maximise profitability. Agricultural policy has been increasing focussed on extensification and results from the model suggest that silvopasture would integrate well with such a policy