Sustainable livestock Intensification in Mediterranean agrosilvopastoral systems - PhD thesis defence

© A. Ripamonti, 2025 1PhD Thesis Defence – Alice Ripamonti – 30th May 2025 The current PDF contains the presentation delivered during the public PhD thesis defence of “Sustainable livestock intensification in Mediterranean agrosilvopastoral systems”, held on May 30, 2025, at the Department of Agriculture, Food, and Environment, University of Pisa. The PhD thesis is under a three-year embargo, as it includes unpublished data intended for future publication. However, this presentation primarily features data already published in peer-reviewed journals. For citation purposes, please refer to the corresponding published papers. Where unpublished data are presented, this is clearly indicated on the respective slides. Reuse of figures or data should cite the relevant published articles. © A. Ripamonti, 2025 © A. Ripamonti, 2025 Sustainable livestock intensification in Mediterranean agrosilvopastoral systems Alice Ripamonti XXXVII PhD Cycle PhD Thesis Defence Supervisors Prof. Marcello Mele Dr. Alberto Mantino Pisa, 30th May 2025 Photo credit: J. Goracci © A. Ripamonti, 2025 1. Introduction PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 Provision of 38% of global protein supply •Ritchie and Roser (2019) Animal food rich in essential nutrients (B12, vitamins A, D, iron, Amino Acids) •Beal et al. (2023) Ability to use low-opportunity-cost feed (crop residues, food by-products, grassland) •Gerber et al. (2015) Source of income, especially in rural area 53% of agricultural GHG emissions •FAO (2022) Food-feed competition, 40% of arable land used for feed production •Van Zanten et al. (2018) Risk of water eutrophication and acidification Pro Cons 4 The role of livestock in food systems PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 5PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Technical and economic Social Environmental Agricultural and sectorial policies Societal demand: consumers and citizens Global change Bernués et al. (2011) Context Farm and less-favoured areas in Mediterranean: synergies and trade-offs Pasture-based livestock farming systems © A. Ripamonti, 2025 6 Review: Make ruminants green again PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Sustainable Intensification Agroecology how can sustainable intensification and agroecology converge for a better future? Dumont et al. (2018) Nutrient use efficiency and technology System re-conception and ecosystem services © A. Ripamonti, 2025 7 Agroecology for adaptation to climate change PhD Thesis Defence – Alice Ripamonti – 30th May 2025 and resource depletion in the Mediterranean region. A review Aguilera et al. (2020) Crop Livestock Biodiversity management Increasing soil organic matter Renewable energy Extensive herds, diversification and local breeds Pasture and forage management Agroforestry © A. Ripamonti, 2025 SI can be reach through agricultural systems redesign at landscape levels 7 possible solutions identified 8 Intensification for redesigned PhD Thesis Defence – Alice Ripamonti – 30th May 2025 “The combination of agricultural processes in which production is maintained or increased, while environmental outcomes are enhanced” Pretty (2018) and sustainable agricultural systems © A. Ripamonti, 2025 ‘’the practices of deliberately integrating woody vegetation (trees or shrubs) with crop and/or animal systems to benefit from the resulting ecological and economic interaction‘’ 9 Agroforestry: ‘’a new name for an old practice‘’ PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Nair (1991); Burgess and Rosati (2018) Microclimate regulation and shade Carbon sequestration and storage Air and water purification Food, feed and fodder Protection from soil erosion Tiber and firewood Nutrient cycling and soil fertility Photo credit: A. Ripamonti (modified from Burgess and Rosati 2018) © A. Ripamonti, 2025 16 PRISMA Methodology PhD Thesis Defence – Alice Ripamonti – 30th May 2025 TITLE-ABS-KEY (agroforestry OR silvopast* OR agrosilvopast* OR tree OR shad*) AND (quality OR ndf OR “nutritive value” OR protein OR digestibility OR product* OR “chemical composition” OR yield OR “phenological cycle”) AND (forage OR grassland OR grass OR pasture OR pastureland) Research question definition Research query definition How does tree presence influence forage yield and nutritive value? AND (LIMIT-TO(DOCTYPE, “ar”) OR LIMIT-TO(DOCTYPE, “re”) OR LIMIT-TO(DOCTYPE, “ch”) OR LIMIT-TO(DOCTYPE, “bk”)) AND (LIMIT-TO(SUB-JAREA, “AGRI”)) AND (LIMITTO(LANGUAGE, “English”)) AND (LIMIT-TO(SRC-TYPE, “j”) Automatic filters Picture source: www.distillersr.com Page et al. (2021) © A. Ripamonti, 2025 131 Original articles; 5 Reviews; 4 Metanalysis 17 Identification and screening process PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Studies included in the report: 140 Materials and Method screened: 364 Abstract screened: 513 Titles screened: 6,126 Scopus results: 11,028 (automatically excluded through filleters: 4,902) Studies that consider forage yields solely were excluded Level of the experiment Location of trial site(s) Forage botanical family Soil characteristics Limiting factors controlled Experimental design Forage yield and quality Köppen-Geiger Climate (Tropical, Dry, Temperate, Continetnal) Information retrieved © A. Ripamonti, 2025 18 Study level classification and findings PhD Thesis Defence – Alice Ripamonti – 30th May 2025 8 110 13 Number of records Pot Field Landscape Screening for shade adaptability and resilience to water scarcity Understanding of tree competition with livestock and grass to evaluate whole agricultural system productivity Improving forage yield and quality by testing forage mixtures and different tree planting design © A. Ripamonti, 2025 19 Climate and location classification and findings PhD Thesis Defence – Alice Ripamonti – 30th May 2025 43 19 48 17 4 Number of records A - Tropical B - Arid CTemperate D - Continental n.a. South America, focus on: (i) few key grass species; (ii) profitability and animal performance; (iii) low presence of grass-legume mixture © A. Ripamonti, 2025 20 Controlled factors classification and findings PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Conducting studies in real condition is preferable to evaluate tree-grass interactions but artificial shade is costeffective. No nitrogen limitations to isolate light competition effect. Investigate N dynamics, in relation to light reduction. Explore the trade-offs between reduced light availability and increased water use efficiency. Analysis of tree-grass-animals interactions to optimize grazing management and improve pasture characteristics Light Nutrients Water Grazing 110 on field16 110 on field42 110 on field9 110 on field45 AFS: Agroforestry systems © A. Ripamonti, 2025 21 Botanical family classification and findings PhD Thesis Defence – Alice Ripamonti – 30th May 2025 72 7 52 Number of records Grass Legume Mixture Shade has higher detrimental effect on legumes rather than grasses © A. Ripamonti, 2025 On-farm trial: sampling protocols and data collection Grassland and grazing management Animal welfare and heat stress Data collection for Quantification of ecosystem services PhD Thesis Defence – Alice Ripamonti – 30th May 2025 23 Trial site description PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Location: Maremma area, South of Tuscany Climate: Medium average temperature 15.6°C Medium average precipitation 840 mm Surface: 3.69 ha of temporary grassland 3.31 ha of Turkey oak forest Animal: Steers and heifers of Maremmana breed Treatments: Pastoral PA (only grassland) Silvopastoral SP (grassland and forest) Feed supplements: Oat-vetch hay produced on farm (ad libitum) Mixed grain flour produced on farm (1% of body weight) Photo credit: Ripamonti et al., 2023 24 Animal characteristics PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Year 2021 2022 System Pastoral system Silvopastoral system Pastoral system Silvopastoral system Total head 29 21 20 20 Heifers 16 13 12 13 Steers 13 8 8 7 Average age (day) 314 ± 46 331 ± 66 331 ± 34 330 ± 38 Average weight (kg) 298 ± 57 271 ± 55 286 ± 58 276 ± 57 Stocking rate grassland (LU ha −1 d−1 ) 3.38 2.96 2.15 2.69 Stocking rate forest (LU ha−1 d−1)-0.46 -0.50 Photo credit: A. Ripamonti 𝐻𝑒𝑟𝑏𝑎𝑔𝑒 𝐴𝑙𝑙𝑜𝑤𝑎𝑛𝑐𝑒 𝑔 𝐷𝑀 𝑘𝑔 𝐵𝑊−1𝑑−1 𝑃𝑟𝑒𝐻𝑀 𝑔 𝐷𝑀 𝑚−2 𝐷 𝑑 + 𝐷𝐻𝐺 𝑔 𝐷𝑀 𝑚−2𝑑−1 × 𝐴 𝑚2 BW 𝑘𝑔 𝑃𝑜𝑡𝑒𝑛𝑡𝑖𝑎𝑙 𝐻𝑒𝑟𝑏𝑎𝑔𝑒 𝐼𝑛𝑡𝑎𝑘𝑒 𝑔 𝐷𝑀 𝑑−1 𝐸𝐶𝐻𝑀 𝑔 𝑚−2 − 𝑃𝑜𝑠𝑡𝐻𝑀 𝑔 𝑚−2 ×𝐴 𝑚2 𝐷 𝑑 ×BW 𝑘𝑔 Temperature Humidity Index and Black Globe Humidity Index Buffington et al. (1981); Mader et al. (2004) Average daily gain, Glucose, Serum cortisol and Hair cortisol Belhadj Slimen et al. (2016); Heimbürge et al. (2019); Ghassemi Nejad et al. (2022) Nutritional value, Herbage Allowance and Potential Herbage Intake Undi et al. (2008); Mantino et al. (2021) 25 On field data collection PhD Thesis Defence – Alice Ripamonti – 30th May 2025 MicroclimatePasture Animal HM: herbage mass; DHG: daily herbage growth; EC: Exclusion cage BW: Body weight; A: Area; D: day p-value Herbage mass Nutritive Value Net Energy for growth Herbage Allowance Herbage Intake Pregrazing Post - grazing CP NDF (S) 0.6492 0.658 0.0593 0.073 0.649 0.189 0.57 Period (GP) 0.0193 0.006 <.0001 <.0001 <.0001 <.0001 0.002 x GP 0.8831 0.16 0.526 0.067 0.0336 0.3612 0.61 32 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 - Crude protein (p < .0001) - Neutral Detergent Fibre (p < .0001) - Pre-grazing herbage mass (p = 0.019) - Post-grazing herbage mass (p = 0.006) Significant differences among grazing periods in: 2^ round: lower pre-grazed biomass but not post-grazed Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system 2^ round: higher NDF content 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 Herbage biomass and nutritive value R Core Team; Pinheiro et al. (2021) p-value Herbage mass Nutritive Value Net Energy for growth Herbage Allowance Herbage Intake Pregrazing Post - grazing CP NDF (S) 0.6492 0.658 0.0593 0.073 0.649 0.189 0.57 Period (GP) 0.0193 0.006 <.0001 <.0001 <.0001 <.0001 0.002 x GP 0.8831 0.16 0.526 0.067 0.0336 0.3612 0.61 33 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 - Crude protein (p < .0001) - Neutral Detergent Fibre (p < .0001) - Pre-grazing herbage mass (p = 0.019) - Post-grazing herbage mass (p = 0.006) Significant differences among grazing periods in: 2^ round: lower pre-grazed biomass but not post-grazed Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system 2^ round: higher NDF content 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 Herbage biomass and nutritive value R Core Team; Pinheiro et al. (2021) Overall low pasture productivity and quality Porqueddu et al. (2016) Drought condition (-24% of measured rainfall vs. long-term data) accentuate the effect of the low soil fertility Seligman and Van Keulen (1989) Lack of proper synchronization between grazing and the phenological phase Undersander et al. (2002) Lack of rainfall in spring (-24% of measured rainfall) Fast seasonal advancement Norton et al. (2016) 34 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 With supplements: - Herbage allowance (p = 0.18) - Potential herbage intake (p = 0.57) No significant differences between systems in: - Herbage allowance (p = <.0001) - Potential herbage intake (p = 0.002) Significant differences among grazing periods in: Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 (Data not shown) Animal herbage and feed intake R Core Team; Pinheiro et al. (2021) 35 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 With supplements: - Herbage allowance (p = 0.18) - Potential herbage intake (p = 0.57) No significant differences between systems in: Cattle in pastoral and silvopastoral had the same level of herbage intake - Herbage allowance (p = <.0001) - Potential herbage intake (p = 0.002) Significant differences among grazing periods in: Consequence of lower herbage biomass in second round Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 (Data not shown) Animal herbage and feed intake R Core Team; Pinheiro et al. (2021) 36 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 With supplements: - Herbage allowance (p = 0.18) - Potential herbage intake (p = 0.57) No significant differences between systems in: Cattle in pastoral and silvopastoral had the same level of herbage intake - Herbage allowance (p = <.0001) - Potential herbage intake (p = 0.002) Significant differences among grazing periods in: Consequence of lower herbage biomass in second round Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 (Data not shown) Potential herbage intake less than 1% of the body weight, lower than literature Significantly affected by low herbage allowance Dougherty et al. (1989); Wilkinson et al. (2019) Preferences of concentrate feed over herbage Herbage biomass and nutritive value not sufficient to satisfy animal requirements Animal herbage and feed intake R Core Team; Pinheiro et al. (2021) 37 Animal weight gain and hair cortisol PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑻𝒋 + 𝑺 ∙ 𝑻 𝒊𝒋 + 𝑨𝒌|𝑺 + 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested within system Significant differences between systems in: Significant interaction of system and time in: - Body Weight (p = 0.001) - Hair cortisol accumulation (p = 0.04) - Serum cortisol (p = 0.18) No significant differences between systems or among time in: R Core Team; Pinheiro et al. (2021) 38 Animal weight gain and hair cortisol PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑻𝒋 + 𝑺 ∙ 𝑻 𝒊𝒋 + 𝑨𝒌|𝑺 + 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested within system Significant differences between systems in: Significant interaction of system and time in: - Body Weight (p = 0.001) - Hair cortisol accumulation (p = 0.04) - Serum cortisol (p = 0.18) Cattle started with any weight difference but ended with a significant difference No significant differences between systems or among time in: R Core Team; Pinheiro et al. (2021) 39 Animal weight gain and hair cortisol PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑻𝒋 + 𝑺 ∙ 𝑻 𝒊𝒋 + 𝑨𝒌|𝑺 + 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested within system Significant differences between systems in: Significant interaction of system and time in: - Body Weight (p = 0.001) - Hair cortisol accumulation (p = 0.04) - Serum cortisol (p = 0.18) Cattle started with any weight difference but ended with a significant difference No significant differences between systems or among time in: R Core Team; Pinheiro et al. (2021) Lower and slower growth in silvopastoral system (SP: 1.02 kg d-1 vs. PA: 1.20 kg d-1) Due to possible higher energy expenditure for walking activity in forest area Vandermeulen et al. (2018b); de Oliveira et al (2021) Possible reduction of economic income (slower growth, higher feed intake) No evidence of animal heat stress but possible handling stress For long chronic stress preferable using hair cortisol Bristow and Holmes (2007); Ghassemi Nejad et al. (2019) (2020) 40 Effect on soil cover PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝑁𝐷𝑉𝐼 = 𝑁𝐼𝑅 −𝑅𝑒𝑑 𝑁𝐼𝑅 +𝑅𝑒𝑑 •NDVI = Normalized Difference Vegetation Index •NIR = Near Infrared Image analysis: density chart of pixels subdivided in deciles analysis 41 Effect on soil cover PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝑁𝐷𝑉𝐼 = 𝑁𝐼𝑅 −𝑅𝑒𝑑 𝑁𝐼𝑅 +𝑅𝑒𝑑 •NDVI = Normalized Difference Vegetation Index •NIR = Near Infrared Image analysis: density chart of pixels subdivided in deciles analysis © A. Ripamonti, 2025 48 Potential heat stress classification PhD Thesis Defence – Alice Ripamonti – 30th May 2025 BGHI Scale > 79 - Emergency 75 - 79 - Alert < 75 - Minimal risk 𝑩𝑮𝑯𝑰 = 𝑩𝑮𝑻 + 𝟎.𝟑𝟔 ∗ 𝑫𝑷𝑻 + 𝟒𝟏.𝟓 Where: •BGHI = Black Globe Humidity Index •BGT = Black Globe Temperature (°T) •DPT = Dew Point Temperature (°T) Forest has lower BGHI values in 2021 and 2022 (Significant according to the Chi-square test) Higher potential heat stress in June and July 2022 Lemes et al. (2021) Trees mitigate potential of heat stress until a certain threshold Forest might limit wind circulation Buffington et al. (1981) © A. Ripamonti, 2025 49 Spring herbage intake PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 - Herbage allowance (p = <.0001) - Potential herbage intake (p = 0.07) Significant differences between systems in 2022: Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system R Core Team; Pinheiro et al. (2021) In general, higher herbage allowance compared to 2022 Cattle in pastoral had higher herbage allowance in 2022 compared to silvopastoral © A. Ripamonti, 2025 50 Spring herbage intake PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒚𝒊𝒋𝒌 = 𝝁 + 𝑺𝒊 + 𝑮𝑷𝒋 + 𝑺 ∙ 𝑮𝑷 𝒊𝒋 + 𝑷𝒌|𝑺 + 𝜺𝒊𝒋𝒌 - Herbage allowance (p = <.0001) - Potential herbage intake (p = 0.07) Significant differences between systems in 2022: Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •GP = Grazing Period – Fixed (n = 6) •P = Paddock – Random nested within system R Core Team; Pinheiro et al. (2021) In general, higher herbage allowance compared to 2022 Cattle in pastoral had higher herbage allowance in 2022 compared to silvopastoral Grazing trial delayed and to beneficial rainfalls before the start date In 2021 the higher stocking rate used in pastoral system allows to equalize the differences between system paddocks © A. Ripamonti, 2025 51 Animal productivity: average daily gain PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒂𝒗𝒆𝒓𝒂𝒈𝒆 𝒅𝒂𝒊𝒍𝒚 𝒈𝒂𝒊𝒏𝒊𝒋𝒌 = 𝝁+𝑻𝒊+𝑺𝒋+ 𝑻∙𝑺 𝒊𝒋 +𝑨𝒌ቚ𝑺 +𝒔𝒆𝒙 + 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested with system Significant interaction of system and time in: -2021 (p = 0.0001) -2022 (p = <.0001) Growth peak at the end of spring, bigger for pastoral system Lower reduction of average daily gain in summer months in silvopastoral system R Core Team; Pinheiro et al. (2021) © A. Ripamonti, 2025 52 Animal productivity: average daily gain PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒂𝒗𝒆𝒓𝒂𝒈𝒆 𝒅𝒂𝒊𝒍𝒚 𝒈𝒂𝒊𝒏𝒊𝒋𝒌 = 𝝁+𝑻𝒊+𝑺𝒋+ 𝑻∙𝑺 𝒊𝒋 +𝑨𝒌ቚ𝑺 +𝒔𝒆𝒙 + 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested with system Significant interaction of system and time in: -2021 (p = 0.0001) -2022 (p = <.0001) Growth peak at the end of spring, bigger for pastoral system Lower reduction of average daily gain in summer months in silvopastoral system Confined rotational grazing reduced energy expenditure without potential heat stress Positive effect of trees on animal growth response to heat stress until a certain threshold R Core Team; Pinheiro et al. (2021) © A. Ripamonti, 2025 53 Animal welfare: hair cortisol accumulation PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒄𝒐𝒓𝒕𝒊𝒔𝒐𝒍𝒊𝒋𝒌 = 𝝁+𝑻𝒊+𝑺𝒋+ 𝑻∙𝑺 𝒊𝒋 +𝑨𝒌ቚ𝑺 + 𝒂𝒅𝒈+ 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested Significant interaction of system and time in: -2021 (p = 0.04) -2022 (p = <0.001) Higher absolute values in 2022 R Core Team; Pinheiro et al. (2021) ADG: Average daily gain © A. Ripamonti, 2025 54 Animal welfare: hair cortisol accumulation PhD Thesis Defence – Alice Ripamonti – 30th May 2025 𝒄𝒐𝒓𝒕𝒊𝒔𝒐𝒍𝒊𝒋𝒌 = 𝝁+𝑻𝒊+𝑺𝒋+ 𝑻∙𝑺 𝒊𝒋 +𝑨𝒌ቚ𝑺 + 𝒂𝒅𝒈+ 𝜺𝒊𝒋𝒌 Linear Mixed-Effect Model: •S = System – Fixed (n = 2) •T = Time – Fixed (n = 3) •A = Animal – Random nested Significant interaction of system and time in: -2021 (p = 0.04) -2022 (p = <0.001) Higher absolute values in 2022 In 2022: because of higher BGHI, higher cortisol levels for both systems Lower hair cortisol accumulation in silvopastoral system Increased resilience of livestock to heat stress and increased animal welfare Broom et al. (2013); Lemes et al. (2021) R Core Team; Pinheiro et al. (2021) ADG: Average daily gain © A. Ripamonti, 2025 5. The use of Yield-SAFE to model animal heat indices on pastoral and forest land under two future climate scenario Quantification of ecosystem services Which approaches can be used to quantify ecosystem services in agroforestry system? PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 56 Background PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Characteristics of Yield-SAFE: Biophysical: ‘simulation of a biological system using mathematical formalizations of the physical properties’ Dynamic: ‘predict how system unfolds with the passage of time’ Aim of Yield-SAFE: Simulate the development, growth and productivity of the tree and crop over the length of a tree rotation Outputs of YieldSAFE: Predictions of tree and crop yield. Used for financial and economic analyses and ecosystem services Yield-SAFE data requirement: Daily temperature, radiation and precipitation, planting densities, initial biomasses of tree and crop species, and soil parameters Graves et al. (2010) © A. Ripamonti, 2025 Climate simulations were conducted using the Regional Atmospheric Climate Model (RACMO) Jacob et al. (2014); Vautard et al. (2021) The model generated meteorological variables for both historical and future periods: -Baseline scenario: 1953–2023 -Future scenarios: 2030–2100 -RCP4.5 (moderate emissions scenario) -RCP8.5 (high emissions scenario) The simulated data were corrected using case-study historical data to improve accuracy. Cannon et al. (2015) 57 Development of future climate scenarios PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 64 Tree density and stocking rate: results and limits PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Need to use a BGT equation and BGHI equation more sensitive to changes in tree density, solar radiation and temperature variations Need to improve the estimation of stocking rate considering seasonal changes in herbage quality and grassland regrowth to simulate the effect of grazing. © A. Ripamonti, 2025 6. Unlocking the potential of infrared thermography: challenges and opportunities in monitoring animal temperature in extensive systems Application of technology Are there technologies applicable in extensive livestock system to monitor animal behaviour and welfare? PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 66 Background PhD Thesis Defence – Alice Ripamonti – 30th May 2025 †Crea Global positioning systems Tri-axis accelerometer Remote sensing Unnamed aerial vehicles Decision support systems Virtual fencing Infrared thermography Main limits to technology spread Costs Battery duration Data analysis Farmers’ age Odintsov Vaintrub et al. (2021); Aquilani et al. (2022) Photo created with ChatGPT © A. Ripamonti, 2025 1. Instrument: Infrared camera FLIR series E75 2. Data collection: three consecutive days per month (June, July, August and September). Twice a day: 9.30-11.30; 16.0018.00 3. Distance from animal: between 5 and 10 meters 4. Number of animals involved: at least 20 animal 5. Identification of animal: with ID ear tag, animal ID associated with photo ID 6. Part of the body to point: left flank, if possible, photo of the entire body. If possible bigger images of the face (forehead and eye) 7. Position of the instrument: if possible infrared camera positioned perpendicular to the left antimer of the animal. 8. Infrared camera parameters: pre-set parameters during data collection oEnvironmental temperature: 20% oRelative Humidity: 50% oReflected temperature: 20°C oEmissivity: 0.95 67 Protocols used on beef cows in summer 2022 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 1. Instrument: Infrared camera FLIR series E75 2. Data collection: three consecutive days per month (June, July, August and September). Twice a day: 9.30-11.30; 16.0018.00 3. Distance from animal: between 5 and 10 meters 4. Number of animals involved: at least 20 animal 5. Identification of animal: with ID ear tag, animal ID associated with photo ID 6. Part of the body to point: left flank, if possible, photo of the entire body. If possible bigger images of the face (forehead and eye) 7. Position of the instrument: if possible infrared camera positioned perpendicular to the left antimer of the animal. 8. Infrared camera parameters: pre-set parameters during data collection oEnvironmental temperature: 20% oRelative Humidity: 50% oReflected temperature: 20°C oEmissivity: 0.95 68 Protocols used on beef cows in summer 2022 PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Missing animal measurement (i.e. rectal temperature, painting score) Operator/researcher’s presence might influence animal natural behaviour Overheating of the IRT camera, especially during summer © A. Ripamonti, 2025 69 Images and data analysis PhD Thesis Defence – Alice Ripamonti – 30th May 2025 1. Software: FLIR Research Studio (2024 © Teledyne FLIR LLC) 2. Picture processing: eight body areas were defined manually (Kotrba et al. 2007): (i) neck, (ii) dewlap, (iii) trunk, (iv) body forepart, (v) barrel, (vi) body hind part, (vii) forelimb, and (viii) rear limb 3. For each region, the following parameters were calculated: (i) minimum, (ii) mean, and (iii) maximum temperature; (iv) standard deviation; (v) number of pixels. 4. Each image was associated with: (i) month, (ii) day, and (iii) hour in which the picture was taken; (iv) subjected animal and (v) distance; measured (vi) air temperature, (vii) relative humidity, (viii) temperature of dew point, (ix) black globe temperature; and lastly thermal indices: (x) THI, and (xi) BGHI 5. During image processing: (i) air temperature, (ii) relative humidity, (iii) distance from the animal, and (iv) emissivity (set as 0.98) were adjusted for each image © A. Ripamonti, 2025 70 Images and data analysis PhD Thesis Defence – Alice Ripamonti – 30th May 2025 1. Software: FLIR Research Studio (2024 © Teledyne FLIR LLC) 2. Picture processing: eight body areas were defined manually (Kotrba et al. 2007): (i) neck, (ii) dewlap, (iii) trunk, (iv) body forepart, (v) barrel, (vi) body hind part, (vii) forelimb, and (viii) rear limb 3. For each region, the following parameters were calculated: (i) minimum, (ii) mean, and (iii) maximum temperature; (iv) standard deviation; (v) number of pixels. 4. Each image was associated with: (i) month, (ii) day, and (iii) hour in which the picture was taken; (iv) subjected animal and (v) distance; measured (vi) air temperature, (vii) relative humidity, (viii) temperature of dew point, (ix) black globe temperature; and lastly thermal indices: (x) THI, and (xi) BGHI 5. During image processing: (i) air temperature, (ii) relative humidity, (iii) distance from the animal, and (iv) emissivity (set as 0.98) were adjusted for each image Methodology as a starting point for standardize protocols to improve consistency of thermal data collection Image and data processing still manual: high workload and time consuming © A. Ripamonti, 2025 6. Conclusions PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 Pay attention to the agroforestry system design Microclimate improvements increase the resilience to climate change Strategic management for long-term sustainability including the use of fodder trees To evaluate animal welfare integrated approaches are needed Agroforestry modelling require improvements to adapt to agrosilvopastoral systems The use of technology is feasible but still limited to research purposes 72 In a nutshell PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 Diversification Local breeds Grazing management 73 Winning agroecological strategies used PhD Thesis Defence – Alice Ripamonti – 30th May 2025 Sustainable Intensification Agroecology Agroforestry Contribute to increase system resilience and sustainability Photo credit: J. Goracci © A. Ripamonti, 2025 Cannon AJ, Sobie SR and Murdock TQ (2015) Bias correction of simulated precipitation by quantile mapping: How well do methods preserve relative changes in quantiles and extremes? Journal of Climate, 28:6938-6959. https://doi:10.1175/JCLI-D-14-00754.1 de Oliveira CC, de Almeida RG, Karvatte Junior N, et al (2021) Daytime ingestive behaviour of grazing heifers under tropical silvopastoral systems: Responses to shade and graz-ing management. Appl Anim Behav Sci 240:105360. https://doi.org/10.1016/j.appla-nim.2021.105360 Dougherty C, Lauriault L, Cornelius P et al (1989) Herbage allowance and intake of cattle. J Agric Sci 112(3):395–401. https://doi.org/10.1017/S0021859600085841 Dumont B, Groot JCJ, Tichit M (2018) Review: Make ruminants green again - how can sus-tainable intensification and agroecology converge for a better future?. Animal 12:S210–S219. https://doi.org/10.1017/S1751731118001350 FAO (2022) FAOSTAT - Emissions totals. Available online at: https://www.fao.org/fao-stat/en/#data/GT Foley JA, DeFries R, Asner GP, et al (2005) Global consequences of land use. Science (80- ) 309:570–574. https://doi.org/10.1126/science.1111772 Gerber PJ, Mottet A, Opio CI, et al (2015) Environmental impacts of beef production: Review of challenges and perspectives for durability. Meat Sci 109:2–12. https://doi.org/10.1016/j.meatsci.2015.05.013 Ghassemi Nejad J, Ghaffari MH, Ataallahi M, et al (2022) Stress Concepts and Applications in Various Matrices with a Focus on Hair Cortisol and Analytical Methods. Animals 12:3096. https://doi:10.3390/ani12223096 Ghassemi Nejad J, Lee B-H, Kim J-Y et al (2019) Comparing hair cortisol concentrations from various body sites and serum cortisol in Holstein lactating cows and heifers during thermal comfort zone. J Vet Behav 30:92–95. https://doi.org/10.1016/j.jveb.2018.12.\007 Ghassemi Nejad J, Park K-H, Forghani F et al (2020) Measuring hair and blood cortisol in sheep and dairy cattle using RIA and ELISA assay: a comparison. Biol Rhythm Res 51:887–897. https://doi.org/10.1080/09291016.2019.16113 35 Graves AR, Burgess PJ, Palma J, et al (2010) Implementation and calibration of the parameter-sparse Yield-SAFE model to predict production and land equivalent ratio in mixed tree and crop systems under two contrasting production situations in Europe. Ecol Modell 221:1744–1756. https://doi.org/10.1016/j.ecolmodel.2010.03.008 Heimbürge S, Kanitz E, Otten W (2019) The use of hair cortisol for the assessment of stress in animals. Gen Comp Endocrinol 270:10–17. https://doi.org/10.1016/j.ygcen.2018.09.016 Heimburge S, Kanitz E, Tuchscherer A, Otten W (2020) Within a hair’s breadth—factors influ-encing hair cortisol levels in pigs and cattle. Gen Comp Endocrinol 288:113359. https://doi.org/10.1016/j.ygcen.2019.113359 80 References PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 Hoffmann I, From T, Boerma D (2014) Ecosystem services provided by livestock species and breeds, with special consideration to the contributions of small-scale livestock keep-ers and pastoralists. Comm Genet Resour Food Agric 158. Available online at: https://openknowledge.fao.org/server/api/core/bitstreams/d24ef49d-0df2-465a8bef-a0acd0f00236/content Jacob D, Petersen J, Eggert B, Alias A, Christensen OB, Bouwer LM, Braun A, et al. (2014) EURO-CORDEX: New high-resolution climate change projections for European impact research. Regional Environmental Change 14(2):563-578. jian V, Bhatta R, Gaughan JB, et al (2018) Review: Adaptation of animals to heat stress. An-imal 12:S431–S444. https://doi.org/10.1017/S1751731118001945 Kotrba, R., Knížková, I., Kunc, P., & Bartoš, L. (2007). Comparison between the coat tem-perature of the eland and dairy cattle by infrared thermography. Journal of Thermal Biology, 32(6), 355-359. https://doi.org/10.1016/j.jtherbio.2007.05.006 Lemaire G, Franzluebbers A, de Faccio Carvalho PC, Dedieu B (2014) Integrated crop-live-stock systems: strategies to achieve synergy between agricultural production and en-vironmental quality. Agric. Ecosyst. Environ. 190:4–8. https://doi.org/10.1016/j.agee.2013.08.009 Mader TL, Davis MS, Brown-Brandl T (2006) Environmental factors influencing heat stress in feedlot cattle. J Anim Sci 84:712–719. https://doi.org/10.2527/2006.843712x Mantino A, Cappucci A, Annecchini F et al (2021) An on-farm rotational grazing trial: restrict-ing access time to pasture did not affect the productivity of a dairy sheep flock in spring. Ital J Agron 16:1. https://doi.org/10.4081/ija.2020.1711 mes AP, Garcia AR, Pezzopane JRM, et al (2021) Silvopastoral system is an alternative to improve animal welfare and productive performance in meat production systems. Sci Rep 11:14092. https://doi.org/10.1038/s41598-021-93609-7 Millenium Ecosystem Assessment (2005) Ecosystem and Human Well-being: Synthesis. Island Press, Washington DC. Available online at: https://www.millenniumassessment.org/documents/document.356.aspx.pdf Nair PKR (1991) State-of-the-art of agroforestry systems. For. Ecol. Manag. 45:5-29. https://doi.org/10.1016/0378-1127(91)90203-8 Norton MR, Malinowski DP, Volaire F (2016) Plant drought survival under climate change and strategies to improve perennial grasses. A review. Agron Sustain Dev. https://doi.org/10.1007/s13593-016-0362-1 Odintsov Vaintrub M, Levit H, Chincarini M, et al (2021) Review: Precision livestock farming, automats and new technologies: possible applications in extensive dairy sheep farm-ing. Animal 15:100143. https://doi.org/10.1016/j.animal.2020.100143 81 References PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 Page MJ, McKenzie JE, Bossuyt PM, et al (2021) The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 29:372:n71. https://doi.org/10.1136/bmj.n71 Porqueddu C, Ates S, Louhaichi M et al (2016) Grasslands in “Old World” and “New World” Mediterranean-climate zones: past trends, current status and future research priori-ties. Grass Forage Sci 71:1–35. https://doi.org/10.1111/gfs.12212 Pretty J (2018) Intensification for redesigned and sustainable agricultural systems. Science (80- ) 362:6417. https://doi.org/10.1126/science.aav0294 Pulina G, Acciaro M, Atzori AS, et al (2021) Animal board invited review – Beef for future: technologies for a sustainable and profitable beef industry. Animal 15:100358. https://doi.org/10.1016/j.animal.2021.100358 Rao I, Peters M, Castro A, et al (2015) LivestockPlus – the sustainable intensification of forage-based agricultural systems to improve livelihoods and ecosystem services in the tropics. Trop Grassl 3:59–82. https://doi.org/10.17138/TGFT(3)59-82 Ritchie H, Roser M (2019) Half of the world’s habitable land is used for agriculture. In: Our World Data. https://ourworldindata.org/global-land-for-agriculture Sanderman J, Reseigh J, Wurst M, Young M (2015) Impacts of rotational grazing on soil carbon in native grass-based pastures in Southern Australia. PLoS ONE. https://doi.org/10. 4225/08/53B0EC7D91D4B Seligman NG, Van Keulen H (1989) Herbage production of a Mediterranean grassland in re-lation to soil depth, rainfall and nitrogen nutrition: a simulation study. Ecol Model 47:303–311. https://doi.org/10.1016/0304-3800(89)90008-2 Undersander DJ, Albert B, Cosgrove D, et al (2002) Pastures for profit: a guide to rotational grazing (A3529), p 1–38 Undi M, Wilson C, Ominski KH, Wittenberg KM (2008) Comparison of techniques for estima-tion of forage dry matter intake by grazing beef cattle. Can J Anim Sci 88:693–701. https://doi.org/10.4141/CJAS08041 Van Zanten HHE, Herrero M, Van Hal O, et al (2018) Defining a land boundary for sustainable livestock consumption. Glob Chang Biol 24:4185–4194. https://doi.org/10.1111/gcb.14321 Vandermeulen S, Ramírez-Restrepo CA, Beckers Y, et al (2018) Agroforestry for ruminants: A review of trees and shrubs as fodder in silvopastoral temperate and tropical produc-tion systems. Anim Prod Sci 58:767–777. https://doi.org/10.1071/AN16434 Baker 2023 Vandermeulen S, Ramirez-Restrepo CA, Marche C et al (2018) Behaviour and browse species selectivity of heifers grazing in a temperate silvopastoral system. Agrofor Syst 92:705–716. https://doi.org/10.1007/s10457-016-0041-x 82 References PhD Thesis Defence – Alice Ripamonti – 30th May 2025 © A. Ripamonti, 2025 Vautard R, Kadygrov N, Iles C, Boberg F, Buonomo E, Bülow, K, et al. (2021) Evaluation of the large EURO-CORDEX regional climate model ensemble. Journal of Geophysical Research: Atmospheres, 126, e2019JD032344. https://doi.org/10.1029/2019JD032344 Wilkinson JM, Lee MRF, Rivero MJ, Chamberlain AT (2020) Some challenges and opportuni-ties for grazing dairy cows on temperate pastures. Grass Forage Sci 75:1–17. https://doi.org/10.1111/gfs.12458 PUBLISHED PAPERS •Chapter 2: Ripamonti A, Finocchi M, Pulina A et al. Agroforestry Systems 99:110 (2025) https://doi.org/10.1007/s10457-025-01214-8 •Chapter 3: Ripamonti A, Mantino A, Annecchini F et al. Agro-forestry Systems 97:1071–1086 (2023). https://doi.org/10.1007/s10457-023-00848-w •Chapter 4: Ripamonti A, Foggi G, Mantino A, et al. Animal. 19(3):101425 (2025). https://doi.org/10.1016/j.animal.2025.101425 83 References PhD Thesis Defence – Alice Ripamonti – 30th May 2025