Comparative Impacts of Agroforestry Practices on Yield Across Climates

Thesis Abstract

Agroforestry systems are increasingly promoted to enhance yield stability and resilience in the face of climate variability, yet comparative evidence across diverse climatic zones remains fragmented, limiting scalable policy and practice recommendations. This study addresses the problem by evaluating how different agroforestry configurations influence crop and tree yields across contrasting climates, thereby advancing understanding of context-dependent performance and informing climate-smart agricultural strategies. The aim is to quantify and compare yield outcomes of traditional, improved, and integrated agroforestry arrangements across temperate, tropical, and arid environments, and to identify the mechanisms driving yield differentials under climate stress. Specific objectives are to (i) characterize agroforestry configurations and management practices in each climate zone, (ii) estimate yield effects of shade, root competition, nutrient fluxes, and microclimatic modification using robust statistical models, (iii) test the moderating role of climatic variables (precipitation, temperature, and seasonality) on yield outcomes, and (iv) generate evidence-based guidelines for optimizing agroforestry designs under climate change. A mixed-methods, cross-sectional research design was employed. The study sampled 60 farms per climate zone (total N = 180) selected through stratified random sampling to represent three predominant agroforestry types boundary/perimeter agroforestry, alley-cropping, and relay intercropping. Primary data were collected via structured farmer surveys (n = 180) and field measurements of crop yields, shade intensity (hemispherical photography and PAR sensors), soil organic matter, and nutrient indices (N, P, K). Additional data were gathered from 36 paired plots within agroforestry and non-agroforestry controls to isolate system effects. Instruments were validated through pre-testing with 20 farmers and pilot measurements, and reliability was assessed using Cronbach’s alpha for survey scales (? ? 0.78) and inter-rater reliability for yield and soil measurements (? ? 0.80). Secondary data included historical climate records (10-year monthly precipitation and temperature) from meteorological stations proximate to each site. Data analysis integrated quantitative and qualitative approaches. Descriptive statistics summarized agroforestry configurations and yield outcomes. Multilevel linear mixed-effects models were employed to assess yield responses while accounting for clustering by farm and climate, with fixed effects for agroforestry type, shade index, soil fertility, and climatic variables, and random effects for site and year. Interaction terms tested climate–agroforestry configurations on yield. ANOVA and post hoc tests compared yields across configurations within and across climates. Structural equation modeling (SEM) explored causal pathways among shading, microclimate modification, soil properties, and yield, while sensitivity analyses tested robustness to missing data and measurement error. Complementary qualitative data from farmer interviews (n = 60) were analyzed using thematic analysis to elucidate perceived mechanisms and management constraints, with triangulation to enhance validity. Expected findings indicate that integrated agroforestry configurations generally improve yield stability and, in tropical and temperate zones, augment mean yields relative to monocropping controls, while arid environments may experience mixed results depending on tree spacing and shade management. Mechanisms are anticipated to include moderated soil moisture regimes, moderated temperatures reducing evapotranspiration peaks, and enhanced soil organic matter contributing to nutrient retention; however, competitive effects such as root competition and light interception are expected to suppress yields where density is high or pruning is inadequate. The SEM is expected to reveal that climate variables moderate the strength and direction of agroforestry effects on yield, with shade intensity and soil fertility acting as proximal drivers. The study contributes to knowledge by providing cross-climate comparative evidence on agroforestry yield performance, informing design guidelines tailored to climate, soil, and crop combinations, and advancing theory on context-dependent agroforestry productivity through integration of ecological and socio-economic perspectives. Policy implications include recommendations for scalable agroforestry adoption in climate-smart agriculture, precise guidelines on tree-crop configurations and spatial arrangement, as well as targeted extension services. Practical implications address optimization of pruning regimes, soil fertility management, and moisture conservation practices to maximize yield gains across climates. Limitations include cross-sectional design constraints on causal inference and potential unmeasured confounding variables; future research directions suggest longitudinal studies incorporating phenological data and broader crop portfolios to validate the generalizability of the findings.