On-Farm Forage Fertilization And Cattle Manure Value Chain Characterization In Vietnam, Low-Infrastructure Fiber Technique, And Image Analysis For Alfalfa-Grass Harvest Management

Abstract

Crop-livestock systems play a key role in food and economic security for smallholder farmers worldwide. Research targeting farmer and agricultural extension educator-identified agroecological problems can help provide empirical solutions to these problems and improve access to evidence-based information. This dissertation includes components of four studies undertaken in Vietnam and New York State. The first objective was to assess the impact of forage fertilization with composted cattle manure and urea on dry matter yield and nutritive value of Brachiaria Cv. Mulato II forage in south-central coastal Vietnam. Highest forage yields were observed with combinations of urea N and compost. Partial nutrient budgets were negative, suggesting that the management system may not be sustainable over time. Few nutritive value effects were observed, suggesting that fertilization management for high yield is more feasible than management for high nutritive value. The second objective was to characterize cattle manure value chains originating in two Vietnamese communes in the south-central coast. Active seasonal manure trade took place between origin communes and end users (i.e., highland pepper and coffee farms, highland rubber plantations, and southeast coastal dragon fruit farms). Manure trade plays an important economic role for manure value chain participants and may affect sustainability of local nutrient management systems in manure origin and destinations. The third objective was to develop a low-infrastructure neutral detergent fiber technique for forage chemical analysis that could provide accessible technology for low budget laboratories such as those found in many developing countries. The technique was tested on a diverse set of temperate and tropical grasses and legumes to assess performance, which overall did not differ from a standard filter bag method (ANKOM). The fourth objective was to develop a field technique to reduce uncertainty in prediction of alfalfa and grass fractions in mixed stands in New York State. Local binary patterns in digital images were processed several ways to assess system potential to improve spring harvest timing and fiber estimates with equations requiring an accurate stand composition estimate. Current results suggest that the method can be useful in the field, but further testing is needed. Data and information generated in these diverse studies are useful to farmers, extension educators, and development practitioners, and also provide direction for future work.