No description available.

Call number: LD2668 .T4 1953 R3 / Master of Science

Potato requires high levels of nitrogen fertilizer. Up to now in Quebec, no nitrogen fertilizer recommendation has been based on nitrogen availability in the soil, even those soil nitrogen testing is available. The objectives of this study were: (1) to determine the relation between soil nitrate-N content, nitrogen fertilization and potato yield; and (2) to determine the time and the sampling depth of soil most appropriate to predict potato yields and response to nitrogen fertilizer under Quebec conditions. Soil samples were taken at two depths at planting and at hilling time in 28 sites fertilized at four levels of nitrogen. Laboratory measures of nitrate-N were carried out on those samples. A combination of nitrate-N and nitrogen fertilisation gave the best explanation of the variation potato yields. A maximum of 46% of yield variation was attributed to the combination of those two factors in general. Best coefficients of determination were obtained when the sampling of nitrate-N was done at planting time compared to time. Regression models predicted nitrogen fertilizer needs of 89 to 200 kg ha$ sp{-1}.$ The same conclusions were found in the case of the variety Superior alone. The variation of yield explained was up to 62%. Regression models predicted nitrogen fertilizer needs of 77 to 224 kg ha$ sp{-1}$ for Superior. It was concluded that the analysis of soil nitrate-N can improve the prediction of nitrogen fertilizer needs for an optimum potato production.

No description available.

A project was developed to determine the feasibility of using liquid hog manure as a nutrient source (Nitrogen) for the potato crop. A survey of liquid hog manure storage facilities on Prince Edward Island (P.E.I.) provided a sampling technique and the range of nutrients found on hog farms. It was concluded from the data that there were large differences between farms and on farm manure sampling was required to determine accurate nutrient applications. An infrastructure was designed, constructed and tested for storage, handling and the application of liquid hog manure at the Harrington Research Farm, Crops and Livestock Research Centre, Charlottetown, P.E.I. Data from the first of two three year experiments determined that the placement of liquid hog manure under the sown potato row and beside the row (0.23 m) provided yields better than manure placed between the sown rows. These yields were not different for the extra Nitrogen fertilizer treatment. The second field experiment examined the placement of liquid hog manure by various injector designs between the rows after the potato crop was planted. Potato tuber yield data over the three years indicated no differences among injector design nor between the injector treatments and the treatment which received the extra Nitrogen fertilizer. Over both experiments there was a decline in the severity of Rhizoctonia ( Rhizoctonia solani) in one year for plots receiving manure compared to those which received only inorganic fertilizer. There were no differences in the incidence or severity of scab (Streptomyces scabies) over the study. In general liquid manure can be used as nutrient source for the potato crop on P.E.I.

A 4.87 hectare potato field at St. Leonard d'Aston, Quebec was instrumented to measure surface runoff and tile drain flow over the 2 year growing season period, 1989-1990. The soil type was a Ste. Jude sandy loam. Several soil and water parameters and NPK concentrations in runoff were measured. The CREAMS (Chemicals, Runoff and Erosion from Agricultural Management Systems) computer simulation model was validated for the study site. / Observed N concentrations in tile drain flow exceeded the Canadian water quality guideline of 10 mg/L. Observed P concentrations were less than 0.01 mg/L; and K concentrations, for the most part, exceeded 10 mg/L. / CREAMS overpredicted event surface runoff depths, and underpredicted event percolation depths. However, total monthly surface runoff and percolation depths closely matched observed values. / CREAMS overpredicted event nitrate concentrations in tile drain flow. There was a poor match between predicted and observed event nitrate concentrations in tile drain flow (coefficient of predictability, CP$ sb{ rm A}$ = 104.95). However, predicted total monthly nitrate load closely matched observed values (CP$ sb{ rm A}$ = 0.84). Total monthly and seasonal nitrate loads in tile drain flow were underpredicted.

No description available.

No description available.

Graduation date: 1995