Date of Award

Fall 12-14-2020

Level of Access Assigned by Author

Open-Access Thesis

Degree Name

Doctor of Philosophy (PhD)


Ecology and Environmental Sciences


Lakesh Sharma

Second Committee Member

Andrei Alyokhin

Third Committee Member

Richard Brzozowski

Additional Committee Members

Chunzeng Wang

Larry Feinstein

Alicyn Smart


Potato (Solanum tuberosum L.) is a major vegetable crop worldwide, including the United States of America. No other crops could equal potato in its production of food in terms of energy and value per unite area. Because potato is a high-value vegetable, farmers apply phosphorus fertilization at high rates despite high soil phosphorus availability. Phosphorus is the most critical major soil nutrient limiting potato growth after nitrogen and potassium. Six rates of P fertilization (0 – 280 kg P ha−1) were applied at twelve different sites across Northern Maine, United States of America. In the present study, soil pH was significantly correlated with total potato tuber yield (R2 = 0.38). Sites with soil pH values < 6 had total tuber yields, marketable tuber yields, tuber numbers per plant, and total tuber mean weights all higher than these same parameters at sites with soil pH ≥ 6. All sites with soil pH< 6 showed a highly correlated relationship between P uptake and petiole dry weight (R2 = 0.76). The Cate-Nelson analysis for this study allowed distinguishing two P fertility classes: Low and High, that is, 0–14.2 and 14.2 – 43.0 mg P kg-1 soil at the early potato stage and 0 –17.0 and 17.0– 42.0 mg P kg-1 soil at potato harvest time, respectively, for the Modified Morgan (MM) extractant method and 0–307.2 and 307.2–844.0 mg P kg-1 soil at the early potato stage and 0–334.0 and 334.0–845.0 mg P kg-1soil at potato harvest time,

respectively, for the Mehlich3 (M3) extractant method. The highest robustness value (R2= 62.0%) was obtained at potato harvest for the Cate-Nelson analysis with the M3 extractant method. The DPS, using the logarithmic model, showed that desorbable P increased from 16 to 29%. Vegetation indices (VIs) and plant pigments were calculated at various time points and correlated with total potato yield and P uptake. Active sensors provided a poor prediction of total potato yields, adjusted R2 ranged ( 0.05 – 0.36 ) for Crop CircleTM and ranged (0.02 – 0.57) for a GreenSeekerTM, and P uptake, adjusted R2 ranged (0.07 – 0.62) for Crop CircleTM ranged (0.01 – 0.44) for a GreenSeekerTM. Passive sensors provided a good prediction of potato yield, with R2adj ranging between 0.44 – 0.63. Their predictive values increased dramatically throughout the season, with the highest R2adj of 0.63 for the relationship between normalized difference vegetation index (NDVI), green normalized difference vegetation index (GNDVI), and chlorophyll green (CHLGR) and total potato yield at the first flight date (25 June), with a log-transformed response variable (log- transformed models). This study demonstrated multi-spectral imaging's potential application by using an unmanned aerial vehicle (UAVs) to predict total potato yield at the early vegetative growth stage with high accuracy. This study was conducted to evaluate the influence of phosphorus (P) application rates and inoculation with arbuscular vascular mycorrhizal (VAM) fungi on tuber yield, specific gravity, petioles dry weight, phosphorus concentration, and uptake. None of the treatments affected any of the yield parameters. At the same time, soil test phosphorus (Modified Morgan and Mehlich 3) was significantly correlated with VAM fungi's root colonization. It appears that the soils with high soil phosphorus test and soil pH higher than 6 was not benefited from being inoculated with additional mycorrhiza.

Files over 10MB may be slow to open. For best results, right-click and select "save as..."