AN ABSTRACT OF THE THESIS OF Richard A. Little, for the Master of Science degree in PLANT, SOIL AND AGRICULTURAL SYSTEMS, presented on 2018, at Southern Illinois University. TITLE: COMPARATIVE FERTILITY EVALUATION FOR PEPPERS AND MELONS ON AN EXTENSIVE GREEN ROOF. MAJOR PROFESSORS: Karen Stoelzle Midden Co- Chair, Dr. S. Alan Walters Co-Chair, Dr. Brian P. Klubek. With the continued expansion of urban centers around the world comes a growing population of urban dwellers who need a reliable supply of produce that is healthy, safe, and locally grown. Urban green roof agriculture provides an alternative growing space to supply healthy and affordable produce to local markets and helps to address food security. However, there have been few studies done on the fertility requirements for fruits and vegetables grown in an extensive green roof environment. The growing media of most extensive green roof systems are composed mainly of light weight clay aggregate and contain a low amount of organic matter. Appropriate fertility treatments are required to optimize the potential for green roof agriculture. This research study evaluated 6 treatments for ‘Sweet Hungarian’ Peppers (Capsicum annuum, L.) and 4 treatments for ‘Savor’ Charentais Melons (Cucumis melo var. Cantalupensis, L.) grown in a three-inch extensive green roof medium located on the Southern Illinois University campus. A randomized complete design was used for both fruit crops. Each treatment for both the melons and peppers were replicated three times. Each pepper and melon plot contained 4 plants. The treatments applied to ‘Sweet Hungarian’ Peppers were: (1) Limited (reduced) fertility (Organic Miracle GrowTM at bi-monthly intervals-200 mls of solution/12 ft2 Plot); 2) Bonemeal at 6 Tbsp/12 ft2 Plot (applied at planting and flowering); 3) Bloodmeal at 6 Tbsp/12 ft2 Plot (applied at planting and flowering); 4) Bonemeal at planting then sidedressed with bloodmeal at flowering (6 Tbsp/12 ft2 Plot); 5) OsmocoteTM applied once at planting and once at flowering (3.8 oz/12 ft2); and 6) Standard granular fertilizer (13-13-13) with a broadcast application of (1.9 oz/12 ft2) complimented with a sidedress of (13% N by volume calcium nitrate at 1.9 oz/12 ft2) at flowering. The Charentais Melon treatments were: (1) Limited (reduced) fertility (Organic Miracle GrowTM at bi-monthly intervals-200 mls of solution/15 ft2 Plot); 2) Standard granular fertilizer (13-13-13) with a broadcast application of (2.4 oz/15 ft2) complimented with a sidedress of (13% N by volume calcium nitrate at 2.4 oz/15 ft2) at flowering; 3) OsmocoteTM applied once at planting and once at flowering (4.75 oz/15 ft2) 4) vermicompost at 25% of medium by total volume of 120L plot complimented by a sidedress of bloodmeal (7.5 Tbsp/15 ft2) at flowering. Watering was applied daily through drip line irrigation as needed. Plant vigor and active chlorophyll content, monthly from June to August, were recorded with a chlorophyll radiospectrometer. At harvest, the weight and yield were recorded for each plot number. In addition, nutrient composition of the green roof medium was analyzed for each treatment at the beginning of each year of replication. The objective of this research was to identify differences between a set of fertility treatments for the optimization of green roof food production and to reduce fertilizer use. The data from the three year study was analyzed using JMP statistical software. Year by treatment interactions were detected (P ≤ 0.05) for Pepper weight and yield parameters. OsmocoteTM and conventional 13-13-13 treatments outperformed the organic treatments. OsmocoteTM had higher yields than any other treatment for all three years. For the melons, environmental factors made the data inconclusive. However, the results suggest that OsmocoteTM and Vermicompost may have shown some positive interactions among the test parameters such as weight and yield.