Nitrogen Relations of Ornamental Trees in Urban Soils: A First Look

Abstract

Ornamental tree fertility studies have historically used measures of growth response to determine appropriate fertilizer N application rates. Little data exists on the fate of exogenous N applied to landscape soils in urban environments. The objective of this study is to determine the effect of varying levels of N input on soil solution N concentration, N output in soil solution and the potential tree litter N contribution for Norway maple (Acer platenoides), green ash (Fraxinus pennsylvanica), littleleaf linden (Tilia cordata), and honeylocust (Gleditsia triacanthos) on clay and sand soils. A split plot design using two soil types, clay and sand, was employed to investigate treatment differences in soil solution N concentration, N output in soil solution and potential litter N contribution. Treatments consisted of sites receiving no supplemental N (non-intensive sites) and sites receiving supplemental N (intensive sites). Forty trees (24 non-intensive, 16 intensive) were evaluated at three clay soil locations in SE Wisconsin. Thirty-two trees (16 non-intensive, 16 intensive) were evaluated at four sandy soil locations in and around Stevens Point, WI. Tension lysimeters were used to collect soil solution samples at 45cm from July through December, 1999. Litter traps installed beneath individual tree canopies were used to collect litter during the same time period. One-way ANOVA indicated mean N concentration (mg/1) and total N output (kg N/ha) in soil solution on intensive sites were significantly different (p = .05) from non-intensive sites on the sandy soil study sites and for the entire data set. Mean N concentration in soil solution on the intensive clay soil study sites was significantly different from non-intensive clay soil study sites. Linear regression of total mean N output and total mean N input indicates a moderate linear relationship (r2 = 0.49) between total N input and total N output in soil solution. Two-way ANOV A (p = .05) indicated species and treatment differences in potential litter N contribution from study trees on sandy soil study sites. All four species were significantly different from one another. Mean potential litter N contributions on sandy soil study sites ranged from 0.4 kg N /100m2 (0.9 lb. N/1000 ft2)to 1.38 kg NII 00m2 (2.8 lb. NII 000 ft2). Two-way ANOVA using the transformed mean litter N contributions from the clay soil study sites indicated a species difference (p = .05) but no treatment difference. Potential litter N contribution from honeylocust was significantly different than the remaining tree species. Mean potential litter N contribution on clay soil study sites ranged from 0.63 kg NII 00m2 (1.3 lb. NII 000 ft2) to 1.29 kg NII 00m2 (2.6 lb. NII 000 ft2).