Plant and soil responses to fertilization of grasslands in Saskatchewan, Canada and Selenge, Mongolia
Studies were conducted at three different sites in Saskatchewan, Canada (Colonsay, Vanscoy and Rosthern) over two years (2005 and 2006) to determine the effect of dribble banded and coulter injected liquid fertilizer applied in the spring of 2005 at 56, 112 and 224 kg N ha-1 with and without P at 28 kg P2O5 ha-1. A similar study was conducted in 2006 at one site in Mongolia to determine the effect of granular N and P fertilizer application on dry matter yield, and N and P concentration in plant biomass in the year of application (2006) only. The three Saskatchewan sites were unfertilized, 7-8 year old stands of mainly meadow bromegrass (Bromus riparius) dominated haylands, while the Mongolia site was mixed species of native rangeland.All fertilization treatments produced significantly (p≤0.05) higher dry matter yield than the control in the year of application at the three Saskatchewan sites. The addition of 28 kg P2O5 ha-1 P fertilizer along with the N fertilizer did not have a significant effect on yield in most cases. In the year of application, increasing N rates above 56 kg N ha-1 did not significantly increase yield over the 56 kg N ha-1 rate in most cases but did increase N concentration, N uptake and protein content. A significant residual effect was found in the high N rate treatments in 2006, with significantly higher yield and N uptake. In 2005, the forage N and P uptake were in all cases significantly higher than the control in the fertilized treatments. The N uptake at the three Saskatchewan sites increased with increasing N rate up to the high rate of 224 kg N ha-1, although the percent recovery decreased with increasing rate. The P fertilization with 28 kg P2O5 ha-1 also increased P uptake at the three Saskatchewan sites. The site in Mongolia was less responsive to fertilization than the three Saskatchewan sites, with only a significant response in yield, N uptake and no significant effect of P fertilization. For incubation soil cores collected in the fall of 2006, mean respiration rates were similar among the fertilized treatments at all the sites and the pattern of CO2 and N2O evolution measured over a two-week period showed similar trend at the three sites, with no significant difference between treatments. However a significant increase in gas production occured as the soils were wetted during the incubation. By the fall of 2005, the fertilization effect had likely diminished along with available substrate for the soil microbial biomass.Overall, rates of fertilizer of approximately 50 kg N ha-1 appear to be sufficient to produce nearly maximum yield and protein concentration of the grass in the year of application for the Saskatchewan and Mongolia sites. Surface banding placement was as effective as in soil placement and there was limited response to P fertilization. A small amount of N fertilizer that is surface-placed on these grass dominated forage systems appears to be an effective means of increasing production in the year of application. Higher rates are needed to sustain the rejuvenation beyond one year.
DegreeMaster of Science (M.Sc.)
SupervisorSchoenau, Jeffrey J. (Jeff); Anderson, Darwin W.
CommitteeLardner, Bart; Knight, J. Diane; Si, Bing C.
Copyright DateMay 2007