| Abstract: | In order to find a nitrogen fertilization regime that is economically feasible in commercial short‐rotation forest stands of basket willow (Salix viminalis L.), a field trial was established on a clay soil near Västerås, central Sweden, in early 1990. A response surface design was used in which three levels of treatment were chosen for each single year: year 1—0, 30, 60, kg N ha?1; year 2—0, 60, 120, kg N ha?1; year 3—0, 90, 180, kg N ha?1; year 4—0, 60, 120, kg N ha?1. Thirty‐two combinations of these levels, out of 81 possible, were chosen and divided into four blocks. Thus, eight unique treatments were randomized to the plots within each block, providing no replicates at the end of the experimental period. This gave us a more flexible and informative experiment than many of those used earlier with rather few treatments, although they were replicated. After 4 years the willow stand was harvested and the accumulated stem production of each treatment was assessed. The effect of nitrogen fertilization on accumulated stem growth over the experimental period was found to be significant only for nitrogen applied in years 2 and 3, with a somewhat (statistically significant) larger effect in year 2. Growth response to nitrogen fertilization was best approximated with a function including terms with applied amount of nitrogen in years 2 and 3 and, furthermore, a term of interaction with negative coefficient between these years. The interaction between these years was interpreted as the system's ability to recycle nitrogen from one year to another. |
