Methods to improve seed yield of leymus chinensis based on nitrogen application and precipitation analysis
Optimum methods to improve perennial grass seed yield are crucial for agronomic practices. Tzvel [Leymus chinensis (Trin.)] seed yield was examined in field experiments from 2005 to 2007 to determine the effects of N application timing (i.e., double-ridge stage, spikelet initiation stage, and later seed-filling stage) and rates (0, 60, 120 kg ha–1) under different precipitation conditions. Meanwhile, the relationship of successive 5-yr precipitation and temperature data and spike number was analyzed. Application times and rates of N did not affect the heading percentage or spike number per square meter, but they were only influenced by year. A strong correlation existed between spike number and rainfall at the after seed-maturing stage during the previous growing season. In contrast to controls, the seed setting percentage, seed number per spike, and thousand seeds weight increased significantly when the highest N rates were applied at the double-ridge stage. These results suggest that the yield components were decided by two critical stages: the after seed-maturing stage of the previous year and the double-ridge stage of the current year. Among seed yield components, the spike number is the most important component, the importance of seed number per spike and thousand seeds weight is less than spike number. Therefore, to improve seed yield, we first recommend methods of watering at the after seed-maturing stage in the previous growing season, and then applied N at double-ridge stage in the current year.