Abstract:

Objective: Larix principis-rupprechtii is one of the main tree species for plantations in north China, but its natural regeneration is generally poor. It is therefore important to identify the main limiting factors and explore practical techniques to promote the natural regeneration. Method: In the mid October of 2018, five sites in Liupan Mountains of Ningxia, China, were selected, each of which 5 habitat types were chosen according to canopy closure (near-mature forests, forest gaps, forest glades, forest edges, and young forests grown by farmland conversion program). A sample plot with a size of 12 m×12 m was set up for each habitat type, and within each sample plot 5 treatments, i.e., cutting grasses, completely removing humus layer after grass cutting, digging holes to expose mineral soil after humus removal, opening ditches into mineral soil after humus removal, and the control (without any treatment). The completely randomized block design was used, and each block was arranged parallel to the contour line. A total of 25 plots, each containing 5 treatments with 3 replications. Larch seeds were sowed by broadcasting (100 seeds·m^-2) after the plantation floor was treated, and two treatments of covering the ground with nylon mesh and not covering as control were employed to evaluate the effect of nylon mesh-covering on seedling density by preventing birds and rats from eating seeds. We counted the number of remaining seedlings in June of 2019 for statistical analyses. Result: In the case of no mesh-covering, the mean seedling density was 2.7 plants·m^-2 for the young forests, 1.8 plants·m^-2 for the forest edges, 1.4 plants·m^-2for the forest gaps, 1.3 plants·m^-2 for the near-mature forest, and 0.8 plants·m^-2 for the forest glades; the mean seedling density was 2.1 plants·m^-2 for the opening ditches, 1.9 plants·m^-2 for the digging holes, 1.8 plants·m^-2 for the humus removal, 1.3 plants·m^-2 for the grass cutting, and 1.1 plants·m^-2 for the control. The highest mean seedling density among the 5 treatments in each habitat type was 4.1 plants·m^-2for the opening ditches at the forest edges, 3.6 plants·m^-2 for the grass cutting in the young forests, 2.4 plants·m^-2 for the opening ditches in the forest gaps, 2.1 plants·m^-2 for the digging holes in the near-mature forest, and 1.3 plants·m^-2 for the digging holes in forest glades, respectively. Furthermore, we found that mesh-covering can increase seedling density of all treatments significantly. The mean seedling density of each treatment in the near-mature forests was increased by 2.3 plants·m^-2 for the humus removal, 1.0 plants·m^-2 for the opening ditches, 0.5 plants·m^-2 for the control, 0.5 plants·m^-2 for the digging holes, and 0.5 plants·m^-2 for the grass cutting, respectively, with a range of increase of 0.5-2.3 plants·m^-2. The mean seedling density of each treatment in the young forests was increased by 2.3 plants·m^-2 for the humus removal, 1.7 plants·m^-2 for the opening ditches, 1.3 plants·m^-2 for the digging holes, 0.6 plants·m^-2 for the grass cutting, and 0.3 plants·m^-2 for the control, with a range of increase of 0.3-2.3 plants·m^-2, showing a similar scope of increase in the near-mature forests. Conclusion: The important factors restricting seedling under larch forests include animal browsing, humus layer too thick to hinder roots into mineral soil, weed competition, shading by over-dense canopy, etc. We herein propose an integrative measure to significantly increase seedling density of larch: thinning to create a medium canopy shading habitat or selecting such existing habitats (e.g., not-closed young stands, forest edges, forest gaps), removing humus layer completely or opening ditches in large areas, and covering nylon mesh after seeding to prevent animal browsing.

Key words: Liupan Mountains, Larix principis-rupprechtii, natural regeneration, forest floor treatments, seedling density