Microsite variability, safe site description and seedbed requirements for western larch germination and initial seedling establishment on a grand fir/ninebark habitat type

Forest floor heterogeneity is an important factor in shaping community diversity and species richness. We measured small scale microsite heterogeneity in an Abies grandis/Physocarpus malvaceus habitat type in northern Idaho. Variation was investigated using principal component analysis (PCA) and chi-square analysis. Burning reduced microsite variability, especially single-occurring microsite types, but the reduction was short-lived. The microsite type most common in all three measurements ({dollar}<{dollar}5 cm duff depth, 21-50% litter cover, 0 sticks, moss, shading, vegetation) was reduced in frequency of occurrence with burning. Locator variables accounted for 98% of the variation.; Western larch (Larix occidentalis) seeds were distributed after burning treatment (seven on each sample point), and the germinations (radicle emerging from seed), germinants (radicle entering forest floor and needles developing) and seedling (seed cap off and needles visible) numbers tracked over two growing seasons. Germinations (741) and germinants (656) from 17,500 seeds were found on a wide variety of microsite types representing "safe sites" (the zone with the characteristics for germination and seedling development to occur) for these stages of development. Safe site requirements became more specific as the seedling developed, with 47 seedlings surviving on 33 microsite types at the end of the first growing season. Significant descriptor variables and interactions were identified from regression; analysis produces high (R{dollar}sp2{dollar} = 0.75) predictability at the end of the first growing season for initial seedling establishment.; The effects of shading, burning and mechanical seedbed treatment on western larch germination and first year establishment were also assessed. Sixteen 0.74 m{dollar}sp2{dollar}-plots were established in 1987 and 1988 and 150 western larch seeds (100 expected to germinate) distributed of randomly applied seedbed treatments (burning or mechanical, mineral soil of duff, shaded or non-shaded, caged or non-caged to protect from predation). Germination was significantly greater on caged units than on uncaged units by mid-season due to seed predation on uncaged units. Shading was not a significant factor. Mortality of seedlings, most likely caused by a dry period with high temperatures from July to mid-August, resulted in no significant differences in seedling numbers between seedbed treatments by mid-August. (Abstract shortened with permission of author.)...