Effect of retained trees on growth and structure

Effect of retained trees on growth and structure of young Scots pine stands Juha Ruuska, Sauli Valkonen and Jouni Siipilehto Finnish Forest Research Institute, Vantaa, Finland

Green tree retention • An integral part of contemporary forestry in Europe Objectives + Maintain specific ecological processes (habitats) + Create structurally complex stands + Maintain aesthetic values + Promote regeneration

Problems - Poorly stocked gaps around - retained trees (poor sites in N. Finland) Loss of harvested volume Logging damage, higher logging costs Poor stability in solitary distributions

Hypotheses In the vicinity of Retained Trees (RT), • Seedling stand density is lower • Seedling height and diameter growth is slower • Influence on diameter growth is greater than on height growth, and seedlings are slimmer (height-diameter ratio) • Seedlings have smaller and less branches and • An aggregated RT distribution is less harmful than a dispersed distribution

Study stands • A sample of planted or naturally generated Scots pine seedling stands with RT – 9 stands in Southern Finland – Most Scots pine upland sites covered – Seedling stand dominant height of 2 -7 m – Scots pines retained for 8 -18 years – Variable RT densities (32 - 117 ha-1) – Variable but mostly scattered spatial distributions – No significant tree removals or mortality

Sampling • All RT mapped, measured for dbh, h • 10 sample RT per stand – measured for trunk and crown variables, t, dbh growth • 8 seedling sample plots per RT sample tree – Plots systematically at 1, 3, 6 and 10 m distance from RT – Seedlings measured for h, dbh, coordinates – One sample seedling per plot per species selected among main crop seedlings (h, dbh, cr, cw, h growth, d growth, branch diameters etc. ) – One additional sample seedling per species to represent nonmain crop seedlings

Approach • Growth models – For pine seedlings, RT – Individual tree, spatially explicit • Models for branching – Max branch diameter and branch cross sectional area of 3 whorls; for pine only • Simulation – Seedling diameter and height growth – Branching variables – Subject to site, density of the seedling stand, number and diameter of RT, and RT spatial distribution

Overstory influence • Light interception is not the critical factor – Low interception rate with pine (15 % with 50 RT ha-1) • Root competition is – Water, nutrients – Greatest on poor sandy soils • Extent – Roots up to >10 m – Greatest near RT base – Great variation in shape Root density and distance from a retained Scots pine tree (Kalela 1954)

Description of tree competition • A spatial competition index • Incl. RT and seedlings • Ecological field theory: resource availability (Wu et al. 1985) • Index = f(diameter, distance) • Max value = 1 (with max RT d in data and zero distance) Competition effect of one tree by diameter and distance

Results 1. RT effect on seedling stand density • Seedling stands were dense – 3, 700 -37, 000 Scots pine seedlings ha-1 • Virtually no RT influence on pine density • Clearly less birch near RT • Greater RT effect in Northern Finland (Niemistö et al. 1993)

2. Seedling height growth • Cumulative growth = height – Clearly smaller pine seedlings near RT + Smaller effect than in Northern Finland (Niemistö et al. 1993) Relative height of pine seedlings and RT distance and diameter

3. Seedling diameter growth • Cumulative growth = diameter – RT effect negligible compared to that of stand density – Equal RT influence on height and diameter growth – Tree form not influenced by RT Diameter of 5 m high pine seedlings, RT distance and diameter, and seedling stand density

4. Maximum branch diameter • Diameter of the thickest branch of a pine seedling – Slightly smaller near RT for given seedling height – Effect was minor compared to that of • Site index • Competition from other seedlings Influence of RT diameter and distance, site (H 100), and seedling stand density on the maximum branch diameter of a 5 m high pin seedling

5. Total branching • Sum cross-sectional area of branches in 3 whorls + Clearly less near RT + More pronounced than on maximum branch diameter + Site index and competition from other seedlings also had a strong effect Influence of RT diameter and distance, site (H 100), and seedling stand density on the total branch Cross-sectional area of 3 whorls of a 5 m high pine seedling

With retention Without

6. Spatial distribution of RT • Simulated alternatives – – 16 to 48 RT ha-1 Average RT dbh 25 cm and height 21 m 15 -year simulation period Random, regular and clustered RT distribution • Results – RT spatial patterns had only a marginal effect on growth and branching of pine seedlings – 48 RT ha-1 reduced average seedling height 15%, diameter 11 -16% and maximum branch diameter 9 -10% compared to no retention

Conclusions (I) • RT do not reduce seedling stand density in S Finland as much as in N Finland • Can help control birch on problem sites • Influence on diameter growth is not greater than on height growth, and height-diameter ratio is not affected much • Seedlings have smaller and less branches, but the maximum branch diameter is not reduced much • Reduction in branching is small compared to that achieved by higher density, site-species match • An aggregated RT distribution is only little less harmful than a dispersed distribution

Conclusions (II) Retention is practiced for ecological and aesthetical purposes. It is not mandated by law but is strongly suggested for example during a certification process. In that sense, the current practice with 5 -10 trees ha-1 seems not to have a great effect on wood production • Growth effects are small • No unstocked patches in the south • Not effective in quality improvement • It is a small improvement that groups are now placed near stand edges, or on special uncut sites