Genetic and soil factors associated with variation in visual magnesium deficiency symptoms in Pinus radiata

Abstract

Magnesium deficiency symptoms are widespread in radiata pine stands throughout New Zealand, though deficiency symptoms have been reported to be more severe during the last two decades particularly on pumice soils in the central North Island, leading to reductions in growth rate and wood density. Large differences in the severity of symptoms occur over small spatial scales, which arise partly because of genetic differences in nutritional traits between trees, however, suspected variations in site factors have eluded detection. Several new trials established in a second-rotation Pinus radiata stand at Puruki forest were measured to investigate the role of soil and genetic factors associated with small-scale spatial variation in Mg deficiency symptoms observed on the pumice soils there in the first-rotation. The trials included seedling and clonally propagated progeny from open pollinated parents. Ten progeny from each of 40 open pollinated families were propagated as fascicle cuttings. In addition to clones, seedlings from a Puruki Control seedlot developed from 26 open pollinated healthy seed-parent phenotypes, and a GF30 seedlot selected for improved growth rate and stem form were included. These genotypes were planted in six plot replicates each containing a single ramet of the 400 clones, and 10 plot replicates each containing a single ramet of 80 clones (a subset of the 400). The latter 10 plots also contained 10 ramets of a single clone, and 60 seedlings (30 each from Puruki Control and GF30). The 16 plots were deployed in areas with moderate to severe Mg deficiency symptoms. Additional seedlings and clones from 20 families were planted in family row plots to test for propagation effects. Magnesium deficiency symptoms in individual trees scored at stand ages 2 and 3 years showed strong plot, clone, family, and seedlot differences. Plot means ranged from 1.6 to 3.1, clone means ranged from 1.3 to 4.1, with a 1999–2000 age–age correlation for clone means of 0.69 ( n=378), and family means ranged from 1.4 to 3.6, with a 1999–2000 age–age correlation of 0.92 ( n=40). Broad sense heritability of Mg deficiency score (and S.E.) was 0.46 (0.03) and narrow sense heritability 0.69 (0.17), based on clone scores averaged over both years. These results suggest the need to avoid seedlots based on a small number of parent trees, unless clones or families of known resistance to Mg deficiency are used. Significant plot differences in solution Ca, Cu, K, Mg, S, exchangeable Na and Al were found at the 0–0.1 m depth, and solution K, total C, N, and exchangeable K at 0.5–1.0 m depth. Mg deficiency symptoms were positively correlated with exchangeable K ( r=0.83) and solution K ( r=0.70) at 0.5–1.0 m depth, and negatively correlated with exchangeable Mg concentration ( r=0.51) at 0–0.1 m depth. Spatial variation in subsoil K primarily determined plot differences in Mg deficiency symptoms at this site. Exchangeable K concentration was positively correlated with the thickness of the lapilli layer at 0.5–1 m depth, which may explain the widespread occurrence of Mg deficiency symptoms on volcanic soil when the surface horizon is dry.