Ecological genetics of Juglans nigra: Differences in early growth patterns of natural populations

Abstract

Many boreal and temperate forest tree species distributed across large geographic ranges are composed of populations adapted to the climate they inhabit. Forestry provenance studies and common gardens provide evidence of local adaptation to climate when associations between fitness traits and the populations' home climates are observed. Most studies that evaluate tree height as a fitness trait do so at a specific point in time. In this study, we elucidate differences in early growth patterns in black walnut (Juglans nigra L.) populations by modeling height growth from seed up to age 11. The data comprise tree height measurements between ages 2 and 11 for 52 natural populations of black walnut collected through its geographic range and planted in one or more of 3 common gardens. We use the Chapman-Richards growth model in a mixed effects framework and test whether populations differ in growth patterns by incorporating populations' home climate into the model. In addition, we evaluate differences in populations' absolute growth and relative growth based on the fitted model. Models indicated that populations from warmer climates had the highest cumulative growth through time, with differences in average tree height between populations from home climates with a mean annual temperature (MAT) of 13°C and of 7°C estimated to be as high as 80% at age 3. Populations from warmer climates were also estimated to have higher and earlier maximum absolute growth rate than populations from colder climates. In addition, populations from warm climates were predicted to have higher relative growth rates at any given tree size. Results indicate that natural selection may shape early growth patterns of populations within a tree species, suggesting that fast early growth rates are likely selected for in relatively mild environments where competition rather than tolerance to environmental stressors becomes the dominant selection pressure.

Keywords: Juglans nigra; adaptation to climate; ecological genetics; juvenile growth patterns; provenance tests.

FIGURE 1

Distribution of evaluated populations and test sites in relation to the mean annual temperature of the species range

FIGURE 2

Observed versus predicted heights using the selected model, (a) conditional predicted heights (using fixed and random effects) versus observed heights, (b) marginal predicted heights (using only fixed effects) versus observed heights

FIGURE 3

(a) Predicted population tree height using the selected model for three hypothetical populations originating in cold (7°C MAT), mild (10°C MAT), and warm (13°C MAT) climates. The 95% prediction confidence interval is represented by the gray band. Gray circles represent observations. (b) Calculated absolute growth and (c) calculated relative growth rates for the same three hypothetical populations

FIGURE 4

Differences in predicted height growth for each test site. Each line represents the predicted average height trajectory for each test site. Gray circles represent observations

FIGURE A1

Using parameters for the base model (Table 2), the base model prediction is presented in solid line, while in (a) the rate‐related (b₁) parameter is modified by ±20% of original value (black and gray dashed lines), and in (b) the shape‐related parameter (b₂) is modified by ±20% of original value (black and gray dashed lines)

FIGURE D1

Diagnostic plots for final model, (a) residuals versus predicted height, (b) residuals versus population home climate colored by test site, (c) normal quantiles plot for the residuals, (d) residuals variation by test site, and (e) residuals variation by population within site