Climate-driven spatial variability of height growth revealed by bitemporal airborne laser scanning data in dry conifer forests

Airborne laser scanning (ALS) provides spatially explicit measurements of forest height change that can support the development of climate-sensitive growth and yield models. We integrated bitemporal ALS acquisitions (2011 and 2022) with climate and topographic variables to develop age-independent height growth models in interior dry conifer forests of British Columbia, Canada. Height and height increment were derived at a 20 m resolution, models were fit for five species groups using nonlinear mixed-effects modelling with stand-level random effects and residual spatial correlation. Climatic and topographic predictors were incorporated as modifiers of the of the asymptotic height and growth-rate parameters. Validation showed a low overall bias and RMSE of 1.06-1.25 m. Compared with baseline dynamic models without environmental predictors, the climate-sensitive formulations reduced the independent validation RMSE by 3.1–9.4% across the five species groups. Height growth rates varied substantially among species groups and along climatic gradients, confirming that (1) maximum height is modulated by persistent site conditions, (2) height growth rate responds to climate gradients, and (3) several species groups exhibit optimal non-linear responses to temperature and precipitation. Summer and autumn temperature and precipitation were key drivers of growth rate for four out of five species groups. Quadratic transformations of climate predictors captured nonlinear, unimodal responses and identified climatic conditions associated with peak height growth. Wetter sites, as assessed using the topographic wetness index, were associated with greater maximum height, highlighting the role of soil moisture availability in regulating site potential in the region. Because climatic variables were summarized over a single 11-year interval, these relationships represent spatial associations with realized growth rather than interannual climate sensitivity. This framework provides a robust basis for developing next-generation growth models that incorporate ALS information and climatic sensitivity capable of supporting adaptive forest management under changing climatic conditions.