This thesis aimed to develop models for the natural regeneration of Sitka spruce that could be applied to stands under continuous cover forestry in the UK. The models followed the development of Sitka spruce seedlings from the establishment phase until their recruitment to the overstorey. First, I prepared models for estimating the presence and density of Sitka spruce seedlings under canopy cover. I simulated the likelihood of seedling presence according to a combination of stand and site characteristics, and stand management history. I estimated the seedling density by generating random numbers from a Weibull distribution. The calibration data came from an existing dataset extensively collected all around the UK, and I carried out an independent validation with newly collected data in two sites in Wales and England. Then, I modelled the early height and diameter growth of Sitka spruce and western hemlock, and the early growth of Douglas fir, as a function of light availability. I compared light-growth models based on non-linear asymptotic functions, including tree size and intra-regeneration competition as predictors. Results showed that different non-linear structures best simulated the growth of different species. I carried out an analysis of the crown plasticity, observing that both the apical dominance and live crown ratios were negatively correlated with light availability for all species, although with some species-specific differences. Overall, I defined a shade-tolerant ranking as Douglas fir ≤ Sitka spruce < western hemlock. For this study I collected data across several forests in Wales, England and Scotland. Given the importance of the below-canopy light regime analysis in this study, I investigated a possible technological improvement for its assessment. I demonstrated that hemispherical photography carried out with smartphone cameras equipped with fish-eye lenses can be an adequate alternative to traditional cameras, with cheaper equipment and faster methods. The data were collected during the above field campaign. Finally, I tested the feasibility of using the models aforementioned for simulating Sitka spruce natural regeneration in a continuous cover forestry scenario in the UK. I integrated the Sitka spruce regeneration presence, density, and early growth models into MOSES_GB, a forest growth simulator under development by the Forestry Commission. I simulated the development over 15 years of a regeneration layer under an overstorey layer, the development of which was simulated by MOSES_GB. I simulated two different virtual stands, and followed the establishment, growth, and recruitment of the regeneration into the overstorey, comparing two different spatial modelling approaches and two thinning regimes. A spatially-explicit approach, which considered the local variability in the light-availability, seemed to be more accurate. Some issues and potential problems in the simulations were identified, together with the future studies that can address them.