Atmospheric stability impacts wind turbine power production (Wharton and Lundquist WE 2012, St, Martin et al. WES 2016) sometimes in contradictory ways (Wharton & Lundquist ERL 2012 vs Vanderwende & Lundquist ERL 2012) depending on local meteorology. Neural networks can extend limited measurements at a site towards improved resource assessment (Clifton et al. ERL 2013). Detailed lidar observations and WRF simulations of nocturnal low-level-jets (LLJ) (Vanderwende et al. MWR 2015) suggest that LLJ-induced wind shear and veer extend into the turbine rotor-layer during intense jets. This wind shear and veer varies with stability and impacts power production (Sanchez Gomez and Lundquist 2020a, 2020b) At large spatial scales, we evaluate the statistical independence of wind generators, and find that higher-rate fluctuations in wind power generation can be effectively smoothed by aggregating wind plants over areas smaller than otherwise estimated (St. Martin et al. ERL 2015). For wind resource assessment of P50 and P90 levels, interannual variability is influenced by data record length, with counter-intuitive suggestions for ideal record length (Bodini et al. WES 2016). We assess methodologies for long-term wind resource assessment (Lee et al. 2018). Further, wind turbine nacelle measurements are also affected by atmospheric stability (St. Martin et al. WES 2017) as is power production (Murphy et al. 2020).