Urban heat is a significant climate hazard for vulnerable communities within cities. It is especially hazardous for elderly and low-income members of society, and heat waves often result in the death of these citizens. Studies have shown that some of the lowest income neighborhoods in cities sustain the highest exposure to heat waves. Part of this increased exposure is due to the neighborhood morphology, or the footprint, height and placement of buildings in these neighborhoods. For appropriate action to be taken to reduce heat exposure across neighborhoods in cities, representative modeling of contributing factors at local scale is critical. We show that differences in meteorological output from the Weather Research and Forecasting (WRF) model run at 270m horizontal resolution using 10m, 100m and 1km resolution 3D neighborhood morphological inputs and with no morphological inputs produce different spatial variability in temperature, humidity, and other meteorological variables across a city. We also determine that understanding the impact of morphology on neighborhood heat exposure, along with the uncertainty across scales of the representation of that morphology, is necessary for producing actionable results for heat wave mitigation in cities.