Variation In within Canopy Attenuation of Wind Speed In Container Grown Trees: Measurement Errors and Their Impact On Canopy Transpiration Estimates

We examined the canopy wind speed extinction coefficient (α) along a canopy depth profile in ten tree species.   Above-canopy wind speed and canopy α values both followed a strong diurnal pattern with higher wind velocities and lower canopy α values during day light hours - indicating that canopy α values scale with wind speed.  Additionally, canopy α values from periods of low above-canopy wind speeds (< 1 m s^-1) often yielded erroneously high estimates and, when averaged over a 24 hour period, were significantly different from day light specific averages.  To test the impact of this variation on estimates of canopy α values, we filtered the data to remove observations at low above-canopy wind speeds and/or to remove night time measurements.  By investigating the average α value over periods ranging from a single day to an entire growing season, we found a marked change in canopy α’s as a result of leaf area development and canopy structure dynamics.  We used this variation in canopy α estimates, from different time scales and filtering methods, to parameterize a three dimensional mechanistic canopy transpiration model (MAESTRA) and assess the impact of different α values on canopy transpiration estimates. Modeled estimates of canopy transpiration varied by as much as 30% between averaging methods, underscoring the importance of carefully characterizing the canopy wind speed extinction profile when above canopy wind speeds are greater than 1 m s^-1.  We discuss the pitfalls encountered in utilizing averaging methods that, due to averaging errors, do not accurately represent how wind interacts with the canopy and propose a simplification in parameterization of the seasonal pattern of α’s among different nursery grown trees.