deletions | additions       diff --git a/section_Discussion_and_Conclusions_label__.tex b/section_Discussion_and_Conclusions_label__.tex  index 6acd237..7b01ba2 100644  --- a/section_Discussion_and_Conclusions_label__.tex  +++ b/section_Discussion_and_Conclusions_label__.tex 
...
Likely the best way to study dust devil formation and dynamics in the field is not statistically, but directly via deployment of sensor networks that produce a variety of data streams with high spatial and time resolution. Field work with small sensor networks has a long history \citep[e.g.][]{Sinclair_1973}. \citet{2004JGRE..109.7001R} conducted a concerted field study, deploying an vast arsenal of sensors to measure surface heat fluxes of heat, water vapor, short and long wave radiation, soil heat flux, pressure, wind, temperature, water vapor and dust concentration, and electric field on Santa Cruz Flats in Arizona. The results showed that terrestrial dust devils produce heat and dust fluxes orders of magnitude larger than their background values and often involve strong electric fields that might play a significant role in dust sourcing.   In the decade since, technological developments in miniaturization and data storage now provide a wealth of robust and inexpensive instrumentation, ideally suited for the long-term field deployment required to study dust devils, without the need for direct human involvement. Expanding on earlier Recently, \citet{Lorenz_2015} deployed an array of ten miniature pressure- and sunlight-logging stations at La Jornada Experimental Range in New Mexico, providing a census of vortex and dust-devil activity at this site. The simultaneous measurements resolved horizontal pressure structures for several dust devils, giving a sense for vortex size and intensity.   Even though such a sensor network can provide spatially resolved information, the resulting estimates of devil structural parameters are still somewhat degenerate as a result of the uncertain encounter geometry. In fact, the distortion biases discussed here can be incorporated via a Bayesian inference scheme to  work out the likely devil structure: Equations \ref{eqn:dpdP_obs} and \ref{eqn:dpdGamma_obs} actually represent likelihood functions, providing the probability to recover $P_{\rm obs}$ and $\Gamma_{\rm obs}$, given a devil with $P_{\rm act}$ and $\Gamma_{\rm act}$.