Sciamachy No Regression Fit Mcmc Samples

<strong>SCIAMACHY mesosphere NO data and regression model samples</strong> SCIAMACHY mesosphere daily zonal mean NO data and Markov-Chain Monte-Carlo samples from the regression coefficient distributions, derived from and for use with the <code>sciapy</code> regression module. This data set contains the following files: <code>NO_regress_output_pGM_Lya_ltcs_exp1dscan60d_km32_float32.nc</code>, <code>NO_regress_output_pGM_Lya_ltcs_exp1dscan60d_km32_float64.nc</code> - samples from the regression coefficient distributions (single and double precision) <code>NO_regress_quantiles_pGM_Lya_ltcs_exp1dscan60d_km32.nc</code> - the 0.1, 2.5, 16, 50, 84, 97.5, and 99.9 percentiles of the sampled distributions <code>scia_nom_dzmNO_2002-2012_v6.2.1_2.2_akm0.002_geomag10_nw.nc</code> - the SCIAMACHY daily zonal mean NO data <code>sciapy_regress_tutorial.ipynb</code> - example ipython notebook <strong>MCMC Samples</strong> The files <code>NO_regress_output..._float32.nc</code> and <code>NO_regress_output..._float64.nc</code> contain MCMC samples of the model as single and double precision floats. The file <code>NO_regress_quantiles....nc</code> contains the 0.1, 2.5, 16, 50, 84, 97.5, and 99.9 percentiles of the sampled distributions and is provided for convenience. The files contain the following parameters: <code>kernel:log_sigma</code>, <code>kernel:log_rho</code> - the "strength" and "lengthscale" of the Matérn-3/2 Gaussian Process kernel <code>mean:offset:value</code> - the constant offset of the NO model in [\(10^6\) cm\(^{-3}\)] <code>mean:Lya:amp</code> - the Lyman-\(\alpha\) coefficient of the mean model in [\(10^6\) cm\(^{-3}\) / Lyman-\(\alpha\)] <code>mean:GM:amp</code> - the geomagnetic coefficient (AE) in [\(10^6\) cm\(^{-3}\) / nT] <code>mean:GM:tau0</code> - the constant lifetime of the geomagnetic lifetime in [d] <code>mean:GM:taucos1</code>, <code>mean:GM:tausin1</code> - cosine and sine amplitudes of the yearly geomagnetic lifetime variation in [d] <strong>Daily zonal mean NO data</strong> The model was trained on the SCIAMACHY mesosphere NO dataset, binned into 10° geomagnetic latitude bins using the provided <code>gm_lat</code> variable and using the standard error of the mean as data uncertainties. The data are uploaded as <code>scia_nom_dzmNO_2002-2012_v6.2.1_2.2_akm0.002_geomag10_nw.nc</code> and were prepared by running (after installing <code>sciapy)</code>: <pre><code class="language-bash">bash&gt; scia_daily_zonal_mean.py -g -b'-90:90:10' -o &lt;daily_zonal_mean_NO.nc&gt; &lt;/path/to/SCIAMACHY_NO_NOM_orbits_20??_v6.2.1.nc&gt;</code></pre> <strong>Regression sampling</strong> The samples were generated by running the following command: <pre><code class="language-bash">bash&gt; python -m sciapy.regress &lt;daily_zonal_mean_NO.nc&gt; --proxies Lya:&lt;Lyman-alpha_file.dat&gt;,GM:&lt;AE_file.dat&gt; -A &lt;altitude&gt; -L &lt;geomag_latitude_bin&gt; -w 14 -b 800 -p 1400 -F \"\" -I GM --fit_annlifetimes GM --positive_proxies GM --lifetime_scan=60 --lifetime_prior exp -k -K Mat32 -O0 -m "nom_pGM_Lya_ltcs_exp1dscan60d_km32" -P</code></pre>