NOAA PSL thermodynamic profiles retrieved from ASSIST infrared radiances with the optimal estimation physical retrieval TROPoe during SPLASH

This dataset contains daily files with thermodynamic profiles retrieved with the optimal estimation physical retrieval TROPoe (TROPoe, Turner and Löhnert 2014; Turner and Blumberg 2019; Turner and Löhnert 2021). The profiles are retrieved every 10 min from instantaneous radiances observed with an Atmospheric Sounder Spectrometer by Infrared Spectral Technology (ASSIST, Rochette et al. 2009).

The ASSIST was deployed at Roaring Judy in the East River Watershed in Colorado (38.7169321 N,  106.853031 W, 2494 m above mean sea level) from 21 October 2021 to 28 January 2022 as part of the National Oceanic and Atmospheric Administration (NOAA) Study of Precipitation, the Lower Atmosphere, and Surface for Hydrometeorology (SPLASH) campaign. 

The spectral bands used in the retrieval are in the wavenumber range from 612 - 905.4 cm-1 and are specified in Turner and Löhnert (2021). Additional input data in TROPoe are cloud base height from a collocated ceilometer, temperature, water vapor mixing ratio, and pressure from colocated near-surface measurements and from hourly analysis profiles from the operational Rapid Refresh (RAP, Benjamin et al. 2021) weather prediction model at the closest grid point. The latter are used only outside the atmospheric boundary layer (ABL) above 4 km above ground level (AGL) and provide information in the middle and upper troposphere where little to no information content is available from the infrared radiances.

In addition to these temporally resolved input data, TROPoe requires an a priori dataset (prior) which provides mean climatological estimates of thermodynamic profiles and specifies how temperature and humidity covary with height as an input (for details see e.g. Djalalova et al. 2022). The prior is a key component of the retrieval and provides a constraint on the ill-posed inversion problem. For this study, we computed the prior from operational radiosondes launched near Denver, CO, and re-centered the mean profiles of water vapor and temperature to account for the elevation difference between the East River Valley and the launch site near Denver to get a more representative prior.

The file format is netcdf and the file naming conventions are

NOAA_PSL_ASSIST_RoaringJudy_yyyymmdd.cdf

with

yyyy: Year

mm: Month

dd: Day

The time stamp of all data is in UTC.

Selected basic variables are (many more provided):

Name

Dimension

Unit

base_time

Single value

Seconds (since 00 UTC 1 Jan 1970)

time_offset

Time

Second (since base_time)

hour

Time

Hours since 00UTC this day

height

Height

km AGL

temperature

Time, Height

C, temperature

waterVapor

Time, Height

g/kg, water vapor mixing ratio

theta

Time, Height

K, potential temperature

pressure

Time, Height

hPa, pressure

rh

Time, Height

%, relative humidity

dewpt

Time, Height

C, dew point temperature

thetae

Time, Height

K, equivalent potential temperature

sigma_temperature

Time, Height

C, 1-sigma uncertainty temperature

sigma_waterVapor

Time, Height

g/kg, 1-sigma uncertainty water vapor

cdfs_temperature

Time, Height

cumulative degrees of freedom for temperature

cdfs_waterVapor

Time, Height

cumulative degrees of freedom for water vapor

Bold variables are the main retrieved profiles, from which the other variables are derived.

Note that the vertical resolution of the retrieved profiles decreases with height, because of the broadening of the weighting function as a function of height. Thus, there are relatively few independent pieces of information in the profiles, this is reflected in the cumulative degree of freedom variables. The majority of the information from the ASSIST is in the lowest 2-3 km, above that most information comes from the RAP model.

Because of strong emission in the infrared from clouds, clouds strongly impact the ability to retrieve profiles from the ASSIST and care should be taken when analyzing the retrievals in the presence of clouds.  

References:

Rochette, L., W. L. Smith, M. Howard, and T. Bratcher, 2009: ASSIST, atmospheric sounder spectrometer for infrared spectral technology: Latest development and improvement in the atmospheric sounding technology. Imaging spectrometry XIV, Vol. 7457 of, SPIE, 9–17.

Turner, D. D., and U. Löhnert, 2014: Information content and uncertainties in thermodynamic profiles and liquid cloud properties retrieved from the ground-based atmospheric emitted radiance interferometer (AERI). J. Appl. Meteor. Climatol., 53, 752–771, https://doi.org/10.1175/JAMC-D-13-0126.1.

Turner, D. D., and W. G. Blumberg, 2019: Improvements to the AERIoe thermodynamic profile retrieval algorithm. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 12, 1339–1354, https://doi.org/10.1109/JSTARS.2018.2874968.

Turner, D. D., and U. Löhnert, 2021: Ground-based temperature and humidity profiling: Combining active and passive remote sensors. Atmos. Meas. Tech., 14, 3033–3048, https://doi.org/10.5194/amt-14-3033-2021.

本数据集包含采用最优估计物理反演算法TROPoe（TROPoe，Turner与Löhnert，2014；Turner与Blumberg，2019；Turner与Löhnert，2021）反演得到的热力学廓线的每日数据文件。该廓线每10分钟从红外光谱技术大气探测光谱仪（Atmospheric Sounder Spectrometer by Infrared Spectral Technology，简称ASSIST，Rochette等，2009）观测到的瞬时辐射亮度中反演得到。

该ASSIST设备于2021年10月21日至2022年1月28日部署于美国科罗拉多州东河流域的Roaring Judy站点（北纬38.7169321°，西经106.853031°，海拔2494米），作为美国国家海洋和大气管理局（National Oceanic and Atmospheric Administration，简称NOAA）水文气象降水、低层大气与地表研究（Study of Precipitation, the Lower Atmosphere, and Surface for Hydrometeorology，简称SPLASH）项目的一部分。

反演所用的光谱波段位于612~905.4 cm⁻¹的波数范围内，具体参数详见Turner与Löhnert（2021）。TROPoe反演所需的额外输入数据包括：协同部署的云高仪获取的云底高度，协同获取的近地面温度、水汽混合比与气压数据，以及由业务化快速刷新（Rapid Refresh，简称RAP，Benjamin等，2021）天气预报模式在最近网格点上输出的逐小时分析廓线。上述模式数据仅用于离地高度（above ground level，简称AGL）4公里以上的大气边界层（Atmospheric Boundary Layer，简称ABL）以外区域，用于补充红外辐射观测无法提供或信息极少的对流层中高层大气信息。

除上述时序输入数据外，TROPoe还需要先验数据集（先验场），该数据集提供热力学廓线的气候态平均估计，并明确温度与湿度随高度的协变关系作为输入（详细说明参见Djalalova等，2022）。先验场是反演过程的关键组成部分，可为不适定反演问题提供约束条件。本研究采用美国科罗拉多州丹佛市附近发射的业务化无线电探空仪数据计算先验场，并针对东河流谷与丹佛附近发射点的海拔差异，对水汽与温度的平均廓线进行重新中心化处理，以获得更具代表性的先验场。

本数据集采用NetCDF（网络通用数据格式）存储，文件命名规则为：NOAA_PSL_ASSIST_RoaringJudy_yyyymmdd.cdf，其中：
- yyyy：年份
- mm：月份
- dd：日期

所有数据的时间戳均采用协调世界时（UTC）。

以下为部分基础变量（数据集提供更多变量）：

| 变量名 | 维度 | 单位 |
| ---- | ---- | ---- |
| base_time | 单值 | 秒（自1970年1月1日00 UTC起算） |
| time_offset | 时间 | 秒（自base_time起算） |
| hour | 时间 | 当日00UTC起算的小时数 |
| height | 高度 | 公里（AGL，离地高度） |
| temperature | 时间、高度 | 摄氏度（温度） |
| waterVapor | 时间、高度 | g/kg（水汽混合比） |
| theta | 时间、高度 | 开尔文（位温） |
| pressure | 时间、高度 | 百帕（气压） |
| rh | 时间、高度 | %（相对湿度） |
| dewpt | 时间、高度 | 摄氏度（露点温度） |
| thetae | 时间、高度 | 开尔文（相当位温） |
| sigma_temperature | 时间、高度 | 摄氏度（温度的1σ不确定度） |
| sigma_waterVapor | 时间、高度 | g/kg（水汽的1σ不确定度） |
| cdfs_temperature | 时间、高度 | 温度累积自由度 |
| cdfs_waterVapor | 时间、高度 | 水汽累积自由度 |

加粗变量为主要反演得到的廓线，其余变量均由其衍生得到。

需注意，由于权重函数随高度展宽，反演廓线的垂直分辨率随高度升高而降低。因此，廓线中包含的独立信息相对较少，这一点可通过累积自由度变量体现。ASSIST观测获取的大部分信息集中在离地2~3公里以下的区域，该高度以上的信息主要由RAP模式提供。

由于云体在红外波段具有强辐射发射特性，云况会显著影响ASSIST的廓线反演效果，因此在存在云体的情况下分析反演结果时需格外谨慎。

参考文献：
1. Rochette, L., W. L. Smith, M. Howard, T. Bratcher, 2009：ASSIST：红外光谱技术大气探测光谱仪——大气探测技术的最新进展与改进. 《成像光谱学XIV》，SPIE会议论文集第7457卷，第9-17页。
2. Turner, D. D., Löhnert, U., 2014：地基大气发射辐射干涉仪（Atmospheric Emitted Radiance Interferometer，简称AERI）反演的热力学廓线与液态云参数的信息含量及不确定度. 《应用气象与气候学杂志》，53卷，第752-771页，https://doi.org/10.1175/JAMC-D-13-0126.1。
3. Turner, D. D., Blumberg, W. G., 2019：AERIoe热力学廓线反演算法的改进. 《IEEE应用地球观测与遥感选刊期刊》，12卷，第1339-1354页，https://doi.org/10.1109/JSTARS.2018.2874968。
4. Turner, D. D., Löhnert, U., 2021：地基温湿度廓线探测：结合主动与被动遥感传感器. 《大气测量技术》，14卷，第3033-3048页，https://doi.org/10.5194/amt-14-3033-2021。