GATNDOR is a research team supporting the development and implementation of GRUAN on scientifically sound foundations. The team performs focused, short-term research to address specific topics identified by the GRUAN science and management community. GATNDOR efforts are coordinated with those other GRUAN Task Teams and with national GCOS programs when appropriate.
Under the auspices of the GCOS (Global Climate Observing System) AOPC (Atmospheric Observation Panel for Climate) Working Group on Atmospheric Reference Observations, a research team will identify and carry out a series of well-defined, limited-scope retrospective analyses of existing observations from established stations that are potential GRUAN (GCOS Reference Upper Air Network) sites, other complementary observations, metadata, and model simulations. The unifying purpose of this series of studies will be to obtain insight from existing information for optimizing the design and implementation of the GRUAN observational program to meet its scientific goals.
At the 1st GRUAN Implementation and Coordination Meeting (ICM-1) in Norman, Oklahoma, in March 2009, an initial research team of four interested volunteers was formed. It included Reinout Boers, Tom Gardiner, Dian Seidel (chair), and Junhong Wang. A preliminary brainstorming session following the meeting, and a team conference call in April 2009, identified an initial set of five research topic areas.
- Topic 1: Collocation of Observations
- Topic 2: Management of Change
- Topic 3: Scheduling Protocol
- Topic 4: Quantifying the Value of Complementary Observations
- Topic 5: Network Configuration
The team held quarterly conference calls to discuss ideas, progress, impediments, and plans.
At ICM-2, in March, 2010 in Payerne, the team reported progress on Topics 1 and 2 and plans for Topic 3. Topics 4 and 5 were placed “on the back burner” for future work. For details, see the GRUAN web page devoted to the agenda of the Payerne 2010 meeting. GATNDOR presentations are in Session 4.
Discussions at ICM-2 resulted in the formation of several new GRUAN Task Teams to address high priority GRUAN needs. One new Task Team was given responsibility for recommending a GRUAN observing schedule protocol, GATNDOR Topic 3. At a lunchtime meeting of all interested ICM-2 participants, these goals were set for the following year:
- Expand the membership of GATNDOR
- Forge stronger links between GRUAN/GATNDOR and the satellite community
- Submit a paper to a peer-reviewed journal on Topic 1
- Present preliminary results to ICM3 on Topics 2, 4 and 5
This work plan outlines how these goals are to be pursued.
At ICM-3, end of February – early March 2011 in Queenstown, NZ, the team reported progress and achieved results on all the Topics. Moreover, a new sub-topic to be developed in the frame of Topic 1 has been identified and presented to the GRUAN community. The sub-topic will be related to the design of a tool for evaluation of instrumental co-location mismatch for GRUAN sites and Franz Immler will lead the investigation of this issue.
Discussions at GATNDOR side meeting, during the break-out session of ICM-3 on March 2, identified the main results to be released before ICM-4 by means of tools, recommendations and strategies provided to the GRUAN stations. Below, all the topics are described according to their work progress and the mentioned deliverables that also represent the core of 2011-2012 work plan. Deliverables will be made available to the GRUAN community though peer-reviewed papers reporting the outcome of proposed investigation.
During the side meeting it has been also reiterated that principle membership is by no means exclusive; interested scientists are welcome to participate and are encouraged to contact the chair. However, contributions from new collaborators should be well focused to urgently address a few research questions/themes, identified as possible new topics to investigate in the frame of GATNDOR or to be addressed in the enlarged GRUAN community:
- Are error covariance matrices mandatory/useful for providing vertically correlated uncertainties when derived products are considered?
- Use of GRUAN data in order to show their value (as soon as the reprocessing of historical radiosounding data will be available) through the investigation of appropriate scientific topics.
- Validating weather and climate models and re-analysis products using GRUAN data;
- Plan for a network design workshop in order to put together all the expertise already existing in this field.
No volunteers have been identified to address these issues.
A closer cooperation among different topics will be fostered in order to address common goals. For example, scientists working on Topic 2 and 3 will collaborate on the management of change for complementary measurements (lidar, microwave, ….). Moreover, GRUAN Task Team 5 on Ancillary Measurements offers its support. However, a closer cooperation between GATNDOR and TT5 will be established.
Funding perspectives have been also discussed. “MarieCurie” funding scheme of EU and a “medium term trust fund” solution have been proposed: both solutions will be discussed during the quarterly teleconference. Currently GATNDOR work is carried out entirely on voluntary basis.
At ICM4, in Tokyo (Japan), during the GATNDOR side meeting, the team agreed on the extension for one year of two topics, while the topic related to the Management of Change will not be part of GATNDOR by this year, since the Team Lead delivered the expected results at the ICM-4 and since the Management of Change will be moved to the duties of Task Teams, including other instruments. It is also worth to be mentioned that other GRUAN partners has shown some interest in the topic “Atmospheric Variability and co-location” and offered a potential support.
The field of research questions/themes, identified as possible new topics to investigate in the frame of GATNDOR, have been restricted and well separated from those to be addressed in the enlarged GRUAN community, mainly by the Task Teams. However, a urgent need of interactions between TT3, TT5 and GATNDOR has been point out by the three leaders of the three teams.
One of the priority research questions to be addressed within GATNDOR will be:
“Assessment of the impact of comparison model error: use of GRUAN uncertainty covariance matrix and assessment of best vertical resolution to use in Radiative Transfer modeling for improving satellite validation”.
At ICM-4, a change in the recruitment strategy has been also proposed: the recruitment will be mainly oriented in finding PhD students that will operate on the GATNDOR topic, selected by the team. PhD funding will be raised in collaboration with the institution participating in GRUAN and the Universities they are linked with. PhDs will work under the supervision of the team members according to their different scientific skills.
Therefore, a call for the recruitment of PhD student working under GATNODR supervision will be delivered in early April.
It is also worth be mentioned that four GRUAN station (ARM, Payerne, Potenza, Sodankyla) already has made available at least one year of high-resolution radiosonde, Raman lidar, microwave radiometer, GPS and sun-photometer data to be investigated in the frame of GATNDOR topics. Data from a fifth station (Beltsville) will be soon available.
Based on recruitment operated till now, and outreach by GATNDOR team members, the current GATNDOR team is listed here. These individuals are responsible for specific GATNDOR projects and collaborate with a wider community to meet GATNDOR goals.
School of Engineering and Applied Sciences
Cambridge, MA 02138 (USA)
Expertise: sensor calibration, satellite remote sensing, radiative transfer, collocation of observations, management of change (especially for remotely sensed and in situ measurements of the same variable)
Principal Research Scientist
Environmental Measurements Group
National Physical Laboratory
Hampton Rd Teddington Middlesex UK TW11 0LW
Tel: +44 20 8943 7143 Fax: +44 20 8614 0455
Expertise: measurement uncertainties for atmospheric sensors, optical sensing of atmospheric constituents (in particular Fourier Transform spectrometers, laser absorption spectrometers, and differential absorption lidar), and uncertainties in long-term trends.
Professor of Statistics
Head of DIIMM, Dept. of Information Technology and Mathematical Methods
University of Bergamo
Via Marconi 5, 24044 Dalmine BG I, Italy.
Phone +39 035 2052 323
Fax +39 035 2052 310
Expertise: statistical methods and application on environmetrics, sensitivity analysis of computer models, environmental time-series, spatio-temporal data, stochastic monitoring, industrial statistics and structural and geotechnical surveillance, quality control and financial time series analysis.
Fabio Madonna (chair) Istituto di Metodologie per l’Analisi Ambientale C.N.R. – I.M.A.A. Contrada S. Loja Tito Scalo zona industriale I-85050 Potenza, Italy
Phone: +39 0971 427252 Fax. +39 0971 427271
Expertise: Lidar and microwave remote sensing, Integration of active and passive remote sensing techniques
Dian Seidel NOAA Air Resources Laboratory (R/ARL)
1315 East West Highway
Expertise: climate variability and change, data quality and continuity, radiosonde observations and metadata, upper-air temperature trends
Junhong (June) Wang
P.O. Box 3000
Boulder, CO 80307-3000 USA
Expertise: radiosonde observations, data quality and climatology, GPS
observations and application, water vapor measurements and variability,
upper-air sounding instruments and data quality.
Greenbelt, MD 20771 USA
office +1 301 614 6703
lab +1 301 614 6775
ALVICE +1 301 614 6515
fax +1 301 614 5492
Expertise: water vapor measurements and instrumentation; particularly lidar
and balloon-borne sensors, satellite retrieval and radiance validation, aerosol
and cloud studies using lidar
Below are descriptions of three GATNDOR topics. For simplicity and flexibility we have dispensed with the previous topic numbers. The titles are:
- Atmospheric Variability and co-location
- Quantifying the Value of Complementary Observations
Title: Atmospheric Variability and co-location
Research Question: comparison of atmospheric observations measured with different techniques at the same site are generally not taken at exactly the same time and the exactly same spot (quasi-redundant). The (natural) variability in the temporal and spatial frame given by the dislocation of the two measurements limits the accuracy of the comparison of these measurements. Based on this variability and the (inherent) uncertainties of the instruments a definition can be derived to what extend quasi-redundant measurements can be considered co-located.
Approach: Based on model or measurement data the variability of temperature and humidity in the temporal and spatial range that is typical for the dislocation of two quasi-redundant measurements (e.g. lidar and radiosonde) is established. A formalism has been already developed and presented to the GRUAN community. The effect of the separation between two or more complementary measurements on the combined measurement uncertainty will be assessed using the statistical distribution of the effect of the separation difference, and hence the uncertainty. Models are being developed of the joint distribution of the measurements and their correlation both in space and in time. Multivariate spatio-temporal stochastic processes are being used to represent this and will be fitted to appropriate measurement data using Bayesian or maximum likelihood technique. Dislocated sonde data from Sterling and Beltsville GRUAN site are made available for this study, as well as sonde and lidar data from ARM, Beltsville, Potenza, Payerne and Sodankyla GRUAN sites.
Previous Relevant Work: This project builds on earlier GATNDOR and GRUAN research to quantify individual components of the overall profile uncertainty. One study (Sun et al., J. Geophys. Res., 2010) presented a methodology for estimating uncertainty in one observation given a second “reference” observation. That study also quantified the effects of time and space collocation mismatch on temperature, humidity and refractivity profiles. A second study (Seidel et al., J. Geophys. Res., 2011) developed a global climatology of radiosonde balloon drift distance and rise rate, with particularly attention to GRUAN sites, for use in quantifying uncertainty associated with collocation. A third paper (Immler et al., Atmos. Meas. Tech., 2010) presented a framework for characterizing overall measurement uncertainty in GRUAN data products. Recently, the formalism at the base of the investigation of “Co-location of observations” topic has been internally circulated to GATNDOR and WG-ARO for discussion. This will be part of a peer review publication at the end of the project.
Team Lead: Alessandro Fassò (University of Bergamo)
Possible Collaborators: Fabio Madonna (Potenza), Belay Demoz (Beltsville), Bomin Sun, Tony Reale (NOAA/NESDIS) and Dian Seidel (NOAA/ARL)
Start Date and Estimated Project Duration: Fall 2011 + 1.5 year
Title: Quantifying the Value of Complementary Observations
Research Question: How much is measurement uncertainty reduced by having redundant or complementary measurements of a given variable?
Approach: This topic aims to estimate uncertainty of vertical profiles of both temperature and moisture using data from highly-instrumented GRUAN sites (e.g., ARM site, Beltsville, Potenza, Payerne, Sodankyla) and to quantify the error reduction resulting from increasing redundancy of measurements. This requires the assessment of the uncertainty of the temperature and moisture vertical profiles retrieved using each of the considered techniques and then the investigation of possible sensors’ synergies to reduce the uncertainty. The investigation will be carried out focusing on the most common instruments at the considered GRUAN sites: for temperature, radiosoundings and microwave profilers; for moisture, radiosoundings, Raman lidars, microwave profilers, and GPS receivers.
The quantification of the value added by complementary observations will be assessed with respect to the following issues:
- Sensor calibration/inter-calibration (here the ARM Value Added Products could be considered as a model)
- Identification of possible biases
- Representativeness of measurements
- Quality control/assurance with a focus on instrument performance in different meteorological conditions.
Level of redundancy achievable using different measurement techniques together is analyzed at both the levels of providing independent observations, able to improve the atmospheric profiling through the merging or integration of different measurements, and of providing duplicative observations, ensuring continuous measurements of an atmospheric parameter without temporal gaps. The real benefit in performing redundant measurements will be outlined in terms of uncertainty reduction, instrument types and equipment costs. Moreover, the improvements in radiative transfer modeling and radiation closure experiments resulting from the use of radiosonde humidity profile alone or in combination with Raman lidar measurements will be discussed.
The main mathematical tools that will be used for this purpose will be canonical correlation, redundancy analysis and principal component analysis.
The final outcome of this investigation is the provision of recommendation on the optimal observation strategy to use at GRUAN sites for increasing accuracy of measured parameters through redundancy and reducing profiling uncertainty. The objective is to provide sort of functions/curves of “redundancy level/number of instruments or “redundancy level/cost” to be used for expansions of existing sites or the establishment of new sites, according to GRUAN standards. Thresholds to be used for the identification of minimum redundancy levels to be considered at GRUAN sites will be also defined. This will also help defining the real complementary in support of the sampling strategy.
Previous Relevant Work: Wang and Young (2005)
Team Lead: Fabio Madonna (Potenza)
Possible Collaborators: Nico Cimini (Potenza), Belay Demoz (Beltsville), Juergen Güldner (Lindenberg), Seth Gutman (NOAA), Rigel Kivi (Sodankyla)
Start Date and Estimated Project Duration: Fall 2010+ 2 years
Work Plan for April 2012-March 2013
GATNDOR Work Plan for the period from April 2012 to March 2013 is constituted by 6 action items. The following table describes the action items, the person responsible for each of them and the expected completion date.
|Publish GATNDOR workplan for coming year||
GATNDOR action items
|Investigate essential “recipe” of GRUAN ancillary and sonde measurements (including uncertainties) to calculate SASBE with focus on atmospheric temperature and moisture for NPP satellite hyper-spectral products validation.||
GATNDOR action items
|Submission of a peer-reviewed paper about the topic “Quantify the value of complementary measurements”||
GATNDOR action items
|Publish radiosonde overlap. Either as peer review (1st preference) or as special report.||
GATNDOR action items
|Submission of a peer-reviewed paper about the topic “Co-location of observations” with an emphasis of how to account for mismatches in quantifying and comparing uncertainties.||
GATNDOR action items
|Circulation of a Phd position for working on a new GATNDOR topic: “Assessment of the impact of comparison model error: use of GRUAN uncertainty covariance matrix and assessment of best vertical resolution to use in RTM modeling for improving satellite validation”.||
WG/GATNDOR action items
This table is part of the master list of GRUAN action items available at: