RMRCO logo   R. Michael Reynolds, 214 Euclid Av., Seattle WA 98122, 206-466-6078      michael@rmrco.com

LastEdit: 2014-11-06

PORTABLE RADIATION PACKAGE (PRP) AND THE ACAPEX CRUISE


CONTENTS

General info:: Ship Info    About PRP2    Links   
Operation Basic information    Parts list    Critical spares    PC Software   
SGP14 Cal/Val sgp14 Deployment    Log    Data    AOD    FRSR Cals    Setup    ARM Hemis    NIMFR    MFRSR   
Mauna Loa MLO Deployment    Log    Data    AOD    Setup   
ACAPEX Project Details Statement of Work, Annotated   
Task log:  0   1   2  3   4  5   6  7   8  9   10  qa1   qa2  qa3   qa4  qa5   qa6

top


LINKS

PRP-2 HANDBOOK (v8)
JTEC-2000 FRSR DESCRIPTION AND THEORY
JTEC-2003 FRSR UNCERTAINTY
ADDING SPN1 TO PRP
INSTALLATION FOR MAGIC CRUISES
PRP CAL/VAL IN BOULDER CO.
JGR-2010 MICROTOPS AT-SEA MEASUREMENTS
Memo#1308 RAD MAINTENANCE
RAD/SPN/GPS/TILT DATA PROCESSING
PRP2 MANUAL
PRP2 DATA PROCESSING
NOAA R/V RON BROWN
NOAA MAUNA LOA OBSERVATORY

top


SHIP INFORMATION AND PRP DEPLOYMENT

The deployment location is the best possible for minimizing shadows. However, shadows are a possibility for solar zenith angles greater than 45 deg from the stren and 70 deg towards the bow.

The above sketch, approx July 2014, was a first suggestion of the location of the AMF2 instruments for ACAPEX.

top


GENERAL DESCRIPTION OF THE PORTABLE RADIATION PACKAGE, MODEL 2 (PRP2)

The above schematic shows the PRP2 system in a stand-alone mode where the LAN is private and the switch and laptop are provided. In the AMF2 deployment the ethernet LAN will be part of a system wide network and the Laptop would be replaced by the AMF2 virtual machine.

top


OPERATIONAL INFORMATION

In this section we provide information and links for operators who are using PRP2 in a field program.

CRITICAL SPARE PARTS

The PRP2 is an operational instrument that is proven to have a long MTBF. With care the system should operate for months to years without failure.

DescriptionPart No.TechSource
1 FRSR - Calibrated MFR head 10.01.06.01 1 ARM
2 FRSR - Cable 10.03.01 1 RMR Co.
3 FRSR - Shadowband 10.01.02.06.01 1 RMR Co.
4 CDU - Ethernetplugs 10.03.08 1 Newark 25H6701
5 CDU - Onset TT8 10.02.01.01 2 ANL cache
6 Hose clamps -- 1 McMaster
7 RAD Preamp INA118 10.03.34 2 DK# INA118P-ND
8 RAD Preamp LTC1050A 10.03.34 3 DK# LTC1050CS8#PBF-ND

Technician levels are borrowed from http://academicaffairs.ucsd.edu/_files/staffhr/classification/concepts-electech.pdf (1) Trainee/Novice (2) ET (3) Senior ET (4) Principal ET top

CENTOS VIRTUAL PC REQUIREMENTS

This section addresses preparation of a new Linux/Unix PC for the data collection software. In particular, if the DAQ sw is going to run on a virtual PC running CentOS, it will need the following software packages.

UNIX/LINUX APPS

PERL MODULES -- Currently (140901) these modules and versions are in use.

top


SGP14—SYSTEM BURN IN, CALIBRATION AND VERIFICATION AT THE SGP

top

SGP14 LOG

2014 08 04216Install PRP2 in SGP radiation site.
2014 08 05217Set up communication, internet connectivity.
2014 08 06218Confirm external ssh and ARM auto data ingest (Swank).
2014 08 07219RMR returns to Seattle. RAD1 unit arrives for calibration and repair.
2014 08 07219 RMR returns to Seattle. RAD1 unit arrives for calibration and repair.
2014 08 10222 Ssh connections to Asus PC works well. Easy data collection. We will plan to pull data each evening.
2014 08 11223 We note occasional breaks in the data collection software. I suspect it is a break/spike in the AC power at the box out in the field.
2014 08 12224 RAD1 repair. ANL ordered parts for the repair but say delivery will be on 8/26. DK should be overnight. I am trying to find out why the delay.
2014 08 13225 No problem. We received the Digikey Parts this afternoon. Maybe tomorrow we can make the repair and get this RAD1 back on line.
2014 08 14226 RAD 1 was repaired. We removed the circuit board and found further evidence of a massive shock. The preamp feedback resistor for U2 amp was completely missing. The shock that sheared off the standoffs knocked the precision 10K resistor right off. The resistor was replaced with one we had in inventory and the unit was working well.
2014 08 15227 RAD 1 was re-calibrated. Packaged and sent by Fed Ex to SGP, Craig Webb. He should have it Monday morning.
2014 08 16228 Upgrade DAQ software.
  • Wrote QA programs that assess if the data collection is updating properly or if data collection is stalled.
  • Found a bug in the data ingest software. Remove stray non-ascii bytes.
  • Ssh all new software to field PC and confirmed operation.
2014 08 18230 RAD1 arrives at SGP. Late delivery so we will install tomorrow.
2014 08 19231 Craig installs RAD1. All is good. RAD sw=777 W/m^2. RA2 sw=772 W/m^2. This is less than 1%. The SPN total values are around 883. Considerable more than the RADs. →I need to sort this out. →Suddenly the MFR temperature is not registering. The ADC read gives zero. Apparently Thead went to open circuit at 2014 08 18 02 14 52. Bin-ascii message = "7652 2014 08 18 02 14 52 ##0017,Loo00000000000010*678##", where "oo" means zero volts. I checked the rsr by direct connection and the ADC channel 7 reads zero.
2014 08 20232 Confirmed that Thead is out. I contacted Webb to swap the head cables today and we'll see if it is the cable.
2014 08 21233 FRSR head cable. Craig removed the cable and checked resistance. It was all okay. He replaced the cable and I checked that the problem was still there.
2014 08 22234 Thead came back at 2014 08 21 19 15 34. That's 1400 local time. Just when Craig was doing some maintenance. Data post-processing upgrades. We now process optical depth for head temperatures within a narrow range around 40 C.
2014 08 23235 Sat 0600 check DAQ. All okay and Thead is good. Download current data. I am plotting the RAD and SPN data and the SPN looks terrible.
2014 08 24236 Sun 0500 check DAQ. All okay and Thead is good. Process RAD data. PSP IR correction software is upgraded for fixed gps and fixed tcm. Acapex data web pages are expanded to show new analysis.
2014 08 27239CDU heater circuit goes out. Swank noted it was running very hot. Now it appears dead because the head temperature is ambient. The CDU is being returned to Seattle for service.
2014 08 29241VM has been set up by D. Swank. I checked out the system and log on is good.
2014 08 30242All PRP2 software was loaded on the VM. Bash is good. Problem: some PERL modules need to be installed.
2014 09 01244Data processed and added to this doc. Daily AOD data are computed.
2014 09 08251CDU was repaired over the weekend. Returned to SGP by FedEx.
2014 09 09252VM is working sort of okay. There are plenty of warnings. See task log below.
Equipment is installed and data collection on the Asus began at 2014-09-09(252)22:30:00Z.
2014 09 102531235h check in at sunrise. All is okay.
20Z--VM now checks out. I stopped Asus operation and started VM data collection.
2014 09 11254 VM is operating well. David confirms data collection. I used the “tail -f” to confirm the data were coming into the ARM files.
2014 09 1425703h-I checked the system last night: reboot required. Data stopped at 2014 09 13 17 24 10. Around noon.
2014 09 14257 1300h-The PRP2 is on line and operating well. I check it each morning and evening.
2014 09 1926213z-I am checking the ChkDaq/Stopprp/Startprp as a candidate for the cron check for PRP.
2014 09 2426714z-We are having trouble with the system. For the past two days there have been hangups and today D. Swank confirms there is power maintenance activity. I have introduced the cron program, UpdateDaq and David will check it out and start a cron job. Hourly.
2014 09 25268 David Swank has installed the cron job "UpdateDaq" to check the data collection and if there is any problem to kill the current screen and start a new screen with all data collection operating. I tested it at 1720Z and it worked. I'll continue testing.
2014 09 29272 Working with cron reboot. I needed to sort out the PATH and non-local job calling.
2014 10 01274 Crontab procedure finally is running okay. This took a considerable amount of time. Absolute addressing, Path, and aliases needed to be refined. Today the daq software will be stopped on a regular basis so it can start up by the cron task.
2014 10 03276 For the last 24 hrs I have been exercising the crontab procedure. Last night it rebooted once (0520z) when the rsr apparently failed and once again (0920z) when the adc failed.
2014 10 04277 1320z data check. No reboots in the last 24 hrs.
2014 10 12285 We're just into watching the data collection. The cron log files are being collected as /tmp/tmpchk. Maybe I can increase the cron rate to say 15 min.
2014 10 13286 END SGP14 DATA COLLECTION AT [2014 10 14 13 07 55 Z].
Proceed to takedown procedure.
2014 11 1305 Langley analysis from a few good sun rise/sets is complete.

top

SGP14 DATA

The major gaps are (240-253) The heater circuit on the FRSR failed and the CDU was returned to the manufacturer for repair. (265-275) The software was upgraded with improved source code and a crontab reboot process was installed. System operation after this was significantly improved.

top

MFR Performance


A set of 10 sweeps delivered on a clear morning. The shadow value, the minimum, is the sky irradiance and is almost constant. The edge values are direct beam and sky irradiance and increase as the sun rises in the sky. The edge and shadow values are used to compute the optical depth.

Left panel: Optical depth (OD) and aerosol optical depth (AOD) for Aug 12, a particularly clear sky day. Right panel: The AOD for channels 2 and 3 on the same day.

top

FRSR TOA CALIBRATIONS

This table compares TOA V0 calibrations.

The Lamp calibration (CAL) is derived by taking the lamp off/on values for each channel. The lamp spectrum is convolved with the filter response for each channel to arrive at a channel gain/offset, the ratio of volts from the head channel to the computed input irradiance. This gain is multiplied by the channel electronic gain and offset from the FRSR electronic calibration.

The Langley value of V0 (SGP14) is derived by a standard Langley analysis of quality clear sky sun rises and sun sets.

CHAN NOM LAMP CAL SGP14 STD MLO14
1 open
2 415 416 572.9 1590.4113
3 500 503 570.7 774.440
4 615 613 496.4 567.820
5 680 673 307.2 333.110
6 870 872 371.2 353.28
7 940 940 920.2 316.452

There are large differences between the lamp cal and the Langley calibration values. This is in keeping with other calibration studies.

top

SETUP FILE

The PRP2 software make use of a text setup file called su.txt. This file has all necessary parameters for running the software for all the modules.

LAST EDIT: 2014,08,08
EXPERIMENT NAME: SGP14
GEOGRAPHIC LOCATION: ARM-SGP
PLATFORM NAME: RAD FIELD
LOCATION ON PLATFORM: RAD FIELD 
HEIGHT ABOVE SEA LEVEL (m): 100
PRP2 SERIAL NUMBER: 201
PRP SERIAL NUMBER: 001
FRSR SERIAL NUMBER: 1
HEAD SERIAL NUMBER: 374
RAD MODEL NUMBER: RAD-17c
RAD SERIAL NUMBER: 209
RAD PSP SERIAL NUMBER: 33841F3 (8.91)
RAD PIR SERIAL NUMBER: 33687F3 (3.63)
RA2 MODEL NUMBER: RAD-17c
RA2 SERIAL NUMBER: 218
RA2 PSP SERIAL NUMBER: 34292F3 (8.67)
RA2 PIR SERIAL NUMBER: 35836F3 (3.23)
SPN SERIAL NUMBER: A349
SP2 SERIAL NUMBER: A925
TCM SERIAL NUMBER: 17272
GPS MODEL NUMBER: none
GPS SERIAL NUMBER: ???

  Heading source = TCM, COG, FIXED
HEADING SOURCE: FIXED
TCM FIXED PITCH: 0.1;
TCM FIXED ROLL: -0.1
TCM FIXED HEADING: 12

  Location source = GPS, FIXED
  SGP=(36.605,-97.485,12.0), 
  SEATTLE=(47.60329,-122.28797,18.1)  
GPS SOURCE = FIXED
GPS FIXED LATITUDE: 36.605
GPS FIXED LONGITUDE: -97.485
GPS FIXED VARIATION: 12

   	Data collection PC
PC IP: 198.124.103.52
PC MASK: 255.255.255.0
PC GW: 198.124.103.1
PC DNS1: 198.124.99.97
PC DNS2: 198.124.99.74
TIME SERVER: 198.124.99.74
  	Main serial server
SERIAL HUB URL: 10.1.0.29
SERIAL HUB MASK: 255.255.255.0
SERIAL HUB GW: 10.1.0.1
	Secondary hub for RAD2 with SPN
SERIAL HUB2 URL:  10.1.0.30
SERIAL HUB2 MASK: 255.255.255.0
SERIAL HUB2 GW: 10.1.0.1

PRP COMMENTS:
  SGP14 calibration/validation exercise.
  Before ACAPEX cruise on Ron Brown
END
RSR COMMENTS:
  140805,10--frsr is running okay but the shadowband has a 
   little wiggle. Needs a look.  
END
RAD COMMENTS:
  140804--setup at SGP14
    The rad cdu had suffered a shock and needed repair. 
    The PSP channel was not operating so it was returned to 
    Seattle for repair.
END
RA2 COMMENTS:
  140805--setup next to RAD at SGP14.
   We had to opened the box and found the power connector had been 
    jarred loose. When plugged back in it worked well.
END
TCM COMMENTS:
 This is a fixed site. The FRSR is aligned to N.
 Head connector pointing to north.
END
GPS COMMENTS
 This is a fixed site and the GPS was not used.
END
ADC COMMENTS
 ADAM 4017 ADC
  Chan0 = spn total volts
  Chan1 = spn diffuse volts
END
SPN COMMENTS:
 Operates with PRP2/RAD setup.
 140805-system setup at SGP14.
 SN# A349
END
SP2 COMMENTS
  With RAD2 system.
  140805--setup at SGP14 next to RAD
END
	

top


HEMISPHERIC BROADBAND COMPARISONS WITH SGP

The NetCDF time series from ARM DMF were downloaded. For the comparisons below, the data files were of the type
    sgpqcrad1longC1.c1.20140821.000000.cdf.

SHORTWAVE DOWNWELLING HEMISPHERIC IRRADIANCE

Two time series were selected:
    BestEstimate_down_short_hemisp
and
    down_short_hemisp

Left panel Comparisons between shortwave time series. The blue points are the difference between the SGP “best estimate” sw and the sw series that are defined above. The red points are the difference between the RAD1 and the RAD2 sw values. Right panel A plot of the differences between the two SGP archive estimates of down_short_hemisp. There is a pronounced deviation for irradiances greater than 800 W/m^2.

All time series were filtered for daytime, clear-sky conditions. (Clear-sky is defined as one-minute standard deviation less than 20 W/m^2.) Left panel Difference in shortwave between RAD1 and RAD2. Linear correlation is good (0.2%) but RAD1 is lower than RAD2 by 4 W/m^2. Middle panel Difference in shortwave between RAD and SGPbest. The same bias applies here. Right panel Difference in shortwave between RAD2 and SGPbest. It seems that, compared to SGPbest, the RAD1 shortwave data are biased by about -4 W/m^2.


    sw1(corrected) = sw1*1.047 + 4.04
    sw2(corrected) = sw2*1.044 + 0.2

After the corrections above the clear day comparisons with SGPbest are unbiased. The one-minute stdevs of approx 10 W/m^2 persist.

LONGWAVE DOWNWELLING HEMISPHERIC IRRADIANCE

The SGP longwave time series was derived from the sgpqcrad1longC1.c1.*.cdf NetCDF file, variable down_long_hemisp.

The scatter plots show the differences between the three data sources. In all cases we compare the derived LW irradiance from the PIR,Tcase,Tdome combination.
Blue points: Compare RAD2 and the SGP LW points. Mean=3.4 ± 3.9 W/m^2.
Red points: Compare RAD1 and the SGP LW points. Mean=1.4 ± 2.5 W/m^2.
Green points: Compare RAD1 and the RAD2 LW points. Mean=1.6 ± 3.0 W/m^2.

In all cases the mean differences are of the order 1%.

top

AOD COMPARISONS WITH NIMFR

Apparently there were issues with the MFRSR for this time period so the NIMFR was recommended as the best choice with which to make a comparison. NetCDF data from the ARM archive was downloaded for analysis. File names were "sgpnimfraod1michC1.c1.20140809.000000.cdf" where each file contains one day's data.


The MFR head calibration is dependent on the internal temperature. (Left panel) The NIMFR temperature varies considerably. We question if this is normal or if there is a problem here. (Right panel) The FRSR does a much better job of controling the head temperature to a few tenths of a degree. Ambient temperatures varied from around 18C at night to 36C in the day.

The direct normal irradiance for each filter is included in the NetCDF files. A Langley analysis was performed on these series for the exact same times as were used for the FRSR. The resulting optical depths are shown below.

We selected a very clear day, 12 Aug, JD224, for comparisons.

The FRSR values are spaced each six seconds while the FRSR are 20 seconds. Thus the FRSR lines are more dense.

IN SUMMARY: The NIMFR and the FRSR give equivalent data but subject to having very good Langley calibration values. In the example below the sensitivity to uncertainties in i0 is examined.

This is an example of the sensitivity of computed OD to the calibration factor. The 500 nm NIMFR channel is shown here. A Langley analysis , based on two B mornings, gave i0 = 1.55-1.81 depending on how one might choose the extrapolation. OD was computed using i0=1.6 and i0=1.7 and the result is plotted here. In this case a 6% difference in i0 results in an OD change of 13%. The calibration table above for SGP14 for seven A or B mornings has stdev uncertainties of 5% to 2%. The upcoming MLO deployment we expect will improve the suite of calibrations.

top

AOD COMPARISONS WITH MFRSR C1

On Friday 10/31 Conner Flynn provided good background on the SGP instrumentation. (a) Better to not analyze NIMFR during this period because it had several technical issues, head changes, etc. (b) The MFRSR data is now available and looks very good. (c) Good series to analyze are sgpmfrsraod1michC1 and sgpmfrsraod1MichE13. These instruments are co-located and pass QA tests.


The MFR head temperature, while not as wild as the MIMFR was still variable. Instead of a set point of 40 C, the mean was 42.7 and the stdev was 0.36 C (compared to the FRSR at 40 +/- 0.28).

Ad Hoc Langley Analysis

The direct-normal time series from mfrsr C1 was searched for good Langley mornings or evenings. Seven were noted and Langley plots were analyzed for each time period. The five plots here are one of the seven Langley time periods.

The Langley calibration values were (mean/stdev):

Filter 1 2 3 4 5
Channel 2 3 4 5 6
Wavelength 415 500 615 670 869
Ad Hoc Mean 1.7936 1.9768 1.6812 1.5019 0.9578
Ad hoc stdev 0.0885 0.0747 0.0450 0.0357 0.0179
NetCDF i0 1.177 1.726 1.586 1.409 0.871

The "ad hoc" data is constructed from the Langley analysis over the SGP14 time period. The NetCDF data are the i0 values computed from the NetCDF file and are

    i0x = direct_normal_narrowband_filterx / nominal_calibration_factor_filterx, x=2,...6

    ODx = log(i0x / nx) / AM

Where AM is the atmospheric mass and nx is the direct normal irradiance for filter x. The ad hoc calibration values are used to compute OD below. We use the same clear day as above, 12 Aug, JD224, for comparisons.

Typically, while the ODs for FRSR are close to the MFRSR, the error is not completely explained by errors in the Langley calibration value. Differences between the FRSR and the SGP MFRSR were typically of the order of 14%. The stdevs in the MFRSR i0 values were on the order of 5% and for the FRSR analysis were of the order 4%. Thus the OD errors on the order of 7-8% are expected. (This is all off the top of the head thinking.)

SUMMARIZE: (1) Based on ad hoc Langley analysis, FRSR agrees with MFRSR to about 15% uncertainty. (2) We expect the uncertainty will improve with improved Langleys from MLO.

top


MLO14—CAL/VAL AT MAUNA LOA OBSERVATORY (MLO)

Photos taken from the MLO web site. Left panel: View from MLO in the morning facing east. Right panel: View from MLO facing west toward the volcano top.

Photos provided by Paul Fukumura, NOAA. (email 141018) Left panel: Shows the entire deck is facing North with Mauna Kea in the back ground. At the far end of the deck is a storage container that contains the network switch where the laptop is connected to the internet. There should be room there to store the laptop. Right panel: Shows the East rail and the aluminum grating where we may set up. The width of the grating is 18 inches. The length of the grating is approximately 30 feet. There are electrical outlets located along the rail.

Country: United States Country Flag
Latitude: 19.5362° North
Longitude: 155.5763° West
Elevation: 3397.00 masl
Time Zone: Local Time + 10.0 hour(s) = UTC

Contact Name: John Barnes
Address: NOAA - Mauna Loa Observatory
1437 Kilauea Ave. #102
Hilo, Hawaii, 96720, United States
Phone: (808)933-6965    Fax: (808)933-6967

Shipping address (Hilo Office):
Mauna Loa Observatory
1437 Kilauea Ave., #102
Hilo, Hi 96720
Attn: Paul Fukumura-Sawada
Phone: 808-933-6965 ext. 223

top

MLO LOG

2014 09 15Visit permit approved.
2014 09 15Plan travel and make reservations.
2014 10 13 Sent Paul.Fukumura@noaa.gov email to introduce the deployment.
2014 10 16Follow up call to Paul. Talked with Daryl. They are aware of our coming and have selected a place.
2014 10 18Received photos and comments from Paul. The deployment looks like a snap.
2014 10 27Planning travel. Request funds for a longer stay on site.
2014 10

top


STATEMENT OF WORK

RFP No. 4-BB47-P-00231-00

Note: The ARM SOW is in blue and RMR responses are in black.

Development, Preparation & Deployment of PRP System:

Task 1:

Provide a list of critical spares for operation of the PRP aboard the R/V Ron Brown.
Completion date: 15 July 2014

This will be a spread sheet with all PRP parts, including procurement information and needed spares will be marked. Note, this is a single leg, 30-day cruise so the list of critical spares will be small.

Task 2:

Work with ARM Site Data Systems (SDS) to ensure all operating systems of instrument virtual machines run on CentOS6 or Windows7.
Completion date: 31 October 2014

Work with the IT people to be sure their CentOS has all needed Linux, Perl and Expect modules installed. If possible, connect remotely to the AMF2 VM and check operation. During the deployment of PRP on the ship we will install PRP software and assure it is operating properly.

Task 3:

Standardize all software to run as non-privilege OPER user account.
Completion date: 31 October 2014

See above. This software is basically the same as was used during MAGIC. More emphasis will be made in this contract to prepare operation manual and guidance. The MAGIC software will be upgraded to make it more user friendly.

Task 4:

Standardize, to ARM standards, all Ethernet hardware, serial to Ethernet hardware.
Completion date: On or before installation, i.e. January 2015

We will do this, but during MAGIC we were given hardware names of equipment that was too large physically to fit into the compact PRP enclosure. If IT cannot provide me with a suitable hardware recommendation we will need to prepare a separate waterproof enclosure. I have bid on this extra effort.

Task 5:

Work with the ARM Data Quality Office (DQO) to review and update the current quality control parameters.
Completion date: 31 October 2014

We will review the entire NetCDF specification and any other QA parameters. If necessary we can adapt the software to produce desired QA output.

Task 6:

Deliver an up-to-date ARM Instrument Handbook (Mentor) in the provided format.
a. Clear and concise installation instructions
b. Written directions, with picture examples, on how to start, stop, switch screens and troubleshooting of any known or previous issues.
c. Written Preventative Maintenance procedures.
Completion date: all sections completed for review by 30 September 2014; final revisions by 31 October 2014

Note the “Instrument Handbook Template” and the “ARM Data Quality Report Guide” were not included with the SOW. However, we are familiar with these. There exists several manuals for the PRP and for PRP data processing. The requirement that these be delivered in Word format is a problem because they are written in LaTeX.

Task 7:

Work with the ARM Data Management Facility (DMF) developers to ensure collections and ingests are working.
Completion date: no later than 1 week before the ACAPEX operational start date, i.e. January 2015

We are experienced with this from MAGIC and anticipate no significant issues.

Task 8:

Deliver all calibration results and corresponding metadata to the Calibration Database in the ARM Operations Status System (OSS).
Completion date: 27 February 2015.

ANL ships the PRP2 to SGP as soon as possible. Two calibrated PSP and PIR radiometers are delivered to SGP as soon as possible. Two calibrated SPN2 systems delivered to SGP. System preparation (July-Oct) Setup the system at SGP and confirm complete system operation. Electronic cal of both RAD systems. Integration of radiometers. Run system for burn-in period. Field calibration (Nov)
(d) Pack and ship to calibration site, Mauna Loa Observatory, Hawaii.
(e) Travel (TRIP1) to calibration site.
(f) Process calibration data. (Note 2) Note 1 — TRIP 1: We propose to field calibrate at the NOAA facility on Mauna Loa Observatory (MLO). In this way the field calibration trip and the ship deployment are combined to a single trip. This will be a significant savings to the program. Note 2 — Calibration data: AOD calculations are done on the FRSR collected during post processing. Therefore we can calibrate just prior to the cruise then apply the calibration data during post processing.

Task 9:

Install to an operational status the PRP system aboard the NOAA R/V Ron Brown.
Completion date: before the ACAPEX operational start date, i.e. January 2015

We will deploy the PRP (including two RAD-SPN modules) in Honolulu and assure data collection and data system connectivity is operational. Operator training will take place during the installation. We will be able to complete the install after the calibration visit to MLO.

Task 10:

Provide hands on training for installation and maintenance aboard the ship during the installation for AMF2 technicians, who will operate and maintain the PRP system.
Completion date: January 2015

Yes. During deployment. An operation manual will be provided to the technicians.

Data Quality Analysis, Processing and Reporting:

Task 1:

Monitor the PRP data quality during ACAPEX operational period daily through quicklooks and available data.
Completion date: February 2015

Yes.

Task 2:

Use Data Quality Problem Reports (DQPRs) to report instrument or measurement problems daily.
Completion date: February 2015 !

Yes.

Task 3:

Respond to DQPRs daily during the ACAPEX operational period.
Completion date: February 2015

Yes.

Task 4:

Resolve all DQPRs for ACAPEX data.
Completion date: 31 Mar 2015 !

Yes.

Task 5:

Complete and submit ARM Data Quality Report for DQPRs as assigned.
Completion date: 31 Mar 2015 !

Yes.

Task 6:

Analyze the ACAPEX PRP data for quality control and report in the ARM Instrument Mentor Monthly Summary (IMMS) Reporting System.
Completion date: 31 Mar 2015

Yes.

top


TASK LOG

Task 0: General effort for all tasks.

Task 1: Critical spare parts.

Task 2: CentOS compatibility.

Task 3: Software Compatibility.

Task 4: ARM Hardware.

Task 5: DQO QC Params.

Task 6: Handbook.

Task 7: DMF Ingest.

Task 8: Calibration and Results.

Task 9: RHB Install.

Task 10: Training.

QA Task 1: Monitor performance.

  • QA Task 2: Daily DQPR Report.

  • QA Task 3: Respond DQPR daily.

  • QA Task 4: Resolve DQPR daily.

  • QA Task 5: Submit DQPR.

  • QA Task 6: IMMS Monthly Report.

  • 140815—Submitted report.
  • 140914—Submitted report.