Docstoc

DRAFT

Document Sample
DRAFT Powered By Docstoc
					Draft Regional Report QA Section Template




                                            DRAFT
          SWAMP REPORT TEMPLATE – QA NARRATIVE




The following is a template for a QA section narrative for reports. Following this template
ensures consistent summary and reporting of QA data.



Regions requiring assistance with the QA section of their annual report may contact the Data
Management or Quality Assurance Team. Availability of assistance is dependent on workload at
the time of request. Regions requesting assistance should submit a list of datasets (by fiscal year)
to be incorporated in the report and the date the narrative should be complete.

                                             Page 1 of 7
Draft Regional Report QA Section Template




QUALITY ASSURANCE & QUALITY CONTROL (QA/QC)

The data generated by SWAMP may be used to determine the status of beneficial uses throughout the
state, assess trends, make regulatory and management decisions, and support enforcement of policies.
Thorough objectives for achieving quality data are outlined in the SWAMP Quality Assurance
Management Plan (QAMP). In general, data quality is demonstrated through analysis of the following
Data Quality Indicators:

       Laboratory method blanks
       Surrogate spikes
       Matrix spikes and matrix spike duplicates
       Certified reference materials/laboratory control spikes
       Laboratory Duplicates
       Field duplicates


Data for Project IDs [identify specific project IDs that have been verified] have been verified according to
SWAMP Standard Operating Procedures (SOPs) for chemistry and field data verification. The data
verification process determines whether the data are compliant with the individual measurement quality
objectives (MQOs) specified in the SWAMP QAMP. Data are classified as compliant with the SWAMP
QAMP, estimated, non-compliant with the SWAMP QAMP, or rejected if the data were rejected by the
reporting laboratory.

[Identify data from other project ID’s that have not been verified, specifying that the data should be
considered draft]

No data have been validated. This section does not attempt to determine whether or not data should be
used. That can only be done after data validation and comparison to project-specific data quality
objectives (DQOs).

Please note that estimated data is defined as data batches or sample results failing to meet all program
DQIs specified in the SWAMP QAMP, have analytes not covered in the SWAMP QAMP, or are
insufficiently documented such that supplementary information is required for them to be used in reports.
During the DQA phase of reporting, end users may find estimated data batches meet project data quality
objectives.

The percent recovery (%R) acceptance criteria for surrogate spikes, matrix spikes and matrix spike
duplicates, certified reference materials/laboratory control spikes, and relative percent difference (RPD)
acceptance criteria for both laboratory and field duplicates are presented in Appendix X Table X.

Laboratory Method Blanks

Laboratory method blanks were used to assess laboratory contamination introduced during sample
preparation and analysis. The method blanks were processed in a manner identical to the associated field
samples. According to the SWAMP QAMP for conventional, organic and inorganic analyses, at least one
                                                Page 2 of 7
     Draft Regional Report QA Section Template


     laboratory method blank should be analyzed per 20 samples or one per batch, whichever is more frequent.
     Batches that did not include laboratory method blanks at the required frequency were classified as
     estimated (Appendix X, Table X) [specify batches that did not include method blanks].

     Water laboratory method blanks were considered acceptable if target analyte concentrations were below
     their respective method detection limit (MDL). All laboratory method blanks were acceptable with the
     exception of XX [numerically identify the total number of blanks above the MDL] blanks in which
     concentrations of target analytes were above the MDL but less than the reporting limit (RL) (Appendix X,
     Table X) [Identify the data batches that contained method blanks above the MDL in Table X]. These data
     were classified as compliant with regard to the SWAMP QAMP MQO for laboratory blanks.

     Also, XX [numerically identify the total number of blanks above the RL] laboratory blanks contained
     detectable levels of target analytes above the RL (Appendix X, Table X)..These data may be flagged as
     estimated.

     Surrogate Spikes

     Surrogate spikes were used to assess analyte losses during sample extraction and clean-up procedures,
     and must be added to every field and quality control sample prior to extraction. Whenever possible,
     isotopically-labeled analogs of the analytes should be used.

     [Identify in Table X data batches that did not contain the required frequency of surrogate spikes and in a
     separate table specify data batches containing surrogate spike concentrations outside of acceptance
     criteria.]

     The associated analytes in Appendix X, Tables X and X were classified as estimated with regard to the
     SWAMP QAMP MQO for surrogates.

     All surrogate %Rs were within the acceptance criteria identified in Table X.

     Matrix Spikes and Matrix Spike Duplicates

     A laboratory-fortified sample matrix (MS) and a laboratory-fortified sample matrix duplicate (MSD) were
     both used to evaluate the effect of the sample matrix on the recovery of the target analyte(s). Individually,
     these samples were used to assess the bias from an environmental sample matrix plus normal method
     performance. In addition, these duplicate samples can be used collectively to assess analytical precision.

     Aliquots of randomly selected field samples were spiked with known amounts of target analytes. This
     process was repeated for a subset of field samples to create MSDs.

     The success or failure of the matrix spikes is evaluated by calculating the percent recovery. Where:



      % recovery
                        vMS    vambient   100
                               vspike



                                                   Page 3 of 7
Draft Regional Report QA Section Template


        vMS: the concentration of the spiked sample

        vambient: the concentration of the original (unspiked) sample

        vspike: the concentration of the spike added

In addition to the recoveries, the relative percent difference (RPD) between the MS and MSD is
calculated to evaluate how matrix affects precision.




                                    RPD 
                                              v MS           100
                                                        vMSD
                                                   mean

.
        vMS: the concentration for the matrix spike
                         
        vMSD: the concentration of the matrix spike duplicate
        mean: the mean of the two concentrations (MS + MSD)
The MS/MSD %R and RPD acceptance criteria are presented in Appendix X, Table X.

According to the SWAMP QAMP for conventional, organic and inorganic analyses, at least one
MS/MSD pair should be performed per 20 samples or one per batch, whichever is more frequent. X
percent of the batches [X out of Y total batches] did not include MS/MSDs performed at the required
frequency. These batches were classified as estimated (Appendix X, Table X). [Identify the total number
of batches that did not include MS/MSDs at the required frequency]

Laboratory batches with MS/MSD %R and RPD values outside of acceptance criteria are presented in
Appendix X, Table X [Identify the total number of batches that contained %R and RPD values outside of
acceptance criteria]. All other MS/MSD %Rs and RPDs were within acceptance criteria. If the MS or
MSD is spiked too high or too low relative to the ambient concentration, the calculated recoveries are no
longer an acceptable assessment of analytical bias.

Certified Reference Materials, Laboratory Control Materials, and Laboratory Control Samples

Certified reference materials (CRMs), laboratory control materials (LCMs), and laboratory control
samples (LCSs) were analyzed to assess the accuracy of a given analytical method. As required by the
SWAMP QAMP, one CRM, LCM, or LCS should be performed per 20 samples or one per batch,
whichever is more frequent. X percent of the batches (X out of Y total batches) did not include CRMs,
LCSs, or LCMs performed at the required frequency. These X batches were classified as estimated
(Appendix X, Table X) [Specify in the table the batches that did not contain CRMs, LCSs or LCMs run at
the required frequency].

The accuracy of the results is assessed through the calculation of a percent recovery.

                                                 Page 4 of 7
Draft Regional Report QA Section Template


                                                             vanalyzed
                                       % recovery                       100
                                                             vcertified
Where:

         vanalyzed: the analyzed concentration of the reference material
                           
         vcertified: the certified concentration of the reference material

The CRM, LCM, and LCS %R acceptance criteria are presented in Table X.

The CRM, LCM, and LCS recoveries outside acceptance criteria are presented in Table X, these data will
be classified as estimated [Identify the batches the contained CRM, LCM and LCS recoveries outside of
acceptance criteria]. All other CRM, LCM, and LCS %Rs were within acceptance criteria.

Laboratory Duplicates

Laboratory duplicates (DUPs) were analyzed to assess laboratory precision. As required by the SWAMP
QAMP, at least one duplicate of a field sample should be performed per 20 samples or one per batch,
whichever is more frequent. X percent of the batches (X out of Y total batches) did not include DUPs
performed at the required frequency. These X batches were classified as estimated (Appendix X, Table X)
[Specify the batches that did not contain laboratory duplicates run at a minimum frequency of 5%].

Following analysis, the results from the duplicate samples are evaluated by calculating the RPD.


                                       RPD 
                                                 v sample    vduplicate    100
                                                        mean


Where:
                           
         vsample: the concentration of the original sample digest
         vduplicate: the concentration of the duplicate sample digest
         mean: the mean concentration of both sample digests

Laboratory batches with RPDs >25% were classified as estimated (Table X) [Identify laboratory batches
containing laboratory duplicates outside of the acceptance criteria]. All other DUP RPDs were within
acceptance criteria.

In order for the RPD to be an indicator of precision, the concentration of the duplicate samples must be
moderately higher than the detection limit of the utilized analytical method. If the sample concentration is
low, and if the difference between duplicate concentrations is approximately 3-5 times the MDL, the
resulting RPD may be higher than the associated SWAMP MQO. This result may lead to the data being
classified as estimated.
                                                    Page 5 of 7
Draft Regional Report QA Section Template


Field Blind Duplicates/Field Duplicates

Field blind duplicates and/or field duplicates were analyzed to assess variability introduced by field
sampling procedures. Field blind duplicates and/or field duplicates were sampled at X stations for waters.
Water samples were taken by collecting a separate grab sample immediately following the collection of
the field sample. Field duplicate values were compared to field sample values from each site and RPDs
were calculated as in Section X.

The precision of field duplicates is evaluated by calculating the RPD between the involved samples.



                                           RPD 
                                                      vfield sample    vfield duplicate     100
                                                                  mean

Where:

                                
         vfield sample: the concentration of the original field sample

         vfield duplicate: the concentration of the field duplicate



RPDs <25% were considered acceptable as specified in the SWAMP QAMP. Laboratory batches with
RPDs >25% were classified as estimated (Table X) [Identify batches containing field blind
duplicates/field duplicates outside of acceptance criteria]. All other RPDs were within acceptance
criteria. If the field duplicate RPD is >25% but the concentration of the duplicate is <RL, the calculated
recovery is no longer an acceptable assessment of analytical precision.

Toxicity Tests

[Identify deviations in water quality parameters or test conditions; including improper temperature upon
receipt and exceedance of holding times. Identify total number of batches that did not meet toxicity
control criteria.]

Field Data Measurements

The procedures followed when conducting routine field data measurements for SWAMP can be found in
Appendix E of the SWAMP QAMP. Field equipment used to take field data measurements is required to
be calibrated within 24 hours of use and within 24 hours after field measurements activities are
performed. Per the SWAMP QAMP the following equipment should be calibrated at a minimum; titration
equipment, thermometers, DO meters, pH meters, conductivity meters, and multi-parameter field meters.
After post-calibration checks are performed, the percent drift should be evaluated. If data has been
collected outside of compliance (% drift is outside criteria found in Appendix E of the SWAMP QAMP),
it should not be reported unless it has been flagged to indicate non-compliance.



                                                      Page 6 of 7
Draft Regional Report QA Section Template


[Identify Project IDs that included field data measurements and identify the measurement type; e.g.
dissolved oxygen, pH, salinity etc. Specify the total number of results that were classified as estimated due
to probe failure, inability to deploy instrumentation, field calibration not performed at the correct
frequency, velocity too low to measure, or no documentation of field measurement collection existed.]

Holding Times

X% of the results (X out of Y total results) in X number of samples (samples per method) were classified
as estimated due to holding time exceedances. These results consisted of ammonia, TSS, TOC, BOD,
orthophosphate, nitrate and nitrite analyses, water and sediment organics (diesel range organics,
orthophosphate pesticides, organochlorine pesticides, PCBs, PAHs, and pyrethroids), metals and water
toxicity analyses.

[Identify sample types that exceeded mandated holding times, specifying the applicable holding time for
each sample type, this is an optional table to be included in the appendices.]

QA/QC Summary

Data that meet all SWAMP Measurement Quality Objectives (MQOs) as specified in the SWAMP
QAMP are classified as “SWAMP-compliant” and considered usable without further evaluation. Data
that fail to meet all program MQOs specified in the SWAMP QAMP, have analytes not covered in the
SWAMP QAMP, or are insufficiently documented such that supplementary information is required for
them to be used in reports are classified as “estimated” or non-compliant with the SWAMP QAMP.
Rejected data batches do not meet minimum requirements and /or have gross errors or omissions; these
data were classified as rejected when the reporting laboratory rejected the data. During the Data Quality
Assessment (DQA) phase of reporting, end users may find estimated data batches meet project data
quality objectives.

There were X [identify total sample count] sample results; including field measures, grab and integrated
samples, field blind duplicates, and field blanks of which X were classified as compliant, Y were
classified as estimated and Z were classified as rejected.




                                                Page 7 of 7

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:108
posted:5/15/2010
language:Galician
pages:7