Analysis of Sediment Samples for USEPA Region IV: Science and Ecosystem Support Division Fenholloway River

Size: px

Start display at page:

Download "Analysis of Sediment Samples for USEPA Region IV: Science and Ecosystem Support Division Fenholloway River"

Abraham Hill
6 years ago
Views:

1 Analysis of Sediment Samples for USEPA Region IV: Science and Ecosystem Support Division Fenholloway River Submitted To : Mr. Archie Lee USEPA Region IV Science and Ecosystem Support Division 980 College Station Road Athens, Georgia (706) Submitted By : Ms. L. Kellie Dixon Mote Marine Laboratory 1600 Ken Thompson Parkway Sarasota, Florida (941) April 27, 1998 Mote Marine Laboratory Technical Report No. 572 This document is printed on recycled paper. Suggested reference Dixon LK Analysis of sediment samples for USEPA Region IV: Science and Ecosystem Support Division Fenholloway River. USEPA Region IV. Mote Marine Laboratory Technical Report no p. and appendices. Available from: Mote Marine Laboratory Library.

2 TABLE OF CONTENTS Table of Contents i List of Tables i Background Summary of Analyses Analyses and Methodology Page Total Solids (Percent Solids) and Percent Moisture... 3 Total Volatile Solids (Percent Organics)... 3 Grain Size Distributions, Raw Sample... 3 Grain Size Distributions, Inorganic Fraction... 4 Literature Cited LIST OF TABLES Page Table 1. Analyses, methodology, holding times, method detection limits (MDL), and units of reporting for sediment samples. Practical quantitation limits (PQL) are defined as four (4) times the MDL Table 2. Data quality objectives for sediment analyses Table 3. Listing of half-phi intervals and equivalent µm sizes Appendix A. Custody Records. Appendix B. Tabular data for sediments from the Fenholloway River, Florida Appendix C. Field structure, units, and descriptions of magnetic files supplied.. Appendix D. Volume percent distributions of the total, inorganic, and organic fractions (by difference) of sediments from the Fenholloway River, Florida. i

3 Background The United States Environmental Protection Agency (USEPA) Region IV, Science and Ecosystem Support Division, collected samples in the Fenholloway River of Florida in March of USEPA requested Mote Marine Laboratory (MML) perform determinations of percent moisture and percent organics (by combustion). Additionally, sediment grain size distributions (as % volume) were to be measured by optical methods on both the raw sample and on a fraction from which organic matter had been removed, with the size distribution of organic material computed by difference. Samples were transferred, frozen, to MML by common carrier (Chain of Custody Records, Appendix A). Analyses were conducted under MML s Comprehensive Quality Assurance Plan (CompQAP), approved by the Florida Department of Environmental Protection (#870216). Summary of Analyses Analyses requested, methodology references, holding times, and method detection limits (MDL) appear in Table 1, while project data quality objectives are summarized in Table 2. No project specific audit requirements are in place for this project as analyses are being conducted under MML s CompQAP. Frequency of acceptable results on QC samples was met or exceeded for all parameters. Tabular data for all parameters appear in Appendix B and field order, units, and descriptions of magnetic files supplied appear in Appendix C. Graphic depictions of sediment grain size results appear in Appendix D.

4 Table 2. Analyses, methodology, holding times, method detection limits (MDL), and units of reporting for sediment samples. Practical quantitation limits (PQL) are defined as four (4) times the MDL. Holding Analyses Method Ref Time MDL Units Total Solids 2540G SM-18 3 mo 0.5 % Total Volatile Solids 2540G SM-18 3 mo 0.5 % Percent Moisture 2540G SM-18 3 mo 0.5 % , Grain Size 12 mo Percent Sand Folk % Percent Silt Folk % Percent Clay Folk % Whole Sample Statistics Coulter 1994 Median 0.5 µm Mean 0.5 µm Mode 0.5 µm Standard Deviation 0.01 µm Skewness NA -- Kurtosis NA -- Volume Percent 0.2 % Table 3. Data quality objectives for sediment analyses. QC Sample Solids Frequency Laboratory Duplicates 1 per 20 samples <20% RSD Grain Size Laboratory Duplicates 1 per 20 samples <20% RSD Data Quality Objectives Continuing Calibration every 20 samples % recovery of certified Verification (CCV) value 2

5 Analyses and Methodology Total Solids (Percent Solids) and Percent Moisture Analysis of samples for total solids (percent solids) and percent moisture followed 2540G of Standard Methods, 18 th Edition (APHA et al., 1992). Aliquots of homogenized sample were apportioned into predried, tared crucibles, dried at C to a constant weight. Total Volatile Solids (Percent Organics) Analysis of samples for total volatile solids (percent organics) also followed 2540G of Standard Methods, 18 th Edition (APHA, 1992). Dried sediments from the total solids determinations were ashed for 1 hour at 550 C ± 50 C. Grain Size Distributions, Raw Sample Grain size distributions of field moist sediment were determined using a laser diffraction instrument (Coulter LS-200), capable of measurement between 0.4 and 2,000 µm equivalent spherical diameters. In this instrument, the angle and intensity of laser light scattered by a solution of sediment sample are selectively measured and converted to volume distributions based on a Fraunhofer optical model. Similar to other methods of particle sizing (pipette or hydrometer analyses), the optical model is based on assumptions of partial sphericity. During operation, filtered tap water is used for background determinations and sample resuspensions. Samples are homogenized and representative portions introduced to the sample chamber. Samples are recirculated for 60 seconds, and then analyzed for 60 seconds. Repetitive analyses of selected sample aliquots confirmed that a 60 second analysis time was sufficient for reproducible data. The recirculation time was determined by reanalysis to be sufficient for samples distributions to stabilize (destruction of loose agglomerates). Surfactants have previously been demonstrated not to affect distributions and so were not employed. Sonication, on the other hand, produces extensive changes in sample size distribution, with larger particles continuing to decrease and smaller ones continuing to increase as continued sonication disrupts more and more of the fragments within the sediment. Extensively sonicated sediments were not considered to be representative of the in situ material collected and so no sonication was employed. Duplicate evaluations are a separate aliquot from a sample jar introduced into the instrument. As sample aliquots are comparatively small (l-2 g wet weight), low or nonrepresentative concentrations of coarser fragments which are not readily homogenized will produce variations which are more extensive than from a more uniform sediment. Continuing calibrations are performed with against glass beads of known mean grain size.

6 Results are presented in 93 logarithmically distributed size channels as the volume percent of the entire sample within that spherical size range. Within rounding error, the sum of volume percents from all size ranges will total 100 %. For purposes of clarity, the 93 channels have been combined into 26 intervals (Table 5), still totaling 100 %, which represent the classical half-phi distribution (Folk, 1974), in which: As the instrument is sensitive only to 2,000 µm (2.000 mm), sediments are sieved through a 2 mm mesh prior to diffraction analysis. If any material is retained by the screen, then a larger sample aliquot is weighed (field moist), wet sieved through a 2 mm mesh, and the coarse material retained is dried and ashed as in the determination of percent solids and percent organics above. (In these samples, the particles which fail to pass a 2 mm sieve are generally either woody debris or large shell fragments or intact shells from small bivalves. It varies by sample as to whether the larger shell fragments are a representative part of the-sample.) For calculations, all sediments greater than 2,000 µm are assigned to the range between and proportionally incorporated into the results of the diffraction analysis, for presentation of the results on the entire sample. Total percent sand, silt and clay are calculated as the sum of volume percent between and 62.5 µm, 62.5 and 3.91 µm, and 3.91 to 0.04 µm, respectively, using the Wenworth size scales and a 8.04 value as the clay-silt boundary. Sand, silt, and clay percentages are provided only for the raw sample. Geometric distributional statistics are computed from the logarithmic center of each size grouping as sediment distributions are typically more log-normal than normal. Statistics provided include mean, median, and modal grain sizes and are in units of µm. The standard deviation is also in µm and is a measure of the spread of the sediment distribution, Skewness, a unitless coefficient, is a measure of the distortion from a symmetrical distribution, with a skewness of zero (where mean, median, and mode coincide) being perfectly symmetrical. Samples with an excess of material in the finer sizes (left-hand skewed) will have negative skewness coefficients, while samples with an excess of coarser material (right-hand skewed) will have skewness values greater than zero. Kurtosis is also unitless and is a measure of the peakedness of a distribution, with kurtosis values of zero representing a normal distribution (mesokurtic), values greater than zero (leptokurtic) indicating a higher sharper peak, and values less than zero (platykurtic) indicating a comparatively broad distribution. Grain Size Distributions, Inorganic Fraction In order to differentiate between the inorganic and organic sediments and their relative distributions, the organic fraction of an aliquot of field moist sediment was digested with hydrogen peroxide (H 2 O 2, 30 %), and gentle heating ( < 70 C), following recommendations 4

7 of Klute (1986) and Kunze and Dixon (1986). Conditions are maintained near neutral ph to prevent the digestion of carbonate sediments or shell fragments. The digested samples are subsequently processed with the laser diffraction instruments, with any inorganic material greater than 2,000 µm similarly included in final sample distributions and statistics. Statistics presented for the inorganic fraction (inorganic mean, median, mode, standard deviation, skewness, and kurtosis) refer to the distribution of the inorganic materials only. Individual volume percents of the various inorganic size classes, have been uniformly adjusted downward such that the sum of all inorganic volume percents is equal to the inorganic volume percent contained in the original raw sample (computed from total sample percent organics, and the specific gravity of dry wood, 0.42 g cm -1 [Faherty and Williamson, 19891]). This permits the total sample distribution and the inorganic distribution to be superimposed on one another, in essence obtaining the organic sediment size distribution in the original sample by difference. In some instances, however, typically in the finer particle sizes, the adjusted inorganic fractions will have a slightly larger volume percent than was percent in the original raw sample. The effect is not very sensitive to the inherent assumptions of the calculations (specific gravity of organic and inorganic fractions) or other contained measurements (percent solids, percent moisture, specific gravity of organic or inorganic fractions) and is attributed to the digestion process removing organic binding materials, thereby releasing smaller inorganic fragments. The effect is unavoidable regardless of digestion method employed, but is small, typically less than 2 % volume.

8 Table 3. Listing of half-phi intervals and equivalent µm sizes.

9 Literature Cited American Public Health Association, American Water Works Association, and Water Pollution Control Federation Standard Methods for the Examination of Water and Wastewater. 18 th Edition. Washington, DC. Environmental Protection Agency Test Methods for Evaluating Solid Waste. Office of Solid Waste and Emergency Response. SW-846. Washington, DC. Faherty, K.F. and T.G. Williamson Wood engineering and construction handbook. McGraw-Hill Publishing Company. New York, NY. Folk, R. L Petrology of Sedimentary Rocks. Hemphill Publishing Company. Austin, TX. Klute, A. (ed) Methods of Soil Analysis: Part 1: Physical and Mineralogical Methods. Second Edition. American Society of Agronomy, Inc. and Soil Science Society of America, Inc. Madison, WI. Kunze, G.W. and J.B. Dixon Pretreatment for Mineralogical Analysis. Chapter 5. In: Methods of Soil Analysis: Part 1: Physical and Mineralogical Methods. Second Edition. A. Klute (ed), American Society of Agronomy, Inc. and Soil Science Society of America, Inc. Madison, WI. 7

Standard Test Procedures Manual

STP 205-10 Standard Test Procedures Manual Section: 1. SCOPE 1.1. Description of Test This method describes the quantitative determination of the distribution of particle sizes in soils. The distribution