VARIABILITY DUE TO SPATIAL LOCATION

ERDC TN-DOER-D2

September 2004

reviewing examples from recently published project literature. The data are not intended to be

statistically robust, but are presented to illustrate general trends in property variability.

VARIABILITY DUE TO SPATIAL LOCATION: Physical property changes within a pre-

dredged, undisturbed, and heterogeneous sediment are often uncertain and difficult to predict,

and when the sediment becomes disturbed, remixed, or remolded, the uncertainty becomes

greater. Often the spatial variability influence is detected during sediment sampling conducted

before, during, or after the dredging operation. Horizontal and vertical heterogeneities affect

spatial variability. The following two examples are from the same project in Boston Harbor and

serve to illustrate the changes in physical properties and engineering behavior encountered sim-

ply because of the samples' horizontal locations.

River Sediment Samples. Surficial grab samples were taken from the bottom of the Mystic

River in Boston, Massachusetts, for the purpose of characterizing the geotechnical properties of

material to be subsequently dredged and placed in subaqueous disposal cells located further

downstream in the Mystic River and Boston Harbor. The samples were taken with small

(150 cu in. or 0.0033 cu m) hand-operated bottom dredge samplers, which scooped surficial

material from the channel bottom at a depth of approximately 40 ft (13.2 m). Several samples

from within a 50-ft (16.5 m) radius of each of three target locations (MRA, MRB, and MRC)

were withdrawn to the surface and placed in a sealed bucket. Each target location was approxi-

mately 1000 ft (330 m) apart. Myre et al. (2000) and Science Applications International Corpo-

ration (SAIC) (2000) describe the geotechnical aspects of the Boston Harbor project including

the sampling methods and locations.

The upstream sample (MRA), the middle sample (MRB), and the downstream sample (MRC),

were analyzed for specific gravity, grain size distribution, Atterberg limits (plastic and liquid

limits), organic content, water content, vane shear strength, and consistency. Visual identification

indicated a fine-grained, dark gray, highly plastic material that appeared to be the same in all

three samples. Laboratory testing indicated that the three samples had distinctly different engi-

neering properties, as indicated in Table 1.

Table 1

Physical Properties of Visually Identical River Bottom Samples

Sample

% sand

% silt

% clay

Class

MRA

2.76

110

MRB

2.74

MRC

2.74

All three samples are classified as high plasticity clay with similar specific gravity (Gs) and

similar plastic limit (PL). The grain size distribution varies significantly for MRA, which also

has the highest liquid limit (LL). Figure 1 shows the cumulative grain size distribution curves for

all three samples, Figure 2 shows the frequency grain size distribution curves, and Table 2 shows

the median and mean grain sizes.