LNAPL thickness

FIGURE 6.3 Sandbox model showing LNAPL overlying capillary fringe, and apparent vs. actual LNAPL thickness. Saturated conditions (water table) are represented by the straight horizontal line. 

FIGURE 6.6 Cause for discrepancies in apparent LNAPL thickness as measures in a monitoring well in comparison with actual thickness in the formation. 

Additionally, vertical fluctuations in the water table due to recovery operations or seasonal variation in precipitation have a direct effect upon the apparent or measured LNAPL thickness (Figure 6.6d). As the water table elevation declines gradually due to seasonal variations, for example, an exaggerated apparent thickness occurs, reflecting the additional hydrocarbon that accumulated in the monitoring well. The same is true for an area undergoing recovery operations where the... 

Summary of Equations Relating Apparent LNAPL Thickness (H) to Actual Thickness (h)... 

FIGURE 6.7 Cross section of an LNAPL pool showing well-to-formation-LNAPL thickness ratios. 

Apparent/actual LNAPL thickness ratios can be very high at the perimeter of LNAPL pools, notably, under low permeability conditions. Once a well is installed, it can take several months before the LNAPL migrates from the formation into the well reflecting the presence of low-permeability soils in the zone of LNAPL occurrence. As clearly shown in Figure 6.8, a well screened at the perimeter of a known LNAPL pool initially had no detectable LNAPL until 4 months after installation, whereas upon detection the apparent thickness slowly increased with time up to 15.71 ft. 

Several equations and empirical relationships have been formulated in an attempt to relate the apparent LNAPL thickness as measured in a monitoring well to that which actually exists in the adjacent formation. These indirect empirical approaches are summarized in Table 6.3 and discussed below. 

Equation 6.2 has been disputed by several investigators. Shepherd (1983) claimed that apparent LNAPL thicknesses in wells may become more exaggerated than accounted for in Equation 6.2 in certain cases. Hall, Blake, and Champlin (1984) described convincing laboratory experiments that contradicted Equation 6.2. Their results can be formalized as follows ... 

Direct Field Methods for Measuring Actual LNAPL Thickness... 

For cases where an aquifer is believed to be contaminated by a product spill but no apparent LNAPL thickness measurements are available, Dietz (1971) presented Table 6.7 relating the maximum expected product thickness to average aquifer grain diameter for various sands. He assumed that the maximum oil pancake thickness equals the capillary zone thickness. For each sand size range, he gave a corresponding range of maximum h values. 

Continuous dry coring (without water or mud) of the interval immediately above and below the mobile LNAPL layer seems an obvious way of determining the mobile LNAPL thickness. Standard split-spoon samples are commonly used for this purpose. The use of clear acrylic shelby tubes have also been used with favorable results. However, several problems are associated with coring which results in overestimating the actual thickness of the product. These factors include ... 

The resulting LNAPL thickness is conservative in that it incorporates both the actual thickness of LNAPL in the adjacent formation and the height of the capillary fringe. For most practical purposes, this level of accuracy is sufficient, although the more complex and extensive the site conditions, the more sophisticated approaches may be warranted, as discussed later in this chapter. 

Determination of actual LNAPL thickness in the formation is important in providing reasonable estimates of the volume of free product in the subsurface. This, in turn, allows for reasonable estimates of time frame for recovery, as well as provides a mechanism for monitoring the efficiency and effectiveness of a recovery operation. The efficiency and effectiveness of many large-scale recovery operations are monitored by the reduction in volume with time. Thus, the percent reduction with time can easily be viewed as insignificant if exaggerated volumes are used. For example, if one estimates a volume on the order of 100,000 barrels, of which 10,000 barrels have been recovered to date, then a 10% reduction over a certain time interval has been achieved. However, if only 50,000 barrels exist, of which 10,000 barrels have been recovered, then a 20% reduction has actually been achieved. 

For relatively low values of apparent LNAPL thickness (100 cm or less) observed in the well, the ratio of hydrocarbon volume to apparent product thickness is very low, in the range of 10 2 to 10 1 or even less. 

For sufficiendy large values of LNAPL thickness, V0IH0 approaches the limiting maximum value of 0(1 - Sr), which is usually in the range of 0.2 to 0.4. 

Preliminary estimates of LNAPL made during the investigative phases of a project are usually based on the results of short-term pumping tests, approximations of actual LNAPL thickness based on gauging data generated from monitoring wells, and other approximate data. The numerical estimates based on this short-term information are often adequate to define the physical parameters sufficiently for prelim-... 

Reductions in apparent LNAPL thickness in the subsurface have been effected over most of the site. In general, apparent thickness after 1 year were reduced 30 to 40% to those apparent thicknesses measured prior to recovery. The reduction in apparent thickness in a representative monitoring well with time is shown in Figure 12.3. However, in one area of the site, hydrocarbon thicknesses actually increased... 

FIGURE 12.3 Apparent LNAPL thickness decreases with time upon initiation of recovery operations. 

FIGURE 12.4 Apparent LNAPL thickness fluctuations with time reflecting installation of additional recovery wells. 

Maximum LNAPL Thickness (ft) Est. LNAPL Subpool Area (acres)... 

---