Citizens For Alternatives to Radioactive Dumping

Dr. Richard Phillips
PhD, Geomorphology
University of Oregon

The Waste Isolation Pilot Plant (WIPP) near Carlsbad, New Mexico, is intended for the permanent disposal of radioactive waste from nuclear weapons production. The WIPP site was selected in 1974. It is now 1997, and the Department of Energy (DOE) still does not know where the groundwater aquifers are recharged. This, at a minimum, must be understood, or DOE's site characterization has no credibility (EEG-32, 1985; Anderson, 1994; Konikow, 1995; EEG-61, 1996; SEIS, 1996, Appendix H).

DOE's failure to grasp the fundamentals of WIPP site hydrology stems not from a lack of evidence, but from an unwillingness to face the truth:

  1. the WIPP site is in karst

  2. the Dewey Lake Redbeds and the
    Rustler Formation are recharged by rainwater, and

  3. groundwater flow at WIPP is three-dimensional

The controversy over karst at the WIPP site dates to a paper by Larry Barrows entitled: “WIPP Geohydrology -- The Implications of Karst” (Barrows, 1982; reprinted in EEG-32, 1985, Appendix A). Barrows cites as evidence of karst geomorphology:

  1. ample precipitation

  2. lack of surface runoff

  3. disappearing arroyos

  4. sink holes, and

  5. underground caverns.

The WIPP site is located in one of the largest karstlands in the world. The Pecos River valley is famous for Santa Rosa Sinks, Bottomless Lakes, and Carlsbad Caverns. Within one mile of the northwest corner of the WIPP site is Nash Draw, a huge depression in the land surface, up to 18 miles long and 10 miles wide. Nash Draw was formed by the coalescence of thousands of sink holes caused by the abrupt collapse or gradual subsidence of overlying rocks into underground caverns beneath them.

Nash Draw is bounded on the east by Livingston Ridge, which is actually a rim, a 100-foot escarpment capped by Mescalero caliche. Livingston Ridge is not a geomorphic divide; it does not represent the eastern edge of karst conditions. It is the eastern extent of widespread collapse of surficial rocks into the voids caused by dissolution of evaporite rocks in the subsurface. Karst exists east of Livingston Ridge, but the karst landforms are not as widespread or as well developed as in Nash Draw.

The WIPP site has almost no surface runoff. This is not due to inadequate precipitation. Rainfall averages 14 inches per year, and 20 inches per year is not uncommon. Rather, the WIPP site is covered with windblown sand in the form of deflation basins and partially stabilized dunes. These sands are transmissive enough to allow infiltration of even the largest storms. “Instead of running off, the precipitation collects in small topographic depressions and rapidly soaks into the ground. The absence of surface runoff is characteristic of a karstland.” (Barrows, 1982)

Most of the depressions are windblown. But some of the larger ones are sink holes, exemplified by WIPP-33 and WIPP-14, located 1.1 mile and 3.4 miles east of Nash Draw, respectively. WIPP-33 is a collapse sink with a disappearing arroyo, underlain by five caverns: one in Dewey Lake siltstone, two in Forty-Niner gypsum, and two in Magenta dolomite. WIPP-14 is a solution-subsidence doline which has held water in the geologic past; now the Culebra dolomite is underlain by 70 feet of mud with gypsum and anhydrite fragments, here interpreted as cave sediments. The cavernous zones at WIPP-33 and WIPP-14 are direct evidence of karst. These zones can be correlated stratigraphically with washouts and loss of core in seventeen other WIPP boreholes and in the WIPP ventilation shaft. The question is not whether karst exists at the WIPP site, but whether karst hydrology is active today.

The Rustler Formation is the most transmissive aquifer and the principal karst horizon at the WIPP site. If karst hydrology is active today, then the Rustler must be recharged by rainwater. A likely process, according to Barrows (1982), is downward infiltration of fresh water through feeders in the overlying Dewey Lake Redbeds to karst channels in the Rustler Formation. Conversely, if karst hydrology is not active today, then the Rustler Formation must not be recharged by rainwater. This would require a continuous impermeable layer, acting as a barrier to rainwater infiltration, somewhere in the stratigraphic column above the Rustler Formation. Bachman (1985) argued that Mescalero caliche forms such a barrier, preventing infiltration and recharge of the Dewey Lake Redbeds and the Rustler Formation.

Caliche is a layer of calcium carbonate that forms in desert soils at the depth of soil water penetration. Where soil cover is thin, the caliche horizon may become plugged and indurated, forming a “hardpan” resistant to erosion and impervious to rainwater. But where soil cover is thick, infiltrating soil water may migrate along the caliche surface until it finds a fracture that allows downward drainage, or a hole where a plant root has penetrated the caliche; or it may collect in a small depression in the caliche surface and begin to dissolve a new hole in the caliche. In the southwestern part of the WIPP site (SW/4 sec 30, T 22 S, R 31 E), where Mescalero caliche is in direct contact with the Dewey Lake Redbeds, trench exposures revealed fifteen solution pipes, 1 to 14 feet in diameter, right through the caliche. Here the Dewey Lake Redbeds are recharged directly by rainwater. These trenches were located in a karst valley, a broad swale one mile long, ten feet deep, trending east-west, and narrowing from 900 feet in the east to 200 feet in the west, where thick groves of mesquite bushes are impenetrable. Other smaller topographic depressions, visible in the WIPP site air photos, shown on USGS topographic maps, lead directly to the deepest fluvial incisions in Livingston Ridge. The air photos reveal ephemeral or near-surface drainage courses expressed at the land surface as vegetation in dendritic patterns.

The Gatuna Formation, consisting of light reddish-brown, poorly consolidated sandstone, is alluvial fill material deposited in ancient sinks and topographic lows by westward-flowing streams. It was exposed in trenches on the slopes of WIPP-33, below the caliche escarpment. The Gatuna sandstone is commonly fractured, jointed, and broken into blocks. As soil water dissolves the carbonate cement, these openings become enlarged by solution, forming solution pans or tinajitas, and solution grooves or slots. The Gatuna is not a barrier to rainwater infiltration.

The Santa Rosa Formation consists of pale orange, coarse-grained sandstone, cemented by dolomite, interbedded with conglomerate lenses containing dolomite, chert, and quartz pebbles. The Santa Rosa has been eroded from the western part of the WIPP site; to the east, where it remains, it protects the underlying Dewey Lake Redbeds from erosion. At WIPP-14, the Santa Rosa was exposed in trenches beneath a leached and degraded caliche profile. The Santa Rosa exhibited carbonate-filled fractures, direct evidence of rainwater infiltration. The Santa Rosa retards, but does not prevent, rainwater recharge to the underlying Dewey Lake Redbeds.

Water has been encountered in the Dewey Lake Redbeds in eleven test wells inside or within one mile of the WIPP site. All are listed in Table 1. According to the neutron log for H-3b4, a down-hole camera recorded “water streaming from fracture.” The water level was 466.85 feet below the surface. Water was also observed in the Dewey Lake Redbeds in the air intake shaft near the center of the WIPP site (EEG-61, 1996, p. 2-6), at WIPP-33 (SAND 80-2011, p. 11), and in three private wells within 2.5 miles of the WIPP site (Ranch, Barn, and Unger). All are shown in Figure 1.

Rainwater Recharge: Figure 1

Table 1 reveals a strong correlation between encounters of water in the Dewey Lake Redbeds and absence of the overlying Santa Rosa sandstone. At least nine of the thirteen test wells where the Santa Rosa is not present produced water in the Dewey Lake Redbeds. It is not certain that the other four (H-6, P-14, WQSP-5, and Cabin Baby) did not produce water in the Dewey Lake Redbeds, because the actual neutron logs for these test wells are unavailable. However, the “abridged drill-hole histories” for P-13 (located 224 feet from the H-6 hydropad) and P-14 do not report water in the Dewey Lake Redbeds. Only two of the twenty test wells where the Santa Rosa is present produced water in the Dewey Lake Redbeds. At these test wells (H-11 and H-16) the Santa Rosa is only 54 feet and 15 feet thick, respectively. This is further evidence that the Santa Rosa retards, but does not prevent, rainwater recharge to the underlying Dewey Lake Redbeds.