Via Fax & U.S. Mail:
(802) 241-2596
Total of 3 pages

OMYA,Inc.
61 Main Street
Proctor, Vermont 05765
802 459 3311
fax 802 459 3428

April 26, 2000

Liz Dickson
Vermont Agency of Natural Resources
Wastewater Management Division
103 South Main Street, Sewing Building
Waterbury, VT 05671-0405

RE: Notification of Non-Compliant Discharge -- Verpol Plant
Discharge Permit Number 3-0395

Dear Ms. Dickson:

In accordance with Part II Condition A.2. of OMYA's current NPDES discharge permit, this written communication follows up my telephone conversations with Carol Fowler of your office and Environmental Enforcement Officer Gary Urich, both on April 24, regarding the unintentional discharge of process water from OMYA's Verpol plant. The following information reiterates and adds to the information previously supplied verbally and is intended to satisfy the written non-compliance notification requirements of the permit:

a. The incident involved the discharge of process water containing suspended calcium carbonate solids leaving the Verpol plant site at the same point as the permitted combined stormwater (#1-0843) and process water (#3-0395) discharge systems into the unnamed tributary of Smith Pond, continuing to Otter Creek. The events leading to the incident began on Thursday 4/20 at 4:00 p.m. when the primary grinding circuit (cascade and ball mills) at the Verpol plant shut down due to a problem.

When the primary grinding circuit is operating, process water is circulated from the cascade mill, to flotation, to the ball mills, to the thickener, and then back to the cascade mill. The undispersed ball-milled material is settled in the thickener with the aid of a flocculent, and then drawn off the bottom as feed for the fine grinding circuit. The thickener overflow is then returned to the cascade mill where new make-up water is added as needed (175 gal/min typical) to replace that leaving the system with the thickened solids.

During the infrequent periods when the primary grinding circuit is out of operation, the fine grinding circuit can still be operated for a limited time from the store of settled solids in the thickener. Under that operating scenario, the level in the thickener will drop as the thickened solids are drawn off. Therefore new make-up water must be introduced directly to the thickener to maintain its level so that it is ready to overflow to the cascade mill immediately upon re-start of the primary grinding circuit. At the time of the incident, the source of that make-up water was the OMYA East quarry. This quarry has historically been used for flotation waste settling and the water is normally relatively low in TSS, although because of recent events TSS is currently higher due to the presence of water containing dispersed solids from the 'dogleg' quarry.

The primary grinding circuit was restarted at about 1:00 a.m. on 4/21. Due to operator error, the make-up water going directly to the thickener was not shut off after the primary grinding circuit was restarted with its make-up water addition. As a result, the water balance was upset and the thickener overflow volume exceeded that being returned to the system.

The Verpol plant has a discharge permit which allows for the off-site discharge of thickener overflow as well as "reject water" from the cogeneration turbine water injection treatment system. Due to the normally negative plant water balance, there has never been a need to discharge thickener overflow (although the permit has been maintained through the years to cover unforseen situations where a discharge might become necessary). Consequently, a conduit to conduct the thickener overflow to the storm drain discharge route identified in the permit was never constructed. During the incident, the water followed a gravity course through a series of stormwater ditches on the east side of the Verpol site along an unpaved plant road, before finally leaving the site at the same point as the permitted combined stormwater (#1-0843) and process water (#3-0395) discharge systems into the unnamed tributary of Smith Pond, continuing to Otter Creek.

b. The suspended solids in the discharge water consist of finely ground calcium carbonate particles. A sample of the discharge was not taken. However, from a visual observation of the discharge at the point it leaves the site, it is believed that it did not comply with TSS and turbidity permit limits. This was due to the use of OMYA East water as make-up in the thickener.

Due to the preceding several days of rainy weather, volumetric flow in the unnamed tributary during the event period was high. Visual observation along several hundred feet of its length found no deposition or sedimentation, as would be expected due to the fineness of the calcium carbonate particles and the stream velocity. Visual observation of Smith Pond on the afternoon of 4/21 found a noticeable discoloration of the pond water attributable to the discharge. Rainy weather continued over the weekend, and visual observation of Smith Pond on the morning of 4/23 found it to have returned to its normal appearance. Considering that finely ground calcium carbonate is applied directly to lakes and ponds in various parts of the U.S. to counter the effects of acid precipitation, it is believed there will be no detrimental impacts on the unnamed tributary, Smith Pond, or Otter Creek.

c. The period of thickener overflow began sometime after 1:00 a.m. on 4/21 and continued through approximately 9:30 a.m. on that same day.

d. As soon as the cause of the problem was realized at approximately 9:30 a.m. on 4/21, the makeup water flow to the thickener was shut off and the thickener stopped overflowing. A temporary pump was placed in one of two small abandoned quarries on the east side of the Verpol site along the route the overflowed thickener water took. These quarries intercept the stormwater ditch along the east side unpaved plant road and in essence act as detention/settling basins. Because natural stormwater runoff also flows into the quarry, it was decided to attempt to pump out the white water that had been retained in the quarry so that in flowing clean runoff would not displace the white water and extend the period of the incident. The water was pumped from the quarry into a sump in the nearby finished product building, and then to the Verpol dogleg quarry.

Although it did not actually reduce or eliminate the discharge, we would like to note that on 4/22 the plant manager went door-to-door to each residence in the vicinity of Smith Pond for the purpose of providing information relative to the incident.

e. The incident brought to light a weak link in the procedural chain of actions relative to re-starting the primary grinding circuit. Obviously there was a previously-unrecognized potential for manual operator error in not shutting down a critical pump in a timely manner. In response, on 4/26 a new instruction set was programmed into the plant operating control logic to automatically reduce the amount of make-up water added directly to the thickener as the level in the thickener rises, eventually shutting off altogether when the level reaches 100% of capacity for ten minutes (the ten minutes is intended to account for typical instrumentation 'noise'). There is no manual override capability. This removes this pump shut-off decision/action from the manual control of the operators and should prevent the possibility of a future incident.

If you have questions or require additional information, feel free to call me at (802) 770-7261.

Sincerely,
Neal Jordan
Manager of Environmental & Regulatory Affairs

cc. L. Gaudreau
S. Thompson