Every year, millions of fish make a strange and harrowing detour through the Skinner Fish Facility, part of the State Water Project's facilities in the Delta.

In my last post, I wrote about my visit to the Banks Pumping Plant, whose giant pumps slurp water from the Delta to help quench California's thirst. As the volumes of water are sucked up, both resident and migrating fish come along for the ride. The Skinner Facility, in operation since 1968, was built to protect fish from being killed at the pumps–an effort that sadly is not as successful as one would hope (more on that below).

I was amazed to learn there is a whole art and science to fish screens, which range from physical barriers–called positive barriers–like perforated plates or wire mesh, to behavioral barriers like sound, light, or other stimuli aimed at keeping fish away. Well-designed screens minimize both entrainment (fish being pulled into the pump or diversion) and impingement (fish being trapped or injured against the screen itself due to water velocity).

Both physical and behavioral barriers are used at the Skinner Facility. Fish being pulled toward the pumps first encounter a trash rack that diverts many bigger fish, along with floating debris. Next, fish encounter a large, v-shaped array of metal louvers. The louvers create turbulence that functions as a behavioral signal, encouraging the fish to swim away into bypass pipes that function, as our tour guide put it, like "a big vacuum system."

From the bypass pipes fish travel to another set of louvers and pipes, concentrating them into a smaller volume of water, and then into holding tanks in a nearby warehouse. Giant, suspended cone-shaped buckets are used to periodically sample the fish, which are identified, counted, and measured. Some 90 species turn up in the facility, including Chinook salmon, steelhead, white sturgeon, and delta smelt. (I asked our guide if delta smelt really do smell like cucumbers. He confirmed it. In fact, when a school of smelt comes through–an event that has become rare–the warehouse smells "like a salad.") When enough fish have been collected, they are loaded into trucks and driven back to the Delta.

Here's the rub. Many fish caught in the pull of the pumps are lost to predation before even reaching the screening facility. Then, the facility does not effectively screen fish smaller than about 1.5 inches, meaning that littler, less powerful species and juveniles are still vulnerable to the pumps. For the fish that make it to the holding tanks, the process is such a trauma–with big and little fish squashed together in the tanks, buckets, and trucks–it's no surprise there are casualties; in fact, the delicate delta smelt often do not survive. And even for fish that make it through the entire process and out the other end, there's a final, fatal hurdle: the trunks routinely dump salvaged fish at the same locations, where more predators have learned to cluster for a free lunch.

Scientists agree that the loss of fish at the huge state pumps–and other pumps and intake pipes throughout the Delta–is a major contributor to plummeting populations. How much water we use makes a difference: The higher the export rates, the more fish are entrained. There also is broad consensus that more state-of-the-art fish screening facilities are needed. That could come with a hefty price tag. But with our fish disappearing, can we afford not to invest in their survival?

Harvey O. Banks Pumping Plant

In my last post I wrote about visiting a treatment plant to see where our water goes after we've washed the dishes. Now, on this tour of the Banks plant, I'm getting a glimpse "upstream" of the kitchen tap and learning more about where our water comes from.

The scale of the SWP is mind-boggling: More than two in three Californians rely on it for at least part of their drinking water. It is the largest publicly built and operated water project in the country, encompassing 17 pumping plants, more than 30 storage facilities, and over 660 miles of canals and pipelines. At the south end of the San Joaquin Valley at the Tehachapi Mountains, the huge Edmonston Pumping Plant raises the water 1,926 feet-the highest single lift in the world. (If you're driving to Southern California, check it out on the right side of I-5 just before the Grapevine). Moving all that water around and hoisting it over mountains doesn't come easy (water is heavy, after all): The SWP is the largest single user of electricity in the state.

The Banks plant is named for Harvey O. Banks, Director of Water Resources when voters approved funding for the SWP in 1960. The project was ostensibly conceived as a solution to the problem that most of California's water is north of the Delta, while most of its people are to the south and west. Big agricultural interests in the southern San Joaquin Valley also benefited-hugely-from "surplus" water shipped south. (And lest we Northern Californians start feeling smug, keep in mind we receive a greater percentage of our total water supply from the Delta than does Southern California.)

The Banks plant draws water from the Delta through intake gates into Clifton Court Forebay. From there, the water is pulled up a channel to the Skinner Fish Facility, where delta smelt, Chinook salmon, and some 90 other species of fish are, theoretically, screened out so they won't get sucked into the pumps (More on fish screening in my next post). But getting squashed in the pumps is not a fish's only worry: The pumping actually alters the habitat by impacting salinity and flow, disrupting natural rhythms that serve as vital cues for migration and spawning. The old joke that in California water flows uphill toward power and money is not far off the mark: The pull of the pumps is so powerful it causes rivers to flow backwards-literally uphill.

Crashing fish populations, poor water quality, the vulnerability of Delta levees and our water supply to earthquakes or other disasters-all have added to the growing realization that we can't keep quenching California's thirst through big straws stuck in the Delta. Obviously the SWP is not going to stop pumping anytime soon. But we do need to find ways to reduce our reliance on the Delta-through conservation, water recycling, and increased regional self-sufficiency-and to restore the functioning of an ecosystem so devastated by our radical retooling of our waterways.

Recent news headlines have been full of Chinook salmon, but sadly the same cannot be said of Central Valley waterways. This fall, only about 90,000 Central Valley Chinook salmon returned to their home rivers and streams to spawn, down from more than 800,000 just a few years ago.

Like most salmon, Central Valley Chinook are anadromous, spending the bulk of their lives in the ocean but hatching and returning to reproduce in freshwater. The journey from the Valley through the Delta and San Francisco Bay to the Pacific, and back again, has always been a long and arduous one. In the past half century it has become even more difficult as the fish have increasingly faced an obstacle course of dams, pumps, and dewatered rivers and creekbeds.

Central Valley Chinook salmon populations include four runs-winter, spring, fall, and late fall-with each spawning not only at different times of the year but in different parts of the watershed. The dawning of the age of dams hit the winter and spring runs the hardest, cutting the fish off from their historic spawning grounds in the upper reaches of the watershed. Both runs are now listed under federal and state Endangered Species Acts.

The fall run, which spawns lower in the watershed, was less impacted by dam construction. In recent decades it has numbered more than 10x all the other runs combined and has been the mainstay of the California coastal salmon fishery. Now, even it appears to be in serious trouble: The count of 90,000 salmon this year was the second lowest on record and well below the minimum conservation target of 122,000 set by the Pacific Fishery Management Council. Also alarming is that the number of 2-year-old "jacks" returning was just 2,000, down from a typical count of 40,000. Since most spawners are 3-year-olds, these early returners are considered a good predictor of the size of next year's run.

Ocean conditions are one factor in the salmon decline, with rising water temperatures and more unpredictable upwellings-possibly resulting from global warming. But scientists are also pointing to overexploitation of our rivers and Delta-the "highway" for migrating salmon. The abrupt decline in the salmon population comes concurrently with the collapse of other fish species dependent on the Delta ecosystem, including delta smelt and longfin smelt. The salmon returning to spawn this year would have been juveniles headed to the ocean in 2005, the year Delta water exports hit a record high.

Between unfriendly ocean conditions and the degraded condition of the watershed, the salmon are facing a double whammy. But, as Bay Institute Senior Scientist Tina Swanson points out, "Apart from rolling back global warming, we can't really control ocean conditions. What we can do is drastically improve conditions within the watershed so that more adults can spawn successfully and more juveniles survive the journey to the ocean."

Ann Dickinson is Communications Manager for The Bay Institute (www.bay.org), a nonprofit research, education, and advocacy organization dedicated to protecting and restoring San Francisco Bay and its watershed, "from the Sierra to the sea."