After the Sheep are Gone

Restoration of Santa Cruz Island turned out to be as much about defining a new ecosystem as about restoring an old one.

By Jason Van Driesche and Roy Van Driesche

It wasn’t until we banked in close that I saw any sign of life. From a thousand feet up, there were a few trees clinging to the ridgetops and a close-shorn stubble of grass on the slopes. But what jumped out were the sheep trails. From the rocky beaches to the 1500-foot peaks, the island was crisscrossed with the traces of Ovis aries-the domestic sheep-a voracious and prolific herbivore whose presence in this predator-free ecosystem has nearly destroyed the island itself. Though Santa Cruz Island (a 100-square mile island off the southern California coast) is a spare and dry place, it is not by nature a wasteland. We were flying over an ecosystem that had been under full frontal assault for well over a century.

But then our plane crossed the line that separated the east end of the island-where sheep still roamed free-to the body of the island, where sheep had been eradicated about ten years before. We flew in low over sage-covered bluffs and deep, tree-lined ravines. The air was tinged with an aromatic haze, and the canyons showed occasional flashes of water. There was life in abundance on this side of the fence. It was hard to believe that only ten years ago, this rich sweep of coastal sage scrub (an area known as Loma Pasture) was chewed just as bare as the far east end of the island was now. This is what Santa Cruz Island was meant to be. Or was it?

We touched down on a dirt airstrip carved out of dense twelve-foot high stands of fennel. This part of the island ought to have been coastal sage scrub, just like the bluffs that we passed over during our descent. Instead it was completely dominated by a garden vegetable gone wild. No one had anticipated that fennel-an incidental weed when the island was still grazed-would spread so explosively once sheep (and cattle as well, on this part of the island) were removed. The area around the airstrip-and, for that matter, a large portion of the island’s central valley-has followed an entirely different recovery trajectory than have the more remote parts of the island. Variation in composition of the island’s post-eradication plant communities makes it clear that recovery of a heavily overgrazed ecosystem is not as simple as removing the non-native herbivores. Sheep and cattle eradication was a vital first step toward recovery-but it was only the beginning. There are no guarantees in an invaded ecosystem, and recovery is more often a drawn-out series of mistakes and course corrections than it is a straightforward shift from invaded to native. What follows is a brief account of the ecological twists and turns since eradication-and of how managers from The Nature Conservancy and the National Park Service have responded as the system has changed.

A Brief History of Santa Cruz Island

Almost all the ecological changes on Santa Cruz Island-past, present, and future-are shaped to a degree by the fact that it is an island. The ecosystem has no large native herbivores, and before sheep were introduced around 1850, it was not subject to significant grazing or browsing pressure. Even more important, the island has no native predators large enough to exert a controlling influence on a sheep population once it is introduced. As on many islands, this combination of lack of large predators and herbivores conspired to create an ecosystem that was exceptionally vulnerable to catastrophic overgrazing.

Once free-ranging sheep were introduced in the mid-1800s, their population grew rapidly in the island’s near-ideal habitat. Sheep numbers surpassed 50,000 by 1870, despite the fact that tens of thousands of the animals were shipped to the mainland every year during that period. A few years later, though, the number of sheep had clearly overshot the carrying capacity of the island.

Following a series of mid-1870s droughts, the population crashed. But the reduction in sheep numbers was too little and too late. Thirty years of unrestrained grazing had dramatically reduced the productive capacity of the vegetation, and grazing pressure from the sheep that remained was more than enough to prevent recovery. A downward spiral of degradation had begun.

The process of mass wasting slowed somewhat across much of the island in the early part of the twentieth century as gullies wore down toward bedrock, forming relatively stable channels that carried water rapidly to the ocean. Soil and vegetation loss was attenuated further still by a change in ownership of the western 90 percent of the island in 1937, when the focus of ranch operations shifted to cattle and the island’s new owners began active sheep control. Between 1939 and 1980, about 260,000 sheep were either transported to the mainland or shot. Although this new arrangement significantly reduced the impacts of sheep on some parts of the island, overgrazing continued unabated in many other areas.

As a consequence, a number of the island’s natural communities continued to deteriorate. In the mid-1980s, a study by D. Van Vuren from The Nature Conservancy and B. E. Coblentz from the Department of Fisheries and Wildlife at Oregon State University found that “long-term overgrazing by sheep [had] resulted in moderate to severe ecological impacts to about one-half of Santa Cruz Island.” (1). They concluded that “feral sheep have severe negative impacts on the native biota of islands and that endemic plants and birds are particularly vulnerable.” Later, in a demographic study, they demonstrated that even in the face of what appeared to be scarce food resources, the sheep population showed no sign of decline (2). While birth rates, survivorship, and overall condition of large mammals normally decline in response to habitat degradation, “feral sheep [on Santa Cruz Island] showed none of these responses… Our data, showing a healthy sheep population thriving in deteriorated habitat, indicated that the continued presence of sheep would lead to further ecological degradation.”

While these two studies focused on the costs of continued overgrazing, it took a third study to show the benefits of sheep removal. R. W. Brumbaugh from the Department of Geography at University of California, Los Angeles fenced off a small plot in an area that was heavily grazed by sheep and found that within a few years, ground cover in the fenced area was markedly thicker and many native plants that were absent in grazed areas had returned (3). He concluded that “sheep exclosure evidence… indicates the possibility of rapid vegetal and soil recovery if soils are not completely eroded away to bedrock.” Spontaneous recovery was still possible-but under what particular conditions, no one could be sure.

This was the situation that The Nature Conservancy found itself in when it took over management of 90 percent of the island in the early 1980s. Sheep eradication was their number one priority. All the evidence indicated that recovery would be possible only if sheep were removed, and fast. But sheep had driven the system’s composition and dynamics for so long that no one really had a firm idea what recovery would look like. So while the Conservancy’s goal was clear-”to preserve and protect in perpetuity and to enhance the natural ecosystems, the unique natural flora and fauna, the hydrologic features and the natural aesthetic values of the Island”-the way there was anything but straightforward. As a consequence, the process of recovery turned out to be as much about defining a new ecosystem as about restoring an old one.

This is why the Conservancy’s eradication campaign began not with bullets but rather with the series of research projects that provided data on sheep impacts and demographics cited above. Once this baseline research was done, the Conservancy spent the next seven and a half years eradicating all the sheep on the western 90 percent of the island. Teams worked systematically, clearing sheep from one pasture before moving on to the next. At the same time, the Conservancy decided to round up and ship out the remaining cattle from the ranching operation that preceded the preserve. By June 1989, there were almost no sheep or cattle remaining on Conservancy lands. The single most powerful force shaping the island for over a century was gone. This was a new ecosystem.

Or, rather, two new ecosystems: one characterized largely by spectacular regrowth of a wide variety of native species, and the other by equally spectacular stands of fennel interspersed with invasive pasture grasses. While the two were in reality part of the same ecosystem, the differences between them were so striking as to give pause. Why had some parts of the island recovered to coastal sage scrub and pine forest, while others had become little more than weed patches? The answer, it turns out, is to be found in a careful examination of the island’s land use history.

Paying Attention to History

Santa Cruz Island’s most prominent topographic feature is its large, flat central valley. This was the base of operations for both the sheep ranch that began in the mid-1800s and the cattle operation that took its place in the 1930s. Intensive human presence in the central valley had two important impacts on its ecological trajectory. First, it is likely that more non-native plants were introduced in this and other settled areas than in the island’s remote canyons and mountainous regions. Second, soil disturbance was probably more intensive and continuous in the central valley than in most of the mountains, especially in those parts of the valley that were at times farmed.

What these differences added up to was a fundamentally different set of baseline vegetative conditions in the central valley (and the few other areas under intensive long-term human use) than on the rest of the island. Though most of Santa Cruz’s vegetation was uniformly chewed down to nothing at the time of eradication, what the roots and stubs of vegetation consisted of varied considerably from one location to another.

Although the Conservancy had considered the possibility that nonnative plants would increase in abundance once grazers were removed, this possibility unfortunately was not given enough weight. The preserve’s managers were operating on the unspoken assumption that-as preserve ecologist Rob Klinger and two others noted in a study of vegetation changes immediately following sheep removal-”removing nonnative grazers [would] lead to recovery of native species.” Implicit in the eradication campaign was the assumption that history did not matter; that once sheep and cattle were removed, the island would once again be a blank slate and nature would write upon it as the “original” native script intended. Experience showed just the opposite: that history is essential, and there is no intention in the composition of natural communities. Removal of grazers simply led to recovery of whatever vegetation was present on a given site (4).

Looking back on the changes following eradication, Klinger saw the complexity of the situation in a new light. He compared the island ecosystem just before the removal of grazers to a set of interlocking coiled springs. “When you release one,” he explained, “they all take off in different directions.” The experience of sheep eradication and cattle removal has made it clear that alien species control must be a means of native ecosystem restoration, not an end in itself.

While control efforts are essential, Klinger continued, even more important is developing specific restoration goals and a broad understanding of ecosystem dynamics. What ecosystem state or complex of states is most likely to sustain the full scope of native diversity over the long run? How are these desirable states related to less desirable states, and what might trigger a transition between them? Unless alien species control is conceived and carried out with the goal of restoring native ecosystem structure and function, the effects of control will persist only so long as we keep our hands on the wheel. In the long run, the system will have to function without human intervention. The more we understand its dynamics, the better equipped we will be to guide it in the direction that most benefits native species-and then to let it go.

Adaptive Management for the Next Century

With this perspective in mind, the Conservancy and the National Park Service began working together in the late 1990s to create an adaptive management strategy to guide restoration on the island. The National Park Service completed purchase of the eastern 10 percent of the island in 1998, and the Conservancy transferred ownership of an additional 14 percent of the island to the Park Service in 2000 for inclusion in Channel Islands National Park.

From 1998 to 2000, the Park Service removed the remaining sheep from the island’s 6000-acre East End, trapping and shipping a total of about 9000 sheep to the mainland. In the year since the last sheep were rounded up, a few native plants have begun to return, but non-native grasses still largely dominate most of the East End. No one is entirely sure why. This is a worrisome state of affairs given that, as Channel Islands National Park’s Chief of Natural Resources Management Kate Faulkner observed, “What we’ve seen on other islands with heavy non-native grasses is a very slow transition back to native species.”

The challenge now, said Faulkner, is to figure out where intensive restoration effort-plantings, soil mats, and so on-is most likely to promote recovery that otherwise would not have happened or that would have been very slow to come about on its own. Just as the island’s managers would rather not waste time and money on areas where speedy recovery is next to impossible, they want to avoid investing limited resources in areas that are poised for the kind of spontaneous rapid recovery to native species that has made the Loma Pasture area such a success. However, making these kinds of distinctions with any kind of consistency has proven quite difficult. The Loma Pasture, for instance, “was not an area that you would have said would recover so fast.” As Faulkner noted dryly, “We’re not very good at predicting which areas are going to respond most favorably.”

Despite the difficulties inherent in predicting the course of recovery at a given site, one thing is clear: eliminating or controlling major invaders is overwhelmingly a net positive for the ecosystem. Even though removing grazers has produced more positive results in some areas than in others, no one who knows the island would ever say that removing sheep and cattle was a bad idea. Moreover, all of the island’s managers are in agreement about the next step-eradication of the feral pigs that also have roamed the island for over a century.

Until the last decade or so, pigs were overshadowed by sheep and cattle. Yet, they are now one of the primary vectors of transport and establishment of non-native plants on the island. Young pigs are also a critical food source for a population of golden eagles, another North American species not native to the island. The eagles in turn are largely responsible for a recent catastrophic decline in the population of the endemic island fox, from an estimated 2000 individuals in 1994 to perhaps as few as 135 in 2000 (5). Though a number of other management actions are needed in conjunction with pig eradication, Faulkner said, “just removing pigs will put us 90 percent of the way there.”

To this end, the Park Service worked in close cooperation with the Conservancy to prepare an Environmental Impact Statement for the entire island that outlines the next steps in the island’s ecological recovery. The preferred alternative calls for repairing fences that divide the island into ten large pastures and then eradicating pigs from one pasture after another until the island is free of pigs.

The plan also calls for simultaneous aggressive control of fennel. It describes fennel control as “a connected action to the proposed pig eradication actions… Because of the density and extent of the fennel on the isthmus of Santa Cruz Island, substantial reduction of the fennel is required to successfully eradicate pigs from this area.” While managers are focused on pig eradication as the next most urgent step in restoring the island’s native ecosystem, they now recognize the dynamic relationships between species that drive the system. Though no one is claiming to know in any detail what will happen next, everyone is a little more ready for whatever it might be.

The air was clear and cloudless the afternoon I left the island. We roared off the airstrip in a swirl of dust, cleared the bluffs, and climbed out over the ocean. But even from a couple of thousand feet up, the island was too massive to take in all at once. There is a lot of room on this island for surprises to hide, and rewriting the ecological score on this grand a scale is the height of ambition. At this point, though, there really is no other choice.

There are no doubt many more surprises to come. But acknowledging and working with every curve ball that the system throws is an essential prerequisite to successful recovery. This requires doing everything possible to anticipate the surprises and-if they turn out to be unpleasant ones-counter their impacts before they compound the effects of past damage. Just as important, of course, is to stay alert for the good surprises and encourage them as soon as they become apparent. The success of recovery on Santa Cruz Island will be determined largely by the ability of the people charged with its care to adapt to whatever opportunities and challenges the island presents.

This, then, is the shape of conservation from here on out-juggling past mistakes with present knowledge in the hope of coaxing something better from the future. Like it or not, we have imposed upon ourselves for many generations the role of stewards of the earth. Santa Cruz Island is as good a place as any to begin to learn how to carry that responsibility well.

Literature Cited:

1. Van Vuren, D. and B.E. Coblentz. 1987. Some ecological effects of feral sheep on Santa Cruz Island, California USA. Biological Conservation 41:253-268.

2. Van Vuren, D. and B.E. Coblentz. 1989. Population characteristics of feral sheep on Santa Cruz Island. Journal of Wildlife Management 53(2):306-313.

3. Brumbaugh, R.W. 1980. Recent geomorphic and vegetal dynamics on Santa Cruz Island, California. Pp.139-158 in D. Power (ed.) A multidisciplinary symposium on the California Islands. Santa Barbara Museum of Natural History, Santa Barbara CA.

4. Klinger, R.C., P.T. Schuyler, and J.D. Sterner. 1994. Vegetation response to the removal of feral sheep from Santa Cruz Island. Pp. 341-350 in Halvorson, W.L. and G.J. Maender (eds.) Fourth California islands symposium: Update on the status of resources. Santa Barbara Museum of Natural History, Santa Barbara, CA.

5. Santa Cruz Island primary restoration plan: Draft environmental impact assessment. 2001. Channel Islands National Park, Santa Barbara County, CA.

Further Information:

This article was adapted from the recent book, Nature Out of Place: Biological Invasions in the Global Age by Jason Van Driesche and Roy Van Driesche (Island Press, Fall 2000).

Restoring Santa Cruz Island at:

About the Authors:

Jason Van Driesche is a freelance writer and a graduate student in the Institute for Environmental Studies and the Department of Urban and Regional Planning at the University of Wisconsin-Madison.

Roy Van Driesche teaches and researches biological control at the University of Massachusetts-Amherst, and is co-author of the textbook Biological Control (Chapman and Hall, 1996). He is Jason’s father.