Succession

Primary succession is the process of lichen, algae, fungi, and eventually plants establishing themselves on new ground without soil. This could be a lava flow, or an area uncovered by a retreating glacier. Secondary succession is the process of plants re-establishing in areas already covered by soil, for example after a fire.

Riverine succession sometimes is primary (on large deep barren sandbar deposits) or secondary (where sediment deposition is shallow or doesn't bury pre-existing vegetation.

Primary succession usually takes longer than secondary succession.

In 2005 I went on a Glen Canyon Institute backpacking trip into the record-low Powell Reservoir. That April it had dropped to the lowest elevation since during its initial filling in 1969. By the time of our September trip, the reservoir had risen to the same level it had been in 1973--that summer, features had emerged from the water that had been submerged for 32 years. We weren't able to visit the now-accessible Cathedral in the Desert, but we did visit a rock arch within an arch that had been submerged by the reservoir for most of my lifetime. All these features have been exposed again continuously since 2020.

Dense willow thickets were growing on newly-exposed mudflats, and Tamarisk was invading as well.

In June 2026 I went on a rafting trip with some Friends of the River colleagues on the Klamath River through two former reservoirs that had been dammed since 1962. The Copco and Iron Gate dams were removed in 2024 (as well as J.C. Boyle Dam, and Copco 2 in 2023). We rafted past six-foot-deep accumulations of sediment. White silt pedons, if knocked into the river, floated on water, and had been colonized by grasses. Dense green vegetation lined the river's edge, but it was mostly small willows, with very few establishing trees. The stumps and snags of the former forest showed that there had been large trees before the dams.

Where Camp Creek and Scotch Creek had been uncovered by the reservoir, the willows were returning. The recovery seemed faster than in the high severity burn area of the Dixie Fire where Scotts John Creek entered Butte Creek. The Dixie burn was 5 years ago, and the reservoir removal was only 2 years ago, yet the low elevation creek in the reservoir was recovering as fast and perhaps faster than the higher elevation burned creek. This was encouraging for the recovery of rivers post-dam removal, but a reminder of the daunting task of better forest and fire management. Thinning of trees near the Dixie burn would hopefully prevent such high intensity fire in the future.

A creek recovering in Iron Gate Reservoir in June 2026, during the third growing season after drawdown.

A creek recovering in the Dixie Fire burn area in May 2026, during the fifth growing season since the fire.

Iron Gate Reservoir after 2.5 years. There may have been some restoration interventions at this site.

Dixie Fire after five years. Cutting of hazard snags along the road was the only apparent intervention at this site. There was one stand of young alders at the road crossing (not visible).

The recovery of shrub willows and forbs is only the first stage of succession, however. The green edges of the river in the former reservoir and the patches of dense willow were an improvement over a barren clearcut mudflat. But how long will it take a mature riparian forest to regrow--one with species diversity and structure mirroring what once grew there and currently grows upstream and downstream of the restoration areas? How will the deep sediment accumulations affect recovery if flooding is less frequent and groundwater is deeper? Essentially these are river reaches incised into the new higher floodplain, where the deposition creates the same challenging conditions as if the river had downcut.

Succession not only occurs in streams that are no longer inundated by reservoirs, but also in streams below dams that had been dry for decades. I've spent my career restoring Mono Basin streams below dams, where I've seen the same successional process play out in incised systems. The willows and forbs return quickly. The birds and fish come back quickly. It is exciting, encouraging, and beautiful.

Rush Creek in the Mono Basin in 1987 (this photo), 1995, 2002, 2009. Series of photos from retired CDFW biologist Gary Smith as shown in the Mono Basin Synthesis Report, 2010.

Rush Creek in 1995, when high flows returned and I arrived in the Mono Basin and began coordinating revegetation projects. Note the remnant Jeffrey pine and black cottonwood trees in the distance. Those large trees are the seed source for recovering the sagebrush-covered foreground areas.

By 2002 we had lined the streambanks with pine seedlings (not visible yet). Downstream in the bottomlands, willow flycatchers had returned, and PRBO (now Point Blue) had documented the densest population of yellow warblers in California.

By 2009 most of our pine seedlings (now visible) were taller than a person. But despite LADWP declaring this area recovered with no need for additional revegetation, the forest structure and species composition had not returned--this reach had no new cottonwoods. We stopped planting pine trees because PRBO had discovered that the presence of pines increased bird species diversity, but the dominance of pines decreased it.

July 9, 2023 - my photo. The pine trees planted by the Mono Lake Committee are now the tallest trees. But there are still no large cottonwoods. Large black cottonwoods were the primary composition of the pre-diversion riparian forest.

The initial recovery--in this case over a quarter of a century--is a vastly degraded condition compared to the pre-diversion cottonwood forest. But it is far better than when recovery began. The trajectory of restoration can be long.

The Mono Lake tributary streams were once deltaic wooded-wetlands with multiple-channeled bottomlands. Mature cottonwood trees and willows with scattered pines spread across the bottomlands. For many years, I coordinated and assisted tree planting efforts that jump-started black cottonwood, willow, Jeffrey pine, and lodgepole pine in certain areas. Incision hindered that progress while beavers and floods and channel rewatering helped in an unpredictable and dynamic way. We didn't plant enough cottonwoods, or put enough large woody debris in the creeks. After 30 years, many areas of Rush Creek (that the L.A. Department of Water and Power declared "recovered") still are a low willow forest, with a few cottonwood seedlings and saplings just getting a start. Many of these reaches still look like Glen Canyon in 2005 or the Klamath River in 2026. Restoration has reached an asymptote. Destroying an ecosystem can happen quickly, but fully restoring it can take lifetimes... especially when perturbations like incision are growing worse and inadequate flow remain.

But it is good to get the process started. It is even better to avoid destroying and degrading ecosystems in the first place.

References
Mono Basin Synthesis Report, 2010. McBain & Trush, Inc. and Ross Taylor and Associates, April 30, 2010. Mono Basin Stream Restoration and Monitoring Program: Synthesis of Instream Flow Recommendations to the State Water Resources Control Board and the Los Angeles Department of Water and Power FINAL REPORT.