As sea-level rises, as it has been throughout the Holocene, and now rises faster due to human-accelerated climate change (yes, MAGA-world, that's a fact), bottomland hardwood swamps near the coast are being converted to oligohaline* and brackish marshes. As water levels rise, trees that can't tolerate regular flooding and constantly saturated soils die off. Then, as saltier water intrudes, trees that can tolerate being wet all the time but cannot tolerate salinity die off. This is creating so-called "Ghost Forests" (see my earlier Ghost Treespost). Generally the last tree standing, except for a few salt-tolerant shrubs and some holdouts on isolated higher spots, is baldcypress (Taxodium distichum) which can tolerate a little bit of salinity, which swamp tupelo and water tupelo (Nyssa biflora, N. aquatica) cannot. Baldcypress wood is famously decay-resistant; and the white trunks of dead trees are called ghost cypress.
Oligohaline marsh and ghost cypress, Upper Broad Creek (top) and Cahoogue Creek (bottom), North Carolina.
You can see this happening throughout the southeastern U.S.A. coastal plains, but the general swamp-to-marsh conversion has been going on, albeit more slowly (see How to Drown a Swamp), for a long time. The evidence is under, or just above the water--stumps and roots of bald cypress at the edge of the marshes, with no living (and often no upright dead) cypress anywhere close by.
Baldcypress stump, Hancock Creek, N.C.
Cypress wood is renowned for its decay resistance, which is attributable largely to an oil called cypressene. In a recently-deceased cypress trunk, log, or root, the bark and sapwood decompose fairly rapidly. The heartwood, however, can persist for very long periods, particularly underwater, or buried in wetlands (fungi and insects can eat away some of the heartwood, creating tree cavities and some interesting patterns in the wood, such as sought-after "pecky" cypress).
Remnants of a cypress stump adjacent to black needlerush (Juncus romerianus) marsh, Upper Broad Creek
How long is very long? Good question. There exists a cottage industry for lumber from cypress "sinker logs." From the 1880s to about 1920, a lot of virgin and old growth cypress was logged in the south, and was typically dragged from the swamps to canals, creeks, and rivers to float in log rafts to market. Some of the logs sank (one source estimates 10%). Because sinkers were difficult and expensive to recover at the time, nobody did. But in the late 20th century it was realized that these often huge, high-commercial-quality logs--with the heartwood typically not noticeably decayed--were worth recovering. Most of this activity is in the Gulf Coast states, but at least one operation is going on the Cape Fear River, N.C. (https://www.oldgrowthriverwood.com), and another in Charleston, S.C. (https://www.facebook.com/HeartwoodSouth/). Obviously cypress wood can last for >100 years underwater.
Sinker cypress wood for sale at Heartwood South
Ancient buried cypress wood is called "subfossil" rather than fossil because the heartwood is preserved more or less as-is (or as-was), and has not been fossilized by mineral replacement. Buried subfossil cypress has been reported in the southern U.S. since the 18th century. Sand mines along the Pee Dee and Lynches Rivers, S.C. have uncovered subfossil logs roughly 12 m below the ground surface, up to 2.4 m in diameter and up to 29 m long (Stahle et al., 2005). Researchers from the University of Arkansas Tree Ring Lab who studied the logs reported that other hardwood logs were also present, but none as well preserved as the Taxodium distichum. Radiocarbon dates indicate their ages range from about 25 to 45 thousand years old. The report also reviewed other accounts of subfossil cypress, including stumps in growth position with roots and knees still attached uncovered by 20th century construction in Washington, D.C., estimated to be about 100 ka. Rooted cypress stumps and fallen logs exposed along the Intracoastal Waterway in Horry County, S.C. are believed to date to 125 to 135 thousand years BP. There's more about subfossil cypress logs and what can be learned from them in Stahle et al. (2012).
Along eroding shorelines of the Neuse River estuary, N.C., a swamp paleosol is exposed with cypress roots and stumps in growth position. This paleosol is under the ~200 ka Flanner Beach formation and is at least that old. It overlies the James City formation, which may be up to a million years old, but is generally thought to be about 700 ka (Miller, 1986a; 1986b).
Without radiocarbon dating the age of the underwater marsh edge cypress cannot be determined, but we can say they could be very old. My guess is less than 4 ka, which is about the time the modern estuaries of the Carolinas became established.
It is not easy, at least if you are no better a photographer than I am, to get good pictures of the underwater cypress, but below are a few attempts. All these examples are adjacent to brackish marshes with no trees nearby. They are all from tributaries of the Neuse River estuary, not because that's the only place this is happening, but because that's where I live and therefore kayak a lot. Early 20th century records of the U.S. Army Corps of Engineers dredging operations in the Pamlico River at Washington, NC, report encountering in situ cypress stumps at the bottom of the river, for one example.
Cahoogue Creek, NC
Upper Broad Creek, NC
A synthesis of Holocene, recent, and predicted future sea-level rise for the N.C. coast was produced by Kopp et al. (2015). Over the past 11 ka, the region experienced episodes of RSL (relative sea-level) rise acceleration and deceleration, but no periods of fall or stillstand. RSL reached to <10 m below modern levels by about 2000 BCE (or about 4 ka BP), when modern estuaries became established in roughly their modern locations. My guess is that at about that time some of the tributaries to the Pamlico Sound estuary were flanked by bottomland hardwood cypress-tupelo swamps. As RSL continued and continues to rise, salinity increased, first killing off the tupelo and other salinity-intolerant trees. In addition to dying sooner, the wood of these trees is less decay resistant than cypress, the last canopy-size tree to go. Tributaries such as the ones where pictures above were taken first saw formation of ghost forests and marsh conversion, eventually reaching a point where nothing is left of the cypress but a few snags (standing dead trees), stumps, and submerged roots and stumps. That process is ongoing and will continue and probably accelerate. Whether there is a net loss of bottomland hardwood swamps will depend on the extent to which they can expand along their inland and upstream margins. In terms of marsh area, the swamp-to-marsh conversion is offset by erosion and drowning. In most areas, including the Neuse River, there has been a net loss.
*Salinity between 0.5 and 5.0 ppt. Fresh water is <0.5; ocean water is 33 to 35 ppt.
References cited
Kopp, R.E. & 3 others. 2015. Past and future sea level rise along the coast of North Carolina, USA. ClimaticChange 132, 693–707.
Miller, W.M., III. 1986a, Community replacement in estuarine Pleistocene deposits of eastern North Carolina. Tulane Studies in Geology & Paleontology 19, 97-122.
Miller, W.M., III. 1986b. Paleoecology of benthic community replacement. Lethaia 19, 225–231.
Stahle, D.W. & 3 others. 2005. Ancient Baldcypress Forests Buried in South Carolina. Tree Ring Laboratory, Dept. of Geosciences, University of Arkansas.
Stahle, D.W. & 9 others. 2012. Tree-ring analysis of ancient baldcypress trees and subfossil wood. Quaternary Science Reviews 34, 1-15.
My own related research
Phillips, J.D. 2024. Sequential changes in coastal plain rivers affected by rising sea-level. Hydrology 11, 124.
Phillips, J.D. 2024. Ghost cypress as indicators of sea-level rise in the Neuse River, North Carolina. Wetlands Ecology and Management 32, 287-302.
Phillips, J.D. 2023. Landscape change and climate attribution, with an example from estuarine marshes. Geomorphology 430: 108666.
Phillips, J.D. 2018. Environmental gradients and complexity in coastal landscape response to sea level rise. Catena 169: 107-118.
Phillips, J.D. 2018. Coastal wetlands, sea-level, and the dimensions of geomorphic resilience. Geomorphology 305: 173-184.
