If you're panning in the upper Susquehanna basin — Meshoppen Creek, Tunkhannock Creek, Wyalusing, the NY Southern Tier creeks, the Unadilla, the Chenango — and you find flour gold, here is what happened, in three sentences:
Now let's walk through the evidence, the ice story, the local stratigraphy, and the practical implications for where to spend your time.
If you're going to claim there's no lode gold source in the upper Susquehanna basin, you better have the receipts. Here they are.
In 1959, the US Geological Survey published Bulletin 1072-F, "Mineral Occurrences of New York State" — a 64-page catalog of every known mineral occurrence in New York. The authors (Luedke, Wrucke, and Graham) spent three years (1954–1956) compiling the data from published literature and the New York State Museum files. The bulletin has 38 commodity chapters — Aluminum, Antimony, Arsenic, Asbestos, Barite, Beryllium, Bismuth, Cadmium, Chromium, Cobalt, Copper, Feldspar, Fluorspar, Garnet, Graphite, Gypsum, Iron, Lead, Lithium, Magnesium, Manganese, Mica, Molybdenum, Nickel, Niobium, Peat, Platinum, Potash, Pyrite, Quartz, Salt, Sand and Gravel, Silver, Stone, Talc, Tin, Tungsten, Zinc.
Notice what's missing? There is no Gold chapter. There is no Gold and Silver chapter. There is no entry for Gold in the commodity list.
The word "gold" appears exactly once in the entire 64-page bulletin. The single mention is in the Pyrite chapter: "Gold Hill prospect, southeast of Oxbow" in Jefferson County, listed under pyrite occurrences. That's it. One mention, under pyrite, in a 64-page catalog. The 1959 USGS compilers found no significant lode gold in the published literature or the NY State Museum files. Arsenopyrite gets zero mentions. The word "sulfide" gets zero mentions.
The absence of a Gold chapter is itself the conclusion. There is no productive lode gold deposit in New York State. Nothing has changed since 1959. No major lode gold discovery has been made in NY in the 67 years since.
You might be thinking: well, the Adirondacks are old metamorphic rocks, surely they're a gold source? No, they're not. The Adirondacks are Grenville-age (1.0 to 1.3 billion years old) anorthosite, mangerite, charnockite, and granite — the "AMCG suite" (anorthosite-mangerite-charnockite-granite). That's the same age and kind of rock as the Canadian Shield basement — but it's the wrong mineralogy for gold.
The Adirondacks are NOT Archean greenstone belt geology. The greenstone belts are what host the giant gold deposits of the Canadian Shield (Abitibi, Red Lake, Hemlo). The Adirondacks are a separate dome of Canadian Shield-affinity Grenville basement, with a completely different mineralogy and metallogeny. There are no gold deposits of any consequence in the Adirondacks. The Adirondacks are not part of the Appalachian chain — they're an outlier of Shield geology pushed up through the surrounding sedimentary cover.
The Abitibi Greenstone Belt. About 400 miles north of where you're panning. Carried south by the Laurentide Ice Sheet. Deposited in the till. Concentrated by water and gravity into the modern stream networks. That's the story.
This is the part where the official modern geological survey confirms what the 1959 bulletin implied. The USGS Scientific Investigations Report 2022-5069 (Heisig and Fleisher, 2022) is the definitive recent study of the Oneonta area — exactly where the user prospects. Here's what they found about the bedrock:
"The study area lies within the dissected northern Appalachian Plateaus Province of central New York State, wherein gently southward-dipping shale, siltstone, sandstone, and conglomerate bedrock units are incised by northeast-southwest- to east-west-oriented valleys. Shale, siltstone, and sandstone of the Hamilton and Genesee Groups underlie all of the valleys and uplands in the northern half of the study area."
Read that again. Hamilton and Genesee Groups. Shale, siltstone, sandstone. Gently southward-dipping. These are unmetamorphosed sedimentary rocks laid down in a Devonian sea roughly 380 million years ago. They are not the kind of rocks that host primary gold deposits. The gold in these rocks, if any, is at parts-per-billion levels, locked in trace pyrite, and is not economically recoverable. The rocks themselves are not a gold source.
Cross-reference with NYSGA 1977 A-5 (Fleisher): "deeply dissected middle to upper Devonian clastic stratigraphy. Bedrock strata include interfingered and discontinuous beds and lenses of sandstone, siltstone, shale and sparse conglomerates of the Hamilton and Genesee Groups." Same story. The local bedrock is the wrong kind of rock to be a lode gold source.
So: the bedrock says no lode gold. The 1959 USGS bulletin says no lode gold. The 2022 USGS report says no lode gold. The Adirondacks are Grenville basement, wrong mineralogy. The only remaining source is glacial drift from somewhere to the north. The only thing north of the upper Susquehanna basin is the Canadian Shield, and the only gold region on the Shield that's the right kind is the Abitibi.
To understand how Canadian gold ended up in your local creek, you have to understand what happened during the last ice age. Specifically, you have to understand the Laurentide Ice Sheet and the Woodfordian Substage of the Wisconsin Glaciation.
Twenty-six thousand years ago, at the Last Glacial Maximum (LGM), the Laurentide Ice Sheet covered everything from the Canadian Arctic to Long Island, northern New Jersey, and southern Pennsylvania. The ice was over a mile thick in places. In the Oneonta area (your home panning ground), the ice surface was approximately 3,000 ± 1,600 feet above current sea level, per Clark (1992) in the GSA Bulletin. That's 3,000 feet of ice on top of what is now a small city in upstate New York.
The ice sheet had several major drainage basins. The Labradorean Sector flowed from Labrador south through Quebec and into New England. The Hudson Sector flowed from Hudson Bay southwest through Ontario and into the Great Lakes / St. Lawrence region. The Keewatin Sector flowed from the Canadian prairies southeast into the Dakotas and Minnesota. The upper Susquehanna basin was at the boundary of the Labradorean and Hudson sectors, with the dominant ice-flow direction being generally south to southwest during the Woodfordian maximum.
The ice picked up everything in its path. As it ground over the Canadian Shield, it entrained gold-bearing rock fragments — including gold from the Abitibi Greenstone Belt, from the Red Lake district, from the Hemlo deposit, and from hundreds of smaller gold occurrences across the Shield. The gold got mixed into the basal till — the debris-rich layer of ice that was in contact with the bedrock.
Here's where the local picture gets nuanced. The upper Susquehanna basin wasn't covered by a uniform continental ice sheet. The Appalachian Plateau created valley ice tongues — 6 to 12 miles long, behaving like alpine glaciers rather than a continental sheet. P. Jay Fleisher (SUNY Oneonta) and his colleagues have documented this in detail since the 1970s. The Finger Lakes — those long, narrow, deep lakes in central New York — were carved by these valley ice tongues, and similar tongues carved the upper Susquehanna, the Chenango, the Unadilla, and their tributaries.
The valley ice tongues matter because they changed how the till was deposited. Instead of a uniform sheet of till across the landscape, the valley ice tongues produced thick valley-fill stratified drift — sorted and layered deposits of sand, gravel, silt, and clay laid down as the ice retreated. The gold-bearing basal till was reworked by meltwater. The gold got concentrated in the same places all placer gold gets concentrated — in basal gravels, in ice-contact deposits, in stream channels, in heavy mineral lags.
Fleisher (1993) — and this is the paper that really addresses your question — applied the modern Bering Glacier (Alaska) analog to argue that most of the upper-Susquehanna glacial sediment is NOT far-traveled Canadian Shield debris. Instead, most of it is local valley-wall and upland material that was eroded by short valley ice tongues. This is important because if the bulk of the sediment is local, then the gold in it has to come from somewhere too. Either (a) it's far-traveled Shield gold that was mixed in by the active ice, or (b) it's trace gold eroded from local sulfidic Devonian shales (the Genesee Formation has black shales that contain pyrite).
This is an open research question. No formal gold-grain morphology study or till geochemistry survey has been done in the upper Susquehanna. The PA Geological Survey doesn't have a till geochemistry program. The Ontario Geological Survey and the Quebec Ministry of Natural Resources routinely publish till geochemistry open-file reports — no equivalent exists for NE PA or the NY Southern Tier. This is a research gap you could potentially help close.
Let's talk about the source. The Abitibi Greenstone Belt.
The Abitibi is in the southern Superior Province of the Canadian Shield, straddling the Ontario-Quebec border. It's one of the largest gold-producing regions on Earth:
The most famous mines: Hollinger, McIntyre, Dome, Porcupine, Kerr Addison, Lamaque, Sigma — these were the giants of the early 20th century. Total production from the Timmins-Porcupine camp alone is over 70 million ounces. Add in the Val-d'Or, Cadillac, and Larder Lake camps and you approach the 170 million ounce figure.
Most of the Abitibi is covered by Quaternary till and lacustrine sediments. Exploration in the Abitibi itself uses the same techniques we use to trace gold back from the NE US: till geochemistry, indicator mineral surveys, gold grain morphology along ice-flow vectors. The Geological Survey of Canada Bulletin 540 "Drift Prospecting" is the classic reference. The Ontario Geological Survey publishes open-file till geochemistry reports for the Abitibi and the rest of the province. The PA Geological Survey does not have an equivalent program. That's a real gap.
For more on the Abitibi, see Monecke et al. (2017) Chapter 32 in "Geology of the World's Major Gold Deposits and Provinces" (SEG Special Publication), available through GeoScienceWorld.
The Abitibi is the canonical source. When the ice sheet ground south over the Abitibi, it entrained gold-bearing rock fragments. Those fragments got incorporated into the basal till. The ice flowed south and southwest. The basal till got deposited in what is now the upper Susquehanna basin. The till got reworked by meltwater. The gold got concentrated. You're panning the result.
If you collect a few gold grains from your local creek and look at them under a microscope, you're looking at the end of a 2,700-million-year journey. The grain started as a vein deposit in Archean greenstone in Canada. It was eroded by ice. It was carried hundreds of miles. It was deposited in till. It was concentrated by water. And now it's in your pan.
This is the practical part. The ice didn't retreat all at once. It paused, readvanced, paused again. Each pause left a recessional moraine — a pile of debris marking where the ice front sat for a while. The pattern of moraines in the upper Susquehanna basin tells you where the ice-margin stood at different times, and the gold-bearing drift is associated with these margins.
From south to north, the six local ice margins in your area are:
| # | Margin | Age (years ago) | Significance |
|---|---|---|---|
| 1 | Wells Bridge | ~16,000 (breached ~14,000–14,500) | Southernmost ice position in your area. Wells Bridge is between Oneonta and Sidney on the Susquehanna. All drift south of Wells Bridge is outwash from this moraine. |
| 2 | Oneonta | ~17,000 | North of Wells Bridge |
| 3 | New Berlin | ~17,500 | North of Oneonta, on the Unadilla |
| 4 | Cassville-Cooperstown | ~18,000 | Formed by a readvance, not simple retreat. Northernmost extent of a brief ice re-advance. Crosses Otsego County in a prominent belt. |
| 5 | Middleburg | ~19,000 | North of Cassville-Cooperstown |
| 6 | Valley Heads | ~19,500 | Northernmost, also a readvance. Marks the southern limit of the Finger Lakes-style valley glaciation. |
The Wells Bridge moraine is the most important for the southern tier. It's where the ice made its last push into the southern Susquehanna valley. It completely blocked the Susquehanna valley after ice retreat, damming glacial Lake Otego to 1,140 feet elevation. The lake silt and clay deposits above the Wells Bridge are distinctive — flat valley floors with heavy clay soils that prospectors know to test carefully (clay is hard to pan through, but the gold can be there underneath).
The glacial drift in the upper Susquehanna basin is not one thing. It's a complex package of different deposits laid down under different conditions at different times. Understanding the package is the key to knowing where the gold is most likely to be.
From the USGS SIR 2022-5069 and the NYSGA 1977 A-5 (Fleisher), here's the stratigraphy:
On the hilltops and upper slopes, the drift is lodgment till — dense, unsorted, overconsolidated material plastered onto the bedrock by the weight of the ice. Up to 30 feet thick on the stoss side of hills, 160+ feet on the lee side (the classic "till shadow" effect described by Coates 1966). Lodgment till has gold in it, but the gold is scattered, not concentrated. Hard to pan effectively.
In the valleys, the drift is 200 to 440 feet of stratified drift — sorted, layered, water-laid deposits. This is where the gold is most concentrated. The package includes:
From NYSGA 1977 A-5: the drift "is almost entirely sorted, even in moraines, and consists of glaciofluvial and glaciolacustrine gravel, sand, silt and clay." Even the moraines are mostly sorted material, not classic unsorted till. That tells you that water was the dominant agent of deposition, and water sorts by density — gold sinks, lighter stuff washes away.
Now the practical question. Given everything we know about the ice, the bedrock, and the drift, where does the gold actually concentrate? Here are the five highest-priority target types, in order of promise.
This is the top target. Subaqueous fans, eskers, and deep deltaic terraces — sorted sand and gravel that was deposited at the ice margin in standing water (glacial Lake Otego, etc.). The gold-bearing basal till was reworked by meltwater and concentrated in these ice-contact deposits. The lacustrine silt and clay that overlie them are a marker — if you see flat valley floor with heavy clay soils and you're in the Lake Otego or Lake Otego-equivalent area, you want to sample through the clay into the sand and gravel below. Most modern creeks have cut down through the clay into the ice-contact gravels — that's where to pan.
Wells Bridge, West Davenport, Portlandville, Cassville-Cooperstown. Moraines represent pauses in ice retreat and therefore concentrate entrained debris — including any gold. The proximal side of a moraine (facing the direction the ice came from) tends to have higher gold concentrations than the distal side. The moraine itself is a high-priority target. The Wells Bridge moraine is the southernmost in your area, and the most accessible from the Susquehanna River main stem.
Where the meltwater stream exited the ice margin and spread out into a wider valley, the water slowed down and dropped its sediment load. The coarsest material (and the heaviest, including gold) was deposited first, in the proximal outwash. The head-of-outwash position is a high-energy environment with good gold trapping. Stream terraces cut into the outwash expose the gold-bearing basal gravels.
Every modern creek in the area has cut down through lacustrine silt into ice-contact gravels. The contact zone is where you want to sample. In some places, ancient stream channels (paleochannels) are preserved beneath till — these can be incredibly rich because they were concentrating gold in a pre-glacial or interstadial stream system, then got buried and preserved. The challenge is finding them — geophysics, drilling, or careful stratigraphic work can identify them.
The Susquehanna itself drains all the moraines and outwash. The modern river and its ancestral channels are the ultimate concentrator of any gold that was liberated by ice retreat. The big river can be hard to work (deep, fast, lots of clay), but the gravel bars and the ancient channel deposits along its course are valid targets. The smaller tributaries that feed the Susquehanna are usually more accessible and more concentrated per cubic yard.
OK. Let's put it all together. Here's how to apply this knowledge to your actual weekend trips.
Find a map of the Wells Bridge, West Davenport, Portlandville, and Cassville-Cooperstown moraines. The NYSGA 1977 A-5 and 1984 B-8 field trip guides have the maps. Pick a moraine that's accessible (public land, road access). Walk the streams that cut through the moraine. Pan the gravels. The proximal side (facing north, toward where the ice came from) is your best bet.
Find a flat valley floor with clay soil — that's lacustrine deposit. Look for a modern creek cutting down through it. Sample the sand and gravel beneath the clay. If you can find a place where the creek is cutting through ice-contact gravels (you'll see well-sorted sand and pebble gravel with occasional larger cobbles), that's your spot. Classify, pan, repeat.
Where a valley narrows upstream and opens out downstream, that's a classic head-of-outwash position. The ice-margin stream debouched into a wider valley and dropped its bedload. Look for terrace remnants on the valley walls — they preserve the outwash surface. The basal gravel of the outwash is your target.
The big river can be intimidating. But the gravel bars and the places where smaller tributaries deliver concentrated material to the main stem are valid targets. Look for areas where the river bends and the inside of the bend has a gravel bar (the classic "point bar" deposit). Look for places where the modern channel is cutting into glacial outwash. Use a highbanker or a suction dredge if you have access to one. The clay is your enemy here — classifier and sluice aggressively.
Steep tributaries draining the uplands have more energy and can carry larger gold particles. Low-gradient tributaries in the valley floor deposit their gold load. The transition zone between steep and low-gradient is a natural gold trap. Pan the transition zones.
New York has "King's Law" — all gold found on state-owned land belongs to the state. Panning for recreational purposes is generally legal and tolerated, but you cannot sell the gold without state permission, and you cannot use mechanized equipment in many waters without a permit. Pennsylvania's rules are different — check the current DCNR rules for the specific stream and equipment you plan to use. Don't be the prospector who gives the rest of us a bad name.
You can recognize the high-potential deposits in the field:
You want the first two. Skip the clay and the dense till unless you have specialized equipment.
Here's the thing that makes this a live problem, partner. The August 2026 research session uncovered something that the published literature hasn't resolved: what is the immediate source of the gold in your local drift?
There are two competing hypotheses:
The classic model. The Laurentide ice sheet ground over the Abitibi, picked up gold-bearing rock fragments, carried them south, and deposited them in the till. The gold you're panning is from Canada, 400+ miles to the north. Gold grain morphology should show abraded, rounded, often flattened grains — the classic "far-traveled" texture from being tumbled in ice and water for hundreds of miles.
The alternative. Most of the glacial sediment in the upper Susquehanna is local valley-wall and upland material (per Fleisher 1993's Bering Glacier model). If the sediment is local, the gold in it might be too. The Genesee Formation contains black shales with pyrite, and trace gold in pyrite is well-documented globally. The gold you're panning could be from these local Paleozoic rocks. Gold grain morphology should show more delicate, less abraded textures — the gold hasn't traveled as far.
If Hypothesis A is correct, the gold should be distributed throughout the drift, with concentrations in the ice-flow paths and the places where ice-margin processes concentrated the basal till. If Hypothesis B is correct, the gold should be concentrated near the outcrop pattern of the Genesee Formation and other sulfidic Devonian units. Different exploration models. Different targets. The till geochemistry would tell us which one is right.
If you collect a few gold grains from different spots — say, from Meshoppen Creek, from Tunkhannock Creek, from the Wells Bridge outwash, and from somewhere north of the Cassville-Cooperstown moraine — and look at them under a microscope or send them to a lab for analysis, you're doing real science. The combination of grain morphology, trace element chemistry, and spatial distribution could resolve this question.
You don't need a PhD. You need:
Compare the grain morphology from your sites. If they're all the same — uniformly abraded, uniformly rounded — that's evidence for far-traveled Shield gold. If they vary — some delicate, some abraded — that's evidence for a mix. If you find delicate, angular, locally-sourced gold in specific spots, you've found evidence for the local Paleozoic source hypothesis.
This is a real research project that doesn't require a million-dollar budget or a university appointment. It requires boots on the ground, a notebook, and a microscope. The PA Geological Survey doesn't have a till geochemistry program. The NY Geological Survey has glacial geology but hasn't done a gold grain morphology study in your area. If you do this work, you could be the first.