Archaeology from Space Page 9
We know so much about the Norse adventures to North America thanks to the Icelandic sagas.11 The Saga of Erik the Red and the Saga of the Greenlanders describe five separate trips to a place they called “Vinland,” between 999 and 1017.12 Also, we see other references to Vinland in the Icelandic Landnámabók and in Icelandic annals.13
Where and what is Vinland remains a subject of much debate in the Norse world,14 and I barely have the guts to dip my pinky toe in the ongoing discussion. If you thought the Vikings were fierce, you should see Norse specialists debating each other.
As the Norse sailed west and south from Greenland, they encountered three diverse landscapes. The first, Helluland, or “flat slab land,” was known to have rocky coasts and no trees, and today, archaeologists believe that area corresponds to Baffin Island in Canada. Markland, or “forest land,” located south of Helluland, is associated with present-day Labrador, where dense forest runs for many miles along the coast. Vinland, the third, is supposedly somewhere farther south.15
Even the translation of the name is up for debate. The “Vin-land” could be a place where the Norse could grow grapes for wine,16 but it could be named for the large number of berries in Newfoundland and elsewhere along the Gulf of Saint Lawrence that were good for making wine, as they are today. Or the vines could be just vines.17
What we do know is that after sailing south from Helluland, the Norse established more than one settlement, with at least one in Newfoundland.18 The big question is just how far south they sailed, and how far south they would have settled, if even for a single season. This question is the one my team and I decided to put to the test using satellite imagery.
First Contact
While so many searches for Norse sites had failed, one succeeded beyond anyone’s wildest expectations—and changed North American history in the process.
It began in 1960 with a Norwegian couple, Helge and Anne Stine Ingstad. In reading the Norse sagas, they thought that Vinland likely referred to Newfoundland, as its northern coast would have been the first logical area for the Norse to make landfall when sailing south from Labrador.19 After they spoke about their search to George Decker, a local fisherman, he led them to a series of foundations made from sod and covered in grass that had a similar shape to Norse longhouses.20 This goes to show you that locals everywhere know what is what.
Over the next few seasons at the site, called L’Anse aux Meadows, what they excavated was so astounding that it took years for the larger community of Norse scholars to believe them.21 Out of the houses came a spindle whorl made of soapstone, essential for spinning wool to make woolen garments.22 Iron boat rivets suggested that larger boats had made the journey there.23 Excavations in other areas revealed a forge, where iron was smelted and iron items manufactured,24 but it was the discovery of a typical Norse metal ring pin that quieted the naysayers.25
Radiocarbon tests provided multiple dates around 1000 AD,26 with the buildings having an identical shape to those in Iceland and Greenland from the same time period.27 It appeared the Ingstads had discovered the first evidence of the Norse in North America.
After the Ingstads came legendary archaeologist Birgitta Wallace from Parks Canada.28 She and her team worked in the swampier area of the site next to the beach, where they found pieces of worked wood.29 Significantly, they also uncovered a butternut, along with wood from a butternut tree, which is a species of walnut not found in Newfoundland. That’s a strong suggestion that the inhabitants of L’Anse aux Meadows sailed across the Gulf of Saint Lawrence, cutting down trees for wood and exploring.30 The evidence added up to show that the Norse lived at L’Anse aux Meadows for a short period of time, with a maximum population of about a hundred people.31 Curiously, no animal bones appeared in the archaeological record, nor was there any evidence for stables.32
Given all the trips described in the sagas, surely there had to be additional settlements. L’Anse aux Meadows could have been Vinland, or Vinland might have been a description for the entire region. The more we investigated, the more we realized that no one had ever conducted a search for potential archaeological sites along the eastern coast of North America using a systematic approach with remote-sensing technologies.
It wasn’t just Norse sites that offered exciting potential. The eastern coast of Canada and the Gulf of Saint Lawrence were home to fascinating ancient native cultures, including the Dorset,33 Beothuk,34 and Maritime Archaic.35 Discovering further traces of these peoples36 would provide valuable additions to the archaeological record.
As we prepared our research design, we wanted the most unbiased approach possible. We were not seeking any particular ancient site, but rather wanted to test the remote-sensing science to see if it would work in identifying any ancient sites, period. Studying Indigenous ancient cultures gave my team and me a good sense of the range and type of structures we might encounter. We also studied Norse longhouse and farm construction from Iceland, as well as the standard types of construction from 18th- and 19th-century settlers in the area.
A Lot of Land to Map
At this point, we had budget enough to pay for Chase’s and Dave’s time and a bit of satellite imagery. For site mapping and discovery, we normally focus on one small area, but getting high-resolution imagery for the entirety of Canada’s eastern coastline, plus down into New England, would have cost tens of millions of dollars. We had to seek an alternative approach.
Thank God for Google Earth and Bing, two open-access imagery platforms, with at least 60 percent of eastern Canada’s coastline available as high-resolution data. However, in areas with low population density, the quality of imagery can be patchy, as there is simply too little demand for the data. Lower-resolution imagery showed trees but no smaller details.
Weeks of searching followed, hours on end scrutinizing the edges of the coastline and the shores of lakes and rivers nearby. Any curious shapes or features got marked with pins for team review the next day.
About 50 sites of interest appeared around Newfoundland, our most likely hot spot for any ancient sites. Rather than purchasing high-resolution satellite imagery, which would have cost us about $30,000—still way too expensive—I found high-resolution aerial photographs from a government environmental mapping agency. A bargain at $1,000. Although these images did not have multispectral data, they did have a resolution of 25 centimeters, allowing us to assess if the features were worth more exploration.
That took us from 50 sites to 6, a number for which we could afford to use multispectral imagery. On closer inspection, four hit the cutting-room floor in swift order; for reasons of elevation or vegetation growth, putative features from the aerial photos turned out to be nothing. That left us with two sites: one about 20 miles to the west of L’Anse aux Meadows, and another about 700 miles to the south, at Point Rosee, almost on Newfoundland’s southernmost tip.
That tip, covered with lighter vegetation, jutted out from a dense forest into the Gulf of Saint Lawrence. In a 120-by-240-meter area of the satellite image, a series of dense black lines appeared, possibly the faint outlines of a few structures 8 meters wide and 20 meters long. In the south and east, a narrow black dotted line surrounded nearly all the “structures,” forming what looked like an enclosure wall of a farm. Something like farmed ridges appeared to the east.
Location of Point Rosee [MAP COURTESY CHASE CHILDS]
The BBC took us up on the suggestion of further, noninvasive investigation on the ground. At this point, I had to figure out how the permitting process worked, because you cannot just show up on a site in Canada and start surveying. It’s against the law. I emailed Martha Drake, who heads the Provincial Archaeology Office of Newfoundland and Labrador, with a report of our initial results.
Sure that I would be dismissed as a crackpot, I dreaded her response, but I could not have been more wrong. Martha was charming, gracious, and very sympathetic to the idea. On the phone, she told me she was supportive not because she thought we would find anything per se, but beca
use we would test a new methodology for the archaeology of eastern Canada, and Newfoundland in particular.
Processed WorldView-2 satellite imagery of Point Rosee [IMAGE COURTESY DIGITALGLOBE]
Permit duly acquired, that October Dave and Chase flew from Birmingham, Alabama, to Deer Lake Regional Airport, about a three-hour drive from Point Rosee. I did not appreciate the real-life consequences of putting together two brilliant people so different from one another. Chase was up early, raring to go, while Dave is more of a coffee-fueled night owl.
Even finding Point Rosee had turned into a disaster and would have been impossible if it weren’t for the help of a kindly store owner, Edwin Gale—known locally as Hockey—whose emporium contains everything from groceries and bait to guitars. He lent Dave and Chase a 4×4 and told them how to avoid the bears. Newfoundland is still wild. Up to my eyebrows in teaching back home, I got occasional texts from Chase, with encouraging messages like, “Dave nearly died today.” The forest we had seen in the satellite imagery was virtually impenetrable, and wind gusts of more than 60 miles per hour nearly blew Dave off the side of a 50-foot cliff.
Nobody ever said this was easy.
Somehow, Dave and Chase set up the necessary grid for Dave’s magnetometer survey.37 After the standard procedure of strapping the magnetometer on his back had turned him into something like a sail in the high winds, Dave had hacked a safer way of carrying it, and they completed the work in the week they had budgeted. Good thing they finished when they did: the day they left Newfoundland, Point Rosee was blanketed in several feet of snow.
Tantalizing Clues
When they returned, both thankfully alive, we sat around the computer as the image processing began. Anxiously, we watched as a series of darker lines appeared with a half-dozen magnetic spikes across the site. These spikes matched with what we had seen on the satellite imagery. Something was there, something worth investigating.
We shared the data in our report with Martha, and she suggested doing a season with limited test excavations in a series of 2-by-2-meter units and connected us with an archaeological veteran of the region, Fred Schwarz, for planning and digging. Fred told us he thought what we had found so far was probably archaeological, but we needed to know if it was an 18th- or 19th-century settlement38 or a trace of one of Newfoundland’s Indigenous cultures.
We also shared the initial results with Doug Bolender, of the University of Massachusetts Boston—the same chap we’d worked with in Iceland. Doug was cautious but excited, explaining that, no matter what, we’d developed a sound scientific approach to mapping potential archaeological anomalies in Newfoundland.
In the short time before the planned June 2015 season, we dove into the settlement history of southwest Newfoundland39—as you do, when you’re an Egyptologist. No homes or settlements could be found on any maps close to Point Rosee. All we could do was hope for the best.
A Grand Adventure
I arrived in Newfoundland a few days after Greg, Chase, Dave, and Fred had started the season. Newfoundland can deliver some brutal temperature shifts and weather changes. I had read about the “wreckhouse winds,” which were so powerful they could knock a train off the tracks, so I was a little wary. But, along with the unfailing kindness and generosity of the Newfoundlanders, I will always remember the smell of the air on the coast, like a Christmas-tree farm along a beach. The freshness and cleanness and sheer wildness were breathtaking.
Every day, our team set out from Hockey’s store on foot, tromping about 3 kilometers over extraordinarily varied terrain. While one of us drove a 4×4 with supplies to the site, the rest hiked through lightly wooded hills, passing sprays of pink, yellow, and white flowers that dotted the landscape. Walking up an old logging road, we entered dense woods with muddy ruts for us to leap over, and then spilled out onto sloping grassy hills that rolled down to the ocean cliffs and deep bogs.
Point Rosee jutted out into the ocean in the distance. So we kept walking, through dense bushes and small trees hugging the coast. Seals sunned themselves on the rocks below. On a really good day, we spotted whales. Finally, we’d enter the point, where knee-high grass and wet sod earth grabbed at our ankles and pulled us down.
Archaeology is all about the details, and any good excavation starts with a grid system for later reference. Greg and the team had begun a grid that measured 120 meters by 240 meters along the western half of the site, based around a 1974 Canadian Geodetic Survey marker, part of an ongoing 100-year-old Canadian government mapping program.40 We had trouble pinpointing the magnetometer spikes and possible satellite imagery features on the ground to test-excavate them, but it was nothing daunting.
We set up our assorted units, while Dave conducted more magnetometry. Fred, a gruff but kindly man with over 30 years’ experience excavating every possible site in eastern Canada, dug a test unit in an area where nothing had appeared on the satellite imagery, to establish the layers of natural soil on the site. Chase and I worked away to the north, where a 1-meter-wide potential “wall” feature had appeared. After a few days, Dave confirmed a find from the initial survey: a major spike of 250 in the magnetometer readings, when other areas ranged from −2 to +2. This suggested a subsurface magnetic anomaly; perhaps an area of burning or a ditch.
We set up another small excavation unit there and immediately uncovered an ovoid boulder, the tip of which pierced the ground surface. When we removed the heavy grass and roots covering the area, the boulder seemed to have been fire-cracked, with dark metal lumps that looked like they had been soldered on starting about 5 centimeters below the top ground level.
We did not think much of it at first. But further down, harder pieces the size of a quarter popped out of the ground as Fred and I dug away. We cleaned them off in one of the clear puddles nearby. They seemed to be burnt bog iron. Bog iron is found in swampy environments, when concentrations of ore form together in clumps. The Vikings smelted down these impure deposits, after various heating processes, to make iron for nails and tools. In our case, some of the chunks had what looked like bubbles—potential evidence of processed metal. Fred told us that in all his years of working in the region, he had never seen anything like it.
Bog iron from Point Rosee [PHOTO BY GREG MUMFORD]
This was a major moment for us. We could have found an area where someone in the past had intentionally heated bog iron and left waste. We expanded the unit to follow the deposit.
Digging Deeper
Since the start of our dig, it had rained nearly every day, with North Atlantic winds and freezing cold. The site, and our team, spent most of the time soaked. Rubbing my hand against the boulder, I noticed a veneer of black come off—maybe charcoal residue. The waterlogged ground had caused lots of organic material to disintegrate, and there were hundreds of what appeared to be tiny charcoal pieces no bigger than a splinter in the earth, which left black streaks against my hand if I ground them against my palm. We sampled everything we could there.
Chase and I found a roughly circular hollow ringed by stones, set into the ground at close intervals, and placed beside a boulder. Inside were nearly 20 pounds of bog iron, going down 20 centimeters. We noted that a thin gray layer covered the deposit, which looked just like ash. We tried not to get too excited.
Potential furnace feature at Point Rosee [PHOTO BY THE AUTHOR]
But Fred was finding that hard. None of the Indigenous ancient cultures known from Newfoundland used bog iron; all the known tools from the Dorset and Inuit people were made from meteoric iron, and there surely would have been some other sign of Indigenous activity, such as flint working, if this were one of their sites.41 Nor did European settlers smelt bog iron, since they imported metal items on the ships they sailed from Europe.42 And if the site had been an 18th-to-19th-century settlement or even a single house as Fred had suggested before we broke ground, we would have found pottery,43 the standard debris from every other known French and English house on Newfoundland.
This left one pos
sible origin: Norse.
With a beach a kilometer away, and a protected inlet, this seemed like an ideal location for a small camp, if not a transitory settlement. We started to dream we might have found the westernmost and southernmost evidence of Norse occupation. L’Anse aux Meadows probably did not stand in isolation: one saga referred to a place called “Hop,”44 which was warm enough for the Norse to cultivate grain, at a settlement next to a protected inlet connected to a river. Coincidentally, the Codroy Valley, where we were working, is the one place in Newfoundland known today for its farming. Conditions might have been warmer 1,000 years ago, but not enough for all of Newfoundland to be farmed.
It was easy to start connecting the dots. And more dots appeared. Greg was working in a unit just south of our “furnace.” When Doug Bolender joined us for a few days at the end of our season, he opened up the area between Greg’s unit and ours, where the darker feature had appeared in the satellite imagery. As he scraped away, exposing the unit, we heard him mutter, “It’s not possible. I don’t believe it.”
He had apparently uncovered the remains of what seemed to be a Norse turf wall—undulating strips of angled brown and black sod. Doug had seen hundreds of them. I later saw examples in Iceland, and they looked the same as what we found in Newfoundland.
While we did not come up with much beyond the bubbled bog iron, this was not surprising. All the specialists with whom I had spoken said Norse sites did not have many artifacts, just as at L’Anse aux Meadows. We left Point Rosee in high spirits, determined to return with a bigger team the next summer. Our test season seemed victorious.