New Technology Reveals Extent of Early Hawaiian Agricultural Systems

Thousands of farmed acres previously unknown to science identified  

HONOLULU, Hawai'i | November 17, 2009

Early Hawaiian agriculture was far more extensive and complex than anyone has fully understood, according to new research by scientists blending state-of-the-art technologies with traditional dirt archaeology.

Until now, it has been difficult to prove the full scope of Hawaiian farming technology, said Samuel M. Gon III, ecologist, cultural advisor and senior scientist with The Nature Conservancy. “At the peak of Hawaiian population, there were perhaps a million people. It takes thousands and thousands of acres to feed all those people,” Gon said. Where was all that farmland?

A new research tool has now identified thousands of farmed acres not previously known to science - including a vast dry-land agricultural field system in the grassy plains of Ka‘ū on the Big Island.

“Traditional agricultural systems were either abandoned or contracted greatly in the early 19th century as a consequence of population collapse in Hawai‘i,” said archaeologist Patrick V. Kirch. “By the time that E.S.C. Handy and (Mary) Pukui did their classic work on Hawaiian agriculture in the 1930s, the whole fact that there had once been huge dry-land systems had virtually been forgotten, and the extent of even the wet-land systems (some of which were still in use) was much smaller than had once been the case.”

In some cases, plantation-scale sugar and pineapple cultivation have destroyed any evidence of Hawaiian agriculture. In other cases, early Hawaiian language newspapers make reference to agricultural systems that aren't known today. And many Hawaiian farming areas simply haven't been found because they're in places were no one has looked.

A team of scientists from different institutions and different disciplines set out to find them. The team included anthropologist Thegn N. Ladefoged of New Zealand's University of Auckland, archaeologist Kirch of the University of California at Berkeley, Gon of The Nature Conservancy of Hawai‘i, soil scientist Oliver A. Chadwick of the University of California at Santa Barbara, environmental scientist Anthony S. Hartshorn of Arizona State's School of Earth and Space Exploration, and Peter M. Vitousek of Stanford's Department of Biological Sciences.

They describe their work in a new paper in the Journal of Archaeological Science, entitled “Opportunities and constraints for intensive agriculture in the Hawaiian archipelago prior to European contact.”

The team used sophisticated Geographical Information Systems (GIS) (modeling) mapping, which provides remarkably detailed information about the surface of the land. They (analyzed) identified the key geographical characteristics of two key kinds of Hawaiian agriculture -- wet-land and dry-land.

“We were testing our ideas to see if we did understand where farming could have taken place,” said Stanford's Vitousek, who is conducting a multi-year investigation of the Kohala Field System on the Big Island.

Using sophisticated computer mapping models, the scientists looked at availability of surface water for irrigation, rainfall, elevation, amount of slope, soil fertility and more, and put the data together to identify places across the state where it seemed possible for early Hawaiians to grow crops with their two dominant agricultural technologies: flooded taro fields like those in Hanalei Valley on Kaua‘i and massive dry-land farming regions like the Kona Field System. (There are other forms of Hawaiian agriculture, but those cover by far the largest area, Vitousek said.)

Then they compared what the computer models turned out with what was known from more than a century of archaeological research on the ground. The work would provide insight into both what hasn't yet been found, and what's been lost.

“We were trying to understand why there is agriculture in places that we knew about, and more about those we didn't,” Vitousek said.

In one finding, the computerized model predicted that a massive part of Ka‘ū was perfectly suited to Hawaiian dry-land agriculture. The model targeted thousands of acres above the South Point wind generator farm and below the Mamalahoa Highway. While the existence of a field system in Ka‘ū was known, archaeology knew of nothing that matched the extent suggested.

“We went to Google Earth to take a look,” said Vitousek. For someone familiar with Hawaiian dry-land agriculture and its intricate web-like network of earthen and stone walls, the imagery leaped out from the screen. “There it was! You could see the walls. They're unbelievable.”

“We know virtually nothing about this. That's where we need to go next,” Kirch said of the Ka‘ū Field System.

In areas where archaeology has already identified Hawaiian farming areas, Kirch said that the model's predictions fit so well that researchers are confident that its predictions are also valid in regions on which no prehistoric farming has been known.

“Our research was designed to figure out where both wet and dry systems were likely to have been distributed at the peak of Hawaiian population. We did the GIS modeling using known parameters, and then compared the results against archaeological evidence for known systems.

“The fit between these two is good, giving us confidence that the predictions for areas where we don't have archaeological data, or where later sugar and pine plantation agriculture destroyed systems, will also be strong.

“Hence, we now have a firm empirical basis for predicting where intensive agriculture was practiced across the archipelago,” Kirch said.

Researchers were shocked at the extent of dry as well as wet-land agriculture -- and how differently they were distributed -- dry-land farming dominant on the newer islands of Maui County and Hawai‘i, and wet-land on the older islands of O‘ahu and Kaua‘i.

“I already had a pretty good idea of the extent of taro lands on O‘ahu, but the area with high potential for wet taro on Kaua‘i blew me away. Combined, O‘ahu and Kaua‘i were clearly the real 'breadbaskets' of the archipelago,” Kirch said.

That food resource may explain why two famous chiefs, Maui's Kahekili and Hawaii's Kamehameha, worked so hard to gain control over the older islands, he said.

The technology that developed this information on Hawaiian agriculture can readily be deployed for a range of different purposes, Gon said. One conservation application would be to predict, based on specific plants' growth requirements, the zones where those plants were most likely to have grown.

Another application would be to predict, in a time of accelerating climate change, where a rare plant's habitat requirements could be met in the future, he said.

A key feature of the new research is how extensively early Hawaiians applied their farming technology. In most cases, if it was possible to grow crops in a location, the early residents of the Islands found ways to do so.

“If a million mouths could be fed back then, this points to a future where we can wean our reliance on food from the outside world,” said Gon. “We could and should look to a future where we can sustain ourselves without our current precarious reliance on a global transportation system driven by fossil fuels.”

The Nature Conservancy is a leading conservation organization working around the world to protect ecologically important lands and waters for nature and people. The Conservancy and its more than 1 million members have protected nearly 120 million acres worldwide. Visit The Nature Conservancy on the Web at

Contact information

Grady Timmons
Communications Director
(808) 587-6237

Related Links

We’re Accountable

The Nature Conservancy makes careful use of your support.

More Ratings