THE REGION THAT CHANGED OVERNIGHT

In the earliest sources Arven examined to reconstruct the region’s historical geography, he found descriptions of lakes, rivers, and broad areas of farmland. Most settlements had been established around water sources. The places where agriculture developed could not be considered separately from the places where water was available. As the researchers tried to understand the distribution of the old settlements, they also traced this relationship. When a settlement’s location was considered alongside the surrounding farmland, the region’s way of life became clearer.

Taren was examining records from a later period. In sources describing the same landscape, most of the lakes were gone. Many of the rivers had dried up, and dry ground had replaced extensive areas of farmland. Some settlements had been completely abandoned. Others that survived had lost much of their former size. After encountering settlements that had developed around water in the earliest sources, the team saw that this arrangement had substantially deteriorated in the later records.

“The places described as farmland in the period you’re examining are described here as dry ground,” Taren said.

Arven considered the records from both periods together. In the earliest accounts, lakes and rivers defined the surroundings of the settlements. The later accounts stated that most of those same water sources had disappeared. The character of the land was not the only thing that had changed. Some settlements established around water had shrunk, while life in others had ended. The work they had begun to reconstruct the region’s earlier condition now required them to compare two very different landscapes.

“According to the earliest sources, this was a region with extensive farmland,” Arven said. “When we read them alongside the later records, we see that most of those areas could no longer be used in the same way.”

“The settlements haven’t retained their former condition either,” Taren said. “We need to consider both the abandoned ones and the ones that shrank as part of this change.”

When the team considered the time separating the two accounts, the contradiction became more pronounced. A transformation as substantial as the disappearance of lakes, the drying of rivers, and the decline of farmland would have been expected to take place over many centuries. Yet comparison of the records showed that the change belonged to a shorter period than they had expected. The difference between the old and new landscapes was considerable, but the time available for that difference to develop did not appear equally long.

As Arven compared the condition described in the later sources with the earlier period, he assessed each change separately. Lakes, rivers, farmland, and settlements were parts of the same investigation. They could all be described as a single environmental transformation. But such a description did not explain its cause or the speed at which it had progressed. Before the team were records showing the existence of the old arrangement and records showing its deterioration.

“I would have expected a change of this magnitude to take longer,” Arven said.

“So would I,” Taren said. “But we can’t explain the difference between the records from the two periods simply as a prolonged decline.”

They first considered the possibility of natural desertification. Reduced rainfall could affect water sources. Declining vegetation and long-term climate changes could also lead to the loss of farmland. The dry ground described in the later records required them to examine this possibility. To explain how the region had lost its former appearance, the researchers first focused on these processes.

Taren noted that a decline beginning with reduced rainfall could alter the region’s water supply and agricultural pattern. Arven also acknowledged that aspects of natural desertification were consistent with the resulting landscape. His objection concerned the explanation of the speed, rather than the extent, of the change. Compared with known desertification processes, the transition between the records still appeared shorter than expected. Finding a possible cause had not been enough to explain the time difference.

“Reduced rainfall, declining vegetation, and climate changes could produce a result like this,” Taren said.

“But they also need to explain the speed of the change between these records,” Arven said. “For now, we can’t account for that part.”

They did not accept natural desertification as a definitive conclusion. Although the region’s later condition had features consistent with that explanation, the speed of known processes was insufficient to explain the change they had encountered. The team continued examining how the same land had been used in two different periods. They compared maps from both periods to see how much the arrangement that had developed around water had changed.

The maps showed that the difference was not limited to a few settlements. When the earlier water sources were examined together with the later distribution of settlements, it became clear that the transformation had affected life across the region. The decline of farmland had been accompanied by the shrinking and abandonment of settlements. Livestock farming had not retained its former pattern either. Within the short period reflected in the records, the distribution of these interconnected activities appeared to have changed.

Arven assessed the changes on the maps alongside the accounts in the records. They were not drawing a conclusion about the entire region from the decline of a single settlement. Between the two periods they were comparing, the distribution of agriculture, livestock farming, and settlements had changed along with the water sources. This scale made it difficult to continue the work solely as an examination of climate history.

“We can’t limit the investigation to climate changes,” Arven said. “We need to examine why the transformation in this region appears to have happened within such a short period.”

Taren agreed. The team decided to continue reconstructing the historical geography along these lines. The records and maps from the two periods would now be considered together to investigate both the cause and the speed of the change.