A river valley in the south-west, between limestone slopes and market-garden plain, 2046
— The pheromone traps are empty. Again.
Maya was putting the tubes away in the cool box, avoiding the eyes of the two seasonal workers. For three weeks now, the adult cotton bollworm moths had stopped flying into the lower fields, even though the temperature sensors were forecasting a warm night. Just upstream, on the left bank, the tomato plants were shredded. Something did not add up.
— Did you check the counts from the footbridge trap? asked Lila, the technician who had just arrived with her soil map.
— Yes. Zero. Zero and zero. And yet the smell of the farm is the same.
To understand, Maya and Lila called on Youssef, who worked on greenhouse midges and had developed a method for reading the genetic identity card of insects. The idea was simple: every butterfly carries in its DNA marks inherited from its ancestors, and some of those marks make it more comfortable in the heat, or more tolerant of humidity. If you read those marks, you could guess where the insect had grown up, and which way it had come.
On paper, it looked like a road map: the roads are the bands of temperature and humidity that the insect follows; the villages are the little green patches where the genetic marks are common. Youssef had explained that if the environment weighs on populations, then the map of genes should look like the map of climate, not just the map of distances. That was the idea to test.
The team took moths from twenty-four plots, from the dry plateau down to the damp land at the confluence. Lila measured the humidity of the air at ground level, the night temperature and the area of host plants around each trap. Then Youssef compared the DNA. The program returned a figure: eighteen per cent of the genetic variation was explained by climate, once distance was neutralised. Maya raised her eyebrows. That was more than the fifteen per cent expected.
— Eighteen per cent, that’s a lot of insects following the weather, she said.
— Or a lot of insects moving together, Youssef replied. The problem is that the program can’t tell the difference between a butterfly that is adapting to the heat and a butterfly that just comes from the same region. If populations mix too much, the map blurs.
They then drew the map of the marks associated with temperature. A hot spot appeared, sharp, along the south-facing slope. Another, fuzzier, to the north. But when Youssef redid the calculation for the last three years, the northern spot had moved. The marks were no longer found in the same places. The method gave a photograph, not a film.
— If the photograph changes every year, how do I decide where to put my traps next season? Maya asked.
— You can’t. That’s the limit. You see today’s corridors, not next year’s. And if the insects move too fast, the signal drowns.
They still found a use for it. Along the southern slope, the hot spot coincided with a band of plots where the larvae survived the winter. Maya suggested concentrating the traps there in spring, instead of treating the whole valley. The deputy mayor, who had come to see the results, asked:
— So what’s the point of your genes?
— Not treating where it’s pointless, Maya replied. And keeping an eye on the hot band before the caterpillars come down.
Youssef put his map away. One unknown remained: the marks associated with humidity did not show up. Perhaps humidity did not matter, or perhaps the team had measured it badly. That year, the traps in the hot band caught three times as many butterflies as the others. The year before, it was the opposite. The valley had changed sides.