We’re all pretty used to the well established idea that a higher diversity of bee species makes crop pollination more certain. Now, new research has shown that that principle is even more important than previously thought.

Many studies on the importance of bee diversity for crop pollination have measured pollination at a single point, or at least a narrow window, in the growing season. In the new research, Rutgers University scientists looked at pollination levels throughout the flowering period, over several years.

The study showed that many different bee species may pollinate a single crop over the course of its flowering period. Over that season-long timescale, it took two to three times as many bee species to achieve a given level of pollination, as compared to looking at just a single date.

From year to year, the same pattern persisted. It turns out that a crop’s main pollinator species in one year may not be the same the next. That being the case, it takes twice as many different species of bee to maintain the same level of pollination over a six-year span as in just one year.

The researchers found that the number of bee species was actually more important for pollination than the total number of individual bees. That is, having lots and lots of individual bees from just a handful of species would result in lower pollination levels than having relatively few bees from each of a larger number of species.

The study measured pollination levels in blueberries and watermelons at a total of 77 farms in New Jersey, Pennsylvania, and California. It was published in the journal Nature Ecology & Evolution.