Pesticides bring major benefits to modern agriculture, keeping dangerous bugs and fungi and pathogens at bay while boosting yields and making farming more efficient. But what about risks? Like any chemicals — manmade or not — pesticides can be bad for human health and ecosystems if they’re toxic enough and the amount that ends up in the environment is high enough. It’s often tough, however, to get a clear picture of the full array of problems a pesticide may cause.

A new “meta-analysis” — a review of existing scientific studies — provides some answers, but raises even more questions in the process. In water bodies near agricultural sites across the world where scientists have detected certain bug-killing pesticides, these substances exceed regulators’ allowed levels more than half the time, the analysis found. But even more concerning, the researchers report Monday in Proceedings of the National Academy of Sciences, is that most of the time we lack any data at all on insecticide levels in waters near these agricultural fields.

Sebastian Stehle and Ralf Schulz of the University Koblenz-Landau in Germany focused their study on 28 common agricultural insecticides. These range from organophosphates, older substances that generally target insects’ nervous systems, to pyrethroids, a relatively new class of pesticides that in many cases have replaced organophosphates and other old pesticides. Many of these pesticides, while toxic to the bugs they’re meant kill, may also be capable of causing harm to “non-target” organisms such as humans.

The toxicity is just one part of the equation, though. Just because pesticides are toxic doesn’t mean they’ll actually pose risk — of causing a disease or damage to the environment. Creatures’ exposures to these pesticides need to be high enough as well. Regulators typically set limits on the amount of a substance that can occur in various settings, including in surface waters, to keep the risk to human health and the environment below a certain level.

So the researchers checked the scientific literature for any data they could find on levels of these 28 insecticides in “surface waters” (such as lakes, ponds, streams and creeks) near agricultural fields across the globe. Their analysis uncovered 838 studies, capturing data from 2,500 aquatic sites across 73 countries between 1962 and 2012. In these studies, the researchers found about 11,300 measurements for pesticide concentrations in surface waters, and compared these values to risk limits set by regulators in the U.S. and the European Union.

In the end, out of these 11,300 values, more than half of them (52 percent) breached U.S. or E.U. limits, the researchers found. These breaches occurred not just in countries with weak regulatory systems but, in similar amounts, in countries with well-established regulatory systems such as the United States, Canada, Australia, EU nations and Japan.

The researchers also raise concerns about what they didn’t find in the scientific literature: Measurements of insecticide concentrations were non-existent in surface waters near 90 percent of agricultural sites. That doesn’t mean that these lands’ surface waters are tainted with pesticides, much less that the pesticide levels are too high. But as the authors put it, we won’t know what the answer is until we get the data.

And in the cases where we do have data, the researchers say, “our results seriously challenge the protectiveness” of the current measures that regulators take to assess and reduce pesticides’ risks.

That doesn’t mean we should pull the plug on pesticides. But it does mean that in many cases, we may not be doing as good a job of keeping these risks in check as we’ve sought to do.