The Hidden Cost of Environmental Neglect:

Groundwater, Insects, and Food Security

I live in a cave. I want to get that out of the way early, because everything that follows will sound like the ravings of a man who’s had too much sun and not enough plumbing, and frankly, that’s a fair cop. But living somewhere whose water comes off a reservoir supply that goes erratic or drops out entirely without much warning has a curious side effect: you start paying attention to water the way you’d pay attention to a dodgy heart valve. You notice the level in the tank. You notice which weeks the supply just… doesn’t show up. You notice, in short, the thing that everyone with a tap that always works has been trained not to notice at all.

So let’s talk about the tap. Specifically, let’s talk about what’s underneath it, because that’s where this whole horror show is quietly playing out, several hundred metres below the nearest branch of Tesco, in the dark, unphotographed, and largely uninsured against.

The aquifer at the end of the world

Between 1990 and 2020, global groundwater withdrawals very nearly tripled, from 227 billion cubic metres a year to 879 billion. A quarter of the world’s river basins are now classed as overexploited, and those basins alone account for 61% of all groundwater taken out of the ground. The worst offenders, unsurprisingly, are also the places doing the actual work of feeding everyone. India, Pakistan, China, Saudi Arabia, Iran, the United States and Egypt, between them, are responsible for 83% of global groundwater depletion. That’s not a rogue’s gallery of villains twirling moustaches over a map. That’s the breadbasket. That’s dinner.

And here’s the bit that ought to make the optimists in the room a little quieter: something like 70% of all groundwater pulled out of the earth globally goes straight into agriculture. We’re not draining aquifers for swimming pools and golf courses, tempting as that villain is. We’re draining them to grow the wheat that becomes the bread that becomes the sandwich you didn’t think twice about at lunch. The very system that’s supposed to guarantee food security is the one hollowing out the ground it stands on. It’s a bit like burning your house down for warmth. Effective in the short term. Poor long-term strategy.

And it isn’t just quantity that suffers. When aquifers are overdrawn, the change in pressure and flow can mobilise nasty stowaways like arsenic, molybdenum and nickel out of the surrounding rock and straight into the water table. So the water gets scarcer and dirtier at the same time, which feels less like an accident and more like a design flaw nobody wants to own.

Here’s the part I find genuinely darkly funny, in the way you laugh at a funeral because the alternative is worse. Researchers modelled what would happen if we actually did the responsible thing and stopped over-pumping groundwater tomorrow. Not “reduce emissions by 2050” is responsible. Actually stop. The answer is that halting groundwater depletion without any other supporting policy would push up to 26 million more people into the risk of hunger by 2050, mostly because rice and wheat production would fall and prices would rise. Rice prices alone could climb by 7.4%, wheat by 6.7%.

Read that again slowly. We have built a food system so dependent on an unsustainable resource that even stopping the harm causes harm. That’s not a crisis with a villain. That’s a trap, and we built it ourselves, brick by cheap, subsidised brick.

The insects nobody invited to the meeting

Meanwhile, underneath everyone’s notice in an entirely different sense, something smaller and stranger has been quietly unravelling. Roughly three-quarters of the world’s crops rely on pollinators to some degree, though that only accounts for around a third of total global crop production, because most of our staple cereals don’t need them at all. Studies suggest that if pollinating insects vanished tomorrow, crop production would drop by around 5% in wealthier countries and 8% in lower and middle-income ones. Numbers like that are easy to shrug off until you remember which countries can least afford an 8% hit to their food supply, and it isn’t the ones running the studies.

There’s a piece of research I keep coming back to, from Western Nepal, that has nothing to do with satellites or global models and everything to do with a man checking on his beehives. Interviews with 116 local beekeepers found a 44% decline in the number of occupied hives and a 50% drop in honey production per hive between 2012 and 2022, with beekeepers themselves pointing to a changing climate and vanishing wildflowers as the cause. In that region, honey sales make up 16% of household income, and the native honeybee is responsible for over half of all flower visits on apple, cucumber and pumpkin crops. That isn’t an abstraction in a PNAS paper. That’s someone’s actual living, disappearing hive by hive, while the rest of us argue about whether insect decline is “really” a crisis.

And here’s where I have to be honest, because sardonic doesn’t mean dishonest. Not every headline holds up. A recent PNAS study found that pollination limitation in crops, meaning yield lost to insufficient pollination, has actually nearly halved since the 1980s, largely because we’ve got much better at deploying managed pollinators like commercial honeybees. In other words, wild pollinator declines are not yet translating into global yield losses. Good news, on paper. But read it the way I read most good news these days: as a description of a coping mechanism, not a cure. We haven’t fixed the ecosystem. We’ve built a life-support machine for it out of trucked-in beehives, and we’re calling the beeping monitor a recovery.

Because the wild side of the ledger tells a different story, one modelling study on Europe found that losing wild pollinators could cut crop yields there by around 8%, shrink exports, push food prices up, and cost the continent roughly €34 billion a year in welfare losses by 2030. That’s not a hypothetical bee-pocalypse. That’s a spreadsheet with a number on it, sitting in someone’s drawer, largely unread.

Where the two holes meet

Here’s the thing about groundwater and insects that nobody puts in the same paragraph, and I genuinely don’t understand why, because it’s not subtle. Both are invisible while they work, and only become visible once they’ve failed. Nobody erects a memorial to an aquifer. Nobody holds a vigil for a wild bumblebee population that quietly thinned out over a decade of pesticide drift and vanishing hedgerows. These aren’t dramatic collapses with a starting gun. They’re erosions. Slow, boring, unglamorous erosions, and boring is precisely the reason they don’t make the news cycle that thrives on things that explode.

And they’re connected in ways that make the whole picture worse than either alone. Intensive irrigation, the kind that’s draining those overexploited basins, tends to go hand in hand with the exact monoculture farming that strips out the hedgerows, wildflowers and rough margins insects need to survive. You don’t get one crisis. You get a matched set, cross-subsidising each other’s collapse, while both get filed under separate government departments that apparently don’t share a hallway, let alone a strategy.

I think about this most mornings, hauling water I’ve had to think about, in a cave I chose partly because it makes these invisible systems impossible to ignore. Most people get to live entirely unbothered by where their water table sits or whether a bee turned up to do its unpaid, unthanked job on the almond blossom down the valley. That’s not a criticism, incidentally. It’s simply what infrastructure is for: making the essential things invisible so you can get on with your life. The trouble starts when the infrastructure itself is running on a system of depletion nobody’s accounted for, and the invoice, when it finally arrives, comes due for everyone at once, insured or not.

I don’t have a tidy conclusion for you. I don’t think the aquifers do either. Some things just keep dripping, quietly, until one day they don’t.

Until Next Time