Eleven in the morning, late May. I was kneeling in the bean row with a soil thermometer when the ground under my right knee gave a hollow thump. Not a rock. Not a root. A void.
Six weeks later that spot swallowed a full spade of compost and never filled back in.
That is how most gardeners meet a woodchuck. Through the floor of their own plot.
I have worked the same half acre for nineteen seasons. Groundhogs have cost me more than rabbits and deer combined. Here is what holds them out. Here is where the usual advice falls apart.
Read the burrow before you plan anything
The main entrance is a clean circle. It runs 10 to 12 inches across, which is 25 to 30 cm, and it sits behind a fan of freshly pushed soil (Internet Center for Wildlife Damage Management). UConn gives the same 10- to 12-inch diameter for the main door, with a mound of soil piled beside it (UConn Home and Garden Education Center).
That mound is not decoration. A woodchuck stands on it to scan the yard before it moves. Fresh soil there is dark and crumbly.
Secondary entrances are the trap for the unwary. They are dug from below, so they carry no mound. Kansas State counts two or three per system, and sometimes as many as five (Kansas State University Extension). Missouri’s conservation department says at least two, and most of those second doors hide in tall grass or tight against a fence line (Missouri Department of Conservation).
I learned that the hard way in 2016. I blocked a single hole with a cinder block and a bag of gravel. The chuck used a back door eleven feet away for a week. Then it reopened the front one around my gravel.
Tunnel dimensions decide what comes next. Burrows run 25 to 30 feet long in the standard accounts, at depths of two to five feet (Kansas State). The older federal handbook gives a wider spread, from 8 feet to 66 feet (ICWDM). You are not dealing with a scrape. You are dealing with a small apartment block.
Where the animal goes, and when
Woodchucks are daylight animals. Anything eating your beans at midnight is not a groundhog.
UConn puts the daily range at 50 to 100 feet from the den (UConn Home and Garden Education Center). That is 15 to 30 metres, and it is the useful number. A fence around the plot matters far more than anything you do at the far end of the property.
Their year runs on a tight clock. They hibernate through winter and wake in late February or March (UConn). Mating follows within weeks. Young are weaned by late June or early July, then strike out for abandoned dens of their own. Fat storage for hibernation begins around the end of October.
My damage log matches that sequence. A burst of clipping in April and May. A second and heavier run in July, when the juveniles disperse. Very little after the first week of September.
Exclusion is the only method that keeps working
Every extension publication I have read says the same thing. Fencing is the only control that lasts (Penn State Extension). Poison, shooting, trapping and gassing all produce short-term relief.
Rutgers is blunt about why. Removing groundhogs from a property does not stop new animals moving in from the surrounding area, and a relocated one may come straight back (Rutgers NJAES).
The scale of the loss is worth stating plainly. Rutgers reports a single groundhog eating up to 1.5 pounds of vegetation a day (Rutgers). My worst spring, two adults took every bean seedling in a 12-foot row over two nights. They also stripped the tops off thirty carrot plants.
A fence is also the cheapest thing I have built. One removal quote came to more than the mesh and posts for the whole plot.
Building the buried L-footer
A vertical fence on its own fails. A woodchuck digs under it in a single morning.
The fix is a footer bent into an L. The standard advice is to bury the mesh 10 to 12 inches deep, then run at least 12 inches of it horizontally out and away from the plot (Rutgers). In metric that is a trench 25 to 30 cm deep carrying a shelf of at least 30 cm.
Cornell’s factsheet describes a shallower variant. Bury 10 to 12 inches, or bend the lower edge outward at an L and set it just 1 to 2 inches down (Cornell Cooperative Extension). Penn State specifies 12 inches down with the lower 6 inches angled out (Penn State).
I use the Rutgers figures because I dig in clay. A shallow shelf in heavy soil gets peeled up by frost. My current trench is 30 cm down and 30 cm out, backfilled and tamped in layers.
Height above ground is the other half of the job. Three to four feet is the range the extension services give, so roughly 90 cm to 1.2 m (Rutgers). I built mine to 1.2 m. I have never regretted the extra timber.
The top section is what defeats climbing. Rutgers suggests leaving the top 12 inches unattached to the posts so it wobbles when the animal grips it, or bending that section outward at 45 degrees (Rutgers). Penn State gives the same 15-inch out-bend (Penn State).
Mesh size and wire choice
Opening size is the specification most people skip. Woodchucks push through anything loose.
Michigan’s wildlife division sets the ceiling at 3 inches or smaller (Michigan Department of Natural Resources). The federal handbook prefers heavy poultry wire or woven mesh with 2-inch openings, about 5 cm (ICWDM).
I use half-inch galvanised hardware cloth on the beds and 2-inch welded wire on the perimeter. The welded wire is stiffer and cheaper by the metre. It also survives a string trimmer better than the light netting I started with.
One detail nobody mentions. Buy plain galvanised, not the green PVC-coated garden mesh. The coating cracks at the ground line within two seasons. The bare wire underneath then rusts through exactly where digging pressure is highest.
Electric strands for the climbers
This is the fact that surprises most gardeners. Woodchucks climb.
Michigan’s wildlife division lists tree climbing among their habits (Michigan DNR). Cornell calls them good climbers that can easily scale a wire fence unless you plan for it (Cornell).
I watched one go over a 4-foot woven fence at the corner post, using the diagonal brace as a ladder. It took about four seconds.
Two approaches work. Bend the top of the fence outward, so there is nothing to grip. Or run electrified wire. Kansas State recommends a fence at least 2 feet above ground with an electrified strand along the top (Kansas State).
Cornell places an electric wire 4 to 5 inches off the ground and the same distance outside the fence. Rutgers uses two strands, at roughly 4 inches and 8 inches above ground (Rutgers).
If you energise a fence, keep the vegetation down. Wet grass touching the wire shorts the whole run (Penn State). I strim a 30 cm strip along the outside every fortnight through summer. Dull work. It keeps the charger honest.
Raised beds, tree guards and pots
Raised beds need the same logic in miniature.
A bed framed in timber will not stop a woodchuck. The animal works the soil line at the base of the board. My first timber beds lost a whole crop of broccoli through a 6 cm gap where the board met the ground.
The fix is a hardware cloth skirt. Line the inside of the frame with half-inch mesh before you fill it. Fold the bottom 20 to 30 cm out flat under the soil. It costs one afternoon. It ends the problem for years.
Young trees need separate protection. Woodchucks gnaw bark near the trunk base. The incisor marks are large, roughly a quarter to three-eighths of an inch wide (UConn). That is the easy way to tell their damage from vole damage. Push a cylinder of hardware cloth 60 cm up the trunk and bury the bottom edge a few centimetres. Michigan’s wildlife division recommends metal skirting on trees for the same reason (Michigan DNR).
Repellents, and the honest record
I have spent money on repellents. I would rather you did not.
The core fact is regulatory. No woodchuck repellent is registered with the US Environmental Protection Agency, so nothing on the shelf has been reviewed for that use (Cornell). The federal handbook says the same in four words. None are registered (ICWDM).
Trials in Connecticut found that some commercial deer and rabbit repellents, and some insecticides sold with repellent claims, were generally ineffective against woodchuck feeding (Cornell).
The one exception in that literature is a predator scent. Bobcat urine sprayed on apple tree bases cut woodchuck gnawing by 98 per cent against untreated trees. A rope fence sprayed with it cut cabbage damage by 90 per cent (Cornell).
Read that result carefully. The predator scent helped most when paired with a physical barrier. Urine on its own is not a fence.
Habituation is fast. I ran a motion-activated sprinkler across a burrow path in 2021. It worked for nine days. On the tenth, the chuck walked through the spray to reach the beans.
Smoke bombs and the legal line
Here is where I have to be careful, because the law varies and I am not your regulator.
Woodchucks hold different legal status from state to state. Pennsylvania classifies them as a game animal, with a season running 1 July to 24 December and then 26 December to 30 June, and no bag limit (Pennsylvania Game Commission). Virginia treats them as a nuisance species with a continuous open season on private land (Virginia Cooperative Extension). New York lists them as unprotected, which still requires a hunting licence to take with a bow or firearm (Cornell).
The rules that catch people out are about disposal. Live-trapped woodchucks in New York must be euthanised humanely or released elsewhere on the landowner’s own property. Only licensed nuisance wildlife control operators may move an animal off site, and they charge for it (Cornell). Michigan will not authorise moving a live-trapped woodchuck off the property at all (Michigan DNR).
Relocation, the method most homeowners picture, is usually illegal without a permit or a licensed operator. Where a permit is required, your state wildlife agency issues it. No article can substitute for their current regulations.
Smoke bombs and burrow flooding deserve a plain answer. Fumigants are registered in some states and restricted in others. The extension fire warning matters more than the label. Never use a gas cartridge in a burrow under a home or shed, because of fire risk and gas accumulation (Penn State). Flooding is unreliable. It is also cruel.
One more caution. Woodchucks are a rabies vector in several states, and a captured animal can bite hard enough to injure through heavy gloves (Cornell). If you have a real conflict and want the animal gone, hire a licensed nuisance wildlife control operator and let them carry that risk (Connecticut DEEP).
Burrows under buildings and infrastructure
This should worry homeowners more than the tomatoes do.
Penn State warns that woodchuck burrowing can weaken building foundations (Penn State Extension). Missouri’s conservation department states that burrows can undermine concrete foundations, porches and driveways (Missouri Department of Conservation). The federal handbook notes the same risk for dikes and foundations (ICWDM).
Set that against the geometry. A tunnel two to five feet below grade, and 25 to 30 feet long horizontally, can pass directly beneath a shed pad, a patio or a septic drainfield (Kansas State). I have watched a shed corner drop 4 cm over two seasons. The soil under the pad had been carried out one mouthful at a time.
Septic systems are the quiet risk. Drainfield trenches sit in the same depth band. A tunnel through one gives effluent a clear path, and the failure shows up as a wet patch or an odour downhill.
Never seal a burrow until you are sure it is empty. Block it loosely with crumpled newspaper or loose soil, then check the next morning for fresh disturbance (Cornell). Late summer and early autumn are the safer windows for permanent closure, because young are not in the den and hibernation has not started.
Fill abandoned burrows with compacted gravel rather than loose topsoil. Loose fill settles into the void and leaves the same weakness behind. Where a tunnel runs under a footing, get the cavity assessed before you backfill.
Protecting the crops that pay
You cannot fence an entire property in most cases. You should not try.
Fence what is worth the money. Rutgers makes the same argument, and recommends exclusion around high-value areas such as vegetable gardens (Rutgers). I have taken the same approach.
My priority list, in order. Beans and peas first. Young fruit trees second. Salad beds near the house third. Then the tomatoes. Everything else gets a chance.
Courgettes and squash are usually left alone in my plot. Bush beans are wiped out. That difference decides where the fence goes.
Call your state wildlife agency before you move or kill anything. Ask whether a permit is needed. Ask what the local ordinances say. Then build the fence.
A perimeter fence around a 20 by 30 foot plot, with a buried L-footer and an electric top strand, costs about the same as three seasons of losing the crop. That is the honest calculation. You are not eliminating the animal. You are buying certainty for the plants you actually care about.
And the hollow thump under your knee in May? That is the ground telling you where to dig the trench.