The weather has already happened to you by the time the forecast arrives. A rain gauge tells you what actually fell on your patch of ground, which is often quite different from what fell three miles away. Building the instruments is a weekend's work, mostly from things already in the recycling bin, and the payoff is a small daily habit that makes you look at the sky properly. Here is how to make three of the classics, get them reading sensibly, and keep a log worth looking back on.

Your basic kit

Nothing here is specialist. Gather it in one place before you start, because a build that stalls halfway tends to sit on the windowsill for a fortnight.

• A sturdy plastic bottle, two litres or so, with a flat base
• A funnel, or the top third of a second bottle cut off cleanly
• Four matching plastic or paper cups
• Two straight sticks, skewers or thin canes, about 30 cm long
• A knitting needle or thin bolt for a pivot
• Stiff card or foam board and a plastic lid for the vane
• A bead, a small bottle cap and a short length of fine tubing for bearings
• Waterproof tape, cable ties and strong glue
• A measuring cylinder or a large syringe
• A ruler, a waterproof marker and a compass

A rain gauge that reads in millimetres

The most accurate homemade gauge is also the simplest: a straight-sided container with vertical walls. Water collected in it sits at exactly the depth of the rain that fell, so no arithmetic is needed. Stand it level, mark a scale up the side, or just dip a ruler in each morning and read from the base.

If you would rather use a bottle, cut the top third off, invert it into the base to make a funnel, and tape the joint. It works well, but a funnel concentrates the rain, so you have to do a little maths to convert millilitres into millimetres.

Calibrating a funnel gauge

Measure the internal diameter of the funnel in centimetres and halve it to get the radius. The area is π × r². One millimetre of rain over that area gives a volume in millilitres equal to the area in square centimetres divided by ten. A 10 cm funnel has an area of about 78.5 cm², so 1 mm of rain produces roughly 7.9 ml in the bottle. Pour your catch into a measuring cylinder and divide by 7.9 for the total in millimetres.

A cup anemometer

Wind speed is measured by counting how fast something spins. Cross two 30 cm sticks into an X, lash or glue the centre, and fix a cup to each of the four ends. The crucial detail is that every cup's open mouth must face the same way around the circle. That asymmetry is what makes it turn: the wind pushes harder into the closed backs than the open fronts.

Push a knitting needle through the centre as an axle. It needs to spin in something smooth, so thread it through a bead glued to the top of your post, or through a short piece of fine tubing set into a bottle cap. Add a dab of paint to one cup so you can watch it go round.

Counting and calibrating

In a steady wind, count how many times the painted cup passes you in 30 seconds, then divide by 30 for revolutions per second. Repeat twice and average, because a single count is noisy.

There is no formula you can trust for a homemade instrument, so calibrate by comparison. Take your count, then look up the wind speed reported for your area by the Met Office or a nearby airport on the same afternoon. Do this on three or four days of differing wind, plot revolutions per second against speed, and draw a straight line through the points. That line becomes your conversion. Cup anemometers stall in very light air, so treat anything under a gentle breeze as "light" rather than chasing a number.

A wind vane that points the right way

Wind direction is named after where the wind comes from, so a westerly blows in from the west. Your vane's arrow should therefore point into the wind, with the tail pushed downwind.

Cut an arrow from stiff card or foam board: a small triangle at the front, a much larger tail fin at the back. The tail must have more surface area than the head or the vane will simply spin. Glue the arrow along a drinking straw or a light cane, then find its balance point by resting it on your finger. The pivot goes there, which is closer to the tail than the middle. Set the pivot into a base that lets it turn in the faintest breeze, and fix a card disc underneath marked with N, E, S and W.

Line the N mark up with true north using a compass, then check it again in a week. Vanes get knocked, and a station that is quietly pointing 20 degrees off will mislead you for months.

Siting and checking your instruments

Location decides whether your readings mean anything. Rain gauges are fussy about shelter: keep an obstacle at least twice its height away from the gauge, and further if you can. A gauge tucked against a fence under-reads in windy weather because the fence disturbs the airflow. Mount it upright with the rim about 30 cm above the ground, check it with a spirit level, and empty it after every reading so a part-full gauge cannot freeze and crack in winter.

Wind instruments want height. Ten metres is the meteorological standard and unrealistic in a garden; two or three metres above a shed roof is a reasonable compromise. Anything lower is slowed and steered by buildings, hedges and fences.

A thermometer belongs in shade and moving air, never in direct sun and not tight against a wall that radiates heat. A white box with slatted or louvred sides, or a well-ventilated white plant pot, gives a far better reading than a thermometer on a sunny windowsill. Once a month, check the anemometer for nesting birds and the vane for spider webs, both of which seize a bearing very effectively.

Keeping a daily record

A log is only useful if the readings are taken at the same time each day. Pick a time you will actually keep, and write the entry even when nothing happened. "No rain" is a data point; a blank line is nothing at all.

1. Date, and the exact time of the reading
2. Rainfall since the last reading, in millimetres
3. Wind direction as a compass point
4. Rotations counted, and the number of seconds you counted for
5. Anything unusual: hail, sleet, a sudden squall, the day the vane stuck

A notebook works, but a simple spreadsheet lets you total the month and draw a graph in seconds. After a few weeks, patterns start to appear: which direction brings your rain, how much more falls on the west side of town, how often the wind drops completely at dusk. Compare your monthly rainfall with the nearest official station, but do not read too much into a single month. Rainfall is genuinely patchy, and a difference of a few millimetres is not a mistake.

Getting into the habit

Start with one instrument and one reading a day. Add the anemometer once the rain gauge has become automatic, because two readings are easy and five are a chore. Write your entries in ink, including the ones that look wrong; a suspicious reading you kept is more useful than a tidy record you adjusted. And take the log outside occasionally, hold it against the sky, and see whether the numbers match what your eyes are telling you.

Photo: SatyaPrem / Pixabay