Open Field Protocol

Rain Drainage and Time to Mold

A feeder that holds water turns a good bag of seed into a disease source. This protocol soaks two or more feeders side by side with a measured volume of water and times how long it takes the seed inside each one to clump, sour and mold.

The question this answers#

Feeders are sold as having "good drainage" and nobody says what that means. This protocol turns it into two numbers: the fraction of applied water leaving the drain holes in ten minutes, and the hours from a soaking until the seed reaches a defined clumping or mold score.

Wet seed is not a tidiness problem. Damp seed in a warm feeder grows the molds behind aspergillosis and gives salmonella a place to persist, and the dead zone below the lowest port, described in how to clean a feeder, is where it starts. If a feeder takes three days to dry and it rains every second day, that space never dries out.

Why the existing evidence is thin#

There is no standard wetting test for bird feeders. Manufacturers claim drainage without publishing a hole count, a hole diameter or a drying time, and reviews report "it stays dry" from a single yard in a single season.

The bigger problem is that time to mold depends far more on temperature and humidity than on the feeder. The same tube can spoil in a humid August week and hold for many days in a dry October. An hours figure with no weather beside it cannot be compared with anyone else's. Add unrecorded seed type, unrecorded fill depth and unstated scoring, and nothing is left to pool.

What you need#

• Two or more feeders, and a mount holding them side by side at the same height.
• A pump sprayer with a fine cone nozzle, and a jug marked in ml and fl oz.
• A rain gauge marked in inches and mm.
• A kitchen scale accurate to 1 g (0.04 oz).
• A shallow tray and graduated container to catch runoff from the drain holes.
• A long-handled spoon or wire probe for sampling at depth, and bags for weighed samples.
• A thermometer and hygrometer, or the nearest weather station's readings.
• A flashlight, gloves, and a fitted N95 mask for inspections that disturb moldy seed.
• Calipers or a drill-bit gauge for drain hole diameter.

How to set it up#

Mount the feeders on the same rail, about 24 in (60 cm) apart, at the same height and exposure to sun and wind. That side-by-side arrangement is the whole design: it makes weather a shared condition rather than a variable to model.

Record each feeder's fixed properties once: type, material, port count, roof overhang in inches and cm, drain hole count, and hole diameter in mm and inches. Fill from the same bag to the same mass, not the same visual level, since a wide hopper and a narrow tube look equally full at very different masses. A fill of 400 g (14 oz) suits small feeders. Mark the seed depth.

The procedure#

1. Weigh the seed in, record the start mass, and probe all three depths for a baseline.
2. Wet every feeder in a run by the same method. Controlled: apply 300 ml (10 fl oz) per feeder from the sprayer on a fine cone, held 18 in (45 cm) above and swept evenly for 60 seconds. Natural: let real rain do it, log the gauge total in inches and mm, and start the clock when the rain stops.
3. For the controlled method, hold the tray under the feeder and measure runoff leaving the drain holes in the first 10 minutes. Divide by the volume applied for the drain fraction.
4. Inspect at 1, 6, 24, 48, 72 and 96 hours after wetting, on schedule even when nothing seems to be happening.
5. At each inspection, probe three fixed depths: the surface, 1 in (2.5 cm) down, and the base plate dead zone below the lowest port. Score clumping and mold separately at each depth, and wipe the probe dry between feeders so you do not carry spoiling seed from one into the next.
6. For a moisture number rather than a score, take a 20 g (0.7 oz) sample from the base depth, weigh it, dry it at 60 C (140 F) for two hours, and reweigh. Moisture percent is wet mass minus dry mass, over wet mass, times 100.
7. Record temperature and relative humidity at every inspection, plus any rain during the run.
8. Stop any feeder that reaches mold score 2. Record the hour, then take it down.
9. Weigh the remaining seed at the end, to know how much birds took.
10. Repeat for 5 wetting events per design, swapping rail positions each time.

What you record#

One row per feeder per inspection: a 96 hour run on two feeders gives twelve rows plus baselines.

• Feeder: an id reused across runs, type, material, port count, roof overhang, drain hole count and diameter.
• Wetting: method, volume applied, rain gauge total, start time, and drain fraction from the 10 minute catch.
• Seed: type, lot, start mass, marked depth, end mass.
• Inspection: hours since wetting, clump and mold scores at each of the three depths, and moisture percent if measured.
• Weather: temperature, relative humidity, rain during the run, sun exposure.
• Outcome: hours to clump score 2, to mold score 1 and to mold score 2, and the action taken.

Controls, and the ways this goes wrong#

Position effects. One end of a rail gets more afternoon sun or wind, which alone can shift drying time by hours. Swap positions every run and check the difference survives the swap.

Fill depth and seed type. A deep column dries far more slowly than a shallow one, so weigh the fill and keep the mass identical across designs even when the visual level differs. Hulled hearts hold moisture and spoil faster than seed in the shell, and nyjer packs tight and drains poorly, so use one seed type throughout and say which.

Birds emptying the feeder. Consumption changes the depth you probe. Weigh at start and end, and treat a run that lost more than a quarter of its seed as compromised for depth measurements.

Bird welfare. Use small fill masses, keep a clean feeder available at all times, and never treat this test as a reason to leave spoiled food in front of birds. If a local disease report lands mid-run, stop and follow a sick bird at my feeder.

How many runs before the number means anything#

Five wetting events per design is the working minimum, and the useful statistic is the paired difference within each event, not each design's average. "Design A reached mold score 2 eleven hours before design B" is a clean comparison, because both got the same weather. "Design A averaged 44 hours" means nothing without the weather that produced it.

Drain fraction settles faster, since it depends on hole count, diameter and roof geometry rather than weather. Three controlled wettings usually suffice.

Handle censored runs honestly. A feeder that finishes 96 hours without reaching mold score 2 is recorded as "greater than 96", not as 96, and must not be averaged in as though it were. To say anything about temperature, spread runs across at least a 15 C (27 F) range, which means more than one season.

Submitting your results#

Download the datasheet for this protocol, keep the column names as they are, and fill one row per feeder per inspection. Leave a cell blank rather than guess, and record censored outcomes as a greater-than value.

Send the file through the Feeder Log with the make and model of each feeder, a photograph of the drain holes, and a photograph of the seed at each score you assigned. Those photographs let others calibrate their scoring against yours; without them a 0 to 4 scale is still an opinion.