Culture simulator

The model follows a culture one day at a time. Eggs hatch, young animals grow up, adults lay eggs and die. Temperature changes how fast each step goes. Crowding, food, wetness and mould change how many eggs become animals and how many animals die.

Several of the figures are not known well. So the page runs the model 64 times, with those figures set to different points inside their ranges. The line is the middle run. The dark band holds the middle half of the runs and the pale band the middle 80%. Ask the same question twice and you get the same answer.

What is measured and what is assumed

What it cannot tell you

• It does not know your room, your substrate or your animals. Two tubs set up the same way can differ by a factor of three.
• Only one species, Folsomia candida, has a measured life history. For every other species the model stretches that pattern by a seller growth rating.
• It leaves out standing water, predators, escapes, contamination and the health of the founding animals.
• Nobody has compared its forecasts with counts from real cultures. The first honest test is a keeper who counts a culture and shares the numbers.

The claims behind it

1. Model assumption The simulator takes its breeding and survival pattern from one laboratory study of Folsomia candida at 21 °C70 °F. A female lays 1,011 eggs in her life. Her first batch comes 17.7 days after she hatches, and she lives about 118 days after that. An egg takes 8.5 days to hatch, inside the 7 to 10 days the Checklist gives. Every other species reuses this pattern, stretched or shrunk by its seller growth rating. Unknown 1, 2 Open question How fast do the other hobby species breed and live?
2. Model assumption A species with no measured life history takes its times from its seller growth rating. Every duration is multiplied by 0.85 for very fast, 1 for fast, 1.3 for moderate, 1.8 for slow and 1.4 for unrated. Egg output is multiplied by 1.4, 1, 0.6, 0.35 and 0.6 in the same order. These are choices made for this site and are not measurements. Unknown Open question How fast do the other hobby species breed and live?
3. Model assumption Of the hatchlings that would live in a laboratory, a keeper's culture turns one in five into animals that breed. The figure is chosen so that a starter portion becomes a working culture in about a month, as sellers say, and so that a healthy culture with room to grow roughly doubles every two weeks once it is under way. No one has logged real cultures to check it. Unknown 3, 4 Open question How fast do keepers' cultures really grow?
4. Model assumption A full culture holds 8 animals for every square centimetre of substrate, counting eggs and young, for a species 1.5 mm long. Smaller species pack denser and larger ones looser, by the square of the length ratio and within a fifth to three times. A rated substrate then scales it, from 1 for excellent down to 0.3 for poor, and 0.7 when the record has no rating. Counting a full culture yourself replaces this. Unknown Open question At what density does each culture stop growing?
5. Model assumption Each wetting brings the substrate back to damp, 0.85 on the model's scale from 0 to 1. It then dries by 0.05 a day in a tub with medium ventilation at 21 °C70 °F, half as fast with low ventilation and 2.2 times as fast with high, and 4% faster for each degree above that. Egg output is best inside a species' wetness band and falls toward its edges. Animals die at up to a fifth a day when the substrate is below the band. Standing water and over-wetting are not modelled. Unknown Open question How fast do culture tubs dry, and how wet is too wet?
6. Model assumption Animals eat 0.03 mg of dry yeast a day each. A feed holds 0.5, 1.2, 2.2 or 4 times what they eat before the next one, for light, matched, generous and heavy. Leftover food spoils by 4 to 30% a day, faster when warm and wet. When more than five days of food sit uneaten, mould load grows. Above 0.3 it kills animals, and at full load it cuts egg output by half. When more than three days of food sit uneaten, mites are counted as a risk, at 15% of that risk if the food was heated or frozen first. Unknown 3 Open question How fast does leftover food turn into mould and mites, and what harm do they do?
7. Model assumption Between the temperatures a species was measured at, development, egg output and life span follow straight lines, in logs for the rates. Beyond the warmest measured temperature, development and egg output fall to nothing over 4 degrees. Above the warmest temperature a species tolerates, animals die faster each day, up to a fifth of them at 4 degrees over. Cold slows animals and never kills them in the model. A species with no measured temperatures is assumed to do best between 20–26 °C68–79 °F. Unknown Open question What are the real temperature limits of each hobby species?
8. Model assumption Each forecast is 64 runs of one model. In each run the uncertain figures sit at different points inside ranges chosen for this site. They are culture yield, capacity, speed of development, egg output, drying, appetite and mould, and the ranges are wider for a species that was not measured. The outer band is the middle 80% of runs and the inner band the middle 50%. A wide band means a figure is poorly known, not that the culture is random. Unknown Open question How wide should the forecast bands be, and when does a risk count?
9. Model assumption A risk is marked watch when at least 15% of runs cross its line, and high when at least half do. A run is dry when a tenth of its days sit below what the species tolerates. It is mouldy when the mould load stays above 0.35 for a week. It is mite-prone when mite exposure averages 0.25 or more. It crashes when animals fall 40% or more in three weeks from a peak above a fifth of a full culture, outside weeks with a harvest. Unknown Open question How wide should the forecast bands be, and when does a risk count?
10. At 21 °C70 °F, females lived 136 days on average, with a range of 6 to 198 days. They laid their first eggs after about 18 days (16 to 25) and laid about 1,000 eggs in a lifetime, spread over about 13 egg-laying events.At 21 °C70 °F (Michigan stock, parthenogenetic females on plaster and charcoal with brewer's yeast), mean life span was 136.2 days (6 to 198), mean first oviposition was 17.7 days after hatching (16 to 25), mean fecundity was 1,011 eggs per female and mean ovipositions per female was 13. Peer-reviewed 1
11. Cooler was better in the classic laboratory study. Females lived about 241 days at 15 °C59 °F, 136 days at 21 °C70 °F and 72 days at 26 °C79 °F, and laid about 1,344, 1,011 and 130 eggs in a lifetime. The authors thought 26 °C79 °F may be close to the upper limit. Peer-reviewed 1
12. Females took 27.5 days from hatching to their first eggs at 15 °C59 °F, 17.7 days at 21 °C70 °F and 20.6 days at 26 °C79 °F. Development was slowest in the cool and a little slower in the warm than at 21 °C70 °F. Peer-reviewed 1
13. Averaged over the egg batches in the study, 94% of eggs hatched at 15 °C59 °F, 90% at 21 °C70 °F and 72% at 26 °C79 °F. The first batch hatched worst at every temperature. At 26 °C79 °F hatching fell with age, to 54% by the eighth batch. Peer-reviewed 1