What Is "Peak" in a Sourdough Starter?
Peak is the moment when a freshly-fed sourdough starter reaches maximum CO₂ production rate and maximum volume. The visual signal is straightforward: the starter has expanded to roughly 2× to 3× its post-feed volume, the top is domed or flat (just before collapsing), the surface and walls show dense bubble structure, and lifting the jar reveals a webby strand pattern indicating active gluten and gas networks.
Below the visual cues, peak is a specific point on the population growth curve of the dominant microorganisms — Saccharomyces cerevisiae (the yeast responsible for most CO₂) and lactic acid bacteria, primarily Lactobacillus sanfranciscensis (now reclassified as Fructilactobacillus sanfranciscensis) along with several supporting species. At peak, the yeast population has reached late exponential phase. Past peak, available simple sugars are depleted, pH drops below 4.0, and yeast metabolism slows sharply due to acid inhibition.
Knowing the peak time is operationally critical because the levain is most predictable when used at peak. Used too early, fermentation power is below maximum and dough rise is sluggish. Used too late, yeast viability drops and acidity dominates the flavor — sometimes desirable, often not.
Try the Peak Calculator →The Three Variables That Determine Peak Time
1. Inoculation Ratio
The ratio of ripe starter to fresh flour and water. A 1:1:1 ratio means equal weights — for example, 50g starter + 50g flour + 50g water. The starter contributes about 33% of the total mass and a corresponding proportion of microorganisms.
Higher inoculation (more starter relative to flour-and-water) gives a denser initial microbial population. With more cells competing for the same carbohydrate pool, exponential growth reaches its plateau faster. Lower inoculation requires more rounds of cell division before population density saturates the food supply.
| Ratio | Time to peak at 24°C | Use case |
|---|---|---|
| 1:1:1 | 4-6 hours | Same-day baking, daily feeds, mild flavor |
| 1:2:2 | 6-8 hours | Standard daily refresh |
| 1:3:3 | 7-10 hours | Convenient morning-to-evening cycle |
| 1:5:5 | 10-14 hours | Overnight builds, full development |
| 1:10:10 | 16-20 hours | Long autolyse, slow flavor builds |
2. Temperature
Temperature is the dominant variable. The microorganisms responsible for sourdough fermentation follow Arrhenius temperature kinetics over the 18-32°C range. The Q10 coefficient — the factor by which the rate changes per 10°C — is empirically about 2.0-2.5 for sourdough microflora (Gänzle 2014, Food Microbiology). This means the metabolic rate doubles to 2.5× when temperature rises 10°C.
| Temperature | Peak time (1:1:1 ratio) | Notes |
|---|---|---|
| 18°C / 64°F | 10-14 hours | Cool kitchen, winter |
| 21°C / 70°F | 7-9 hours | Cool room temperature |
| 24°C / 75°F | 5-6 hours | Standard "room temperature" |
| 27°C / 81°F | 3.5-4.5 hours | Warm kitchen, summer |
| 30°C / 86°F | 2.5-3.5 hours | Proofing box, warm summer |
Note that this is dough temperature, not air temperature. A starter in a kitchen at 22°C ambient will sit at 22°C (roughly), but a starter that just received cold water from the tap can be 5-7°C below ambient and will lag behind the predicted peak by a corresponding amount until thermal equilibrium is reached.
3. Flour Type
Different flours support fermentation at different rates due to two factors: amylase activity (which liberates fermentable sugars from starch) and microbial load (the bran fraction carries both yeasts and LAB).
- White bread flour: baseline. Lowest amylase, lowest microbial load, slowest peak.
- Whole wheat: 20-30% faster. Bran adds amylase enzymes plus bacterial diversity.
- Rye: 30-40% faster. Very high amylase activity, high pentosan content, robust LAB population.
- Spelt: similar to whole wheat but slightly slower; lower amylase than common wheat.
- High-extraction flour (T80, T110): intermediate between white and whole wheat.
Microbiology of the Peak
The two-organism model is a simplification. Real sourdough cultures contain diverse microbial populations — typically 1-3 dominant yeast species and 5-10 LAB species, plus transient organisms. But two species drive most observable behavior in a healthy stable starter:
Saccharomyces cerevisiae (and related S. exiguus, S. uvarum) is responsible for the majority of CO₂ production. Its population increases roughly exponentially during the early fermentation phase, peaks at 10⁷-10⁸ cells per gram of starter (De Vuyst & Neysens 2005), then declines as ethanol and acid accumulate.
Lactobacillus sanfranciscensis is the dominant LAB in mature sourdough cultures. It is heterofermentative — produces lactic acid, acetic acid, ethanol, and CO₂. Its population reaches 10⁸-10⁹ cells per gram, typically 10-100× higher than yeast cells. Peak LAB activity often occurs slightly later than peak yeast activity, which is why the late-peak phase has more pronounced sourness.
Peak time as observable to the baker — maximum volume — is driven by yeast CO₂. The calculator therefore tracks yeast doubling time as the primary mechanism.
The Use-By Window: From Peak to Fall
After peak, available glucose and maltose are exhausted. Yeast cannot maintain CO₂ production rate. Acidity continues to accumulate from LAB metabolism, dropping pH from ~5.5 (post-feed) to 4.0-3.8 (post-peak). The starter visually recedes — the dome flattens, then sinks. Bubble structure becomes coarser as small bubbles coalesce.
The "use-by" window — the period during which the starter is still useful for leavening — extends roughly 2-4 hours past peak at room temperature. Beyond that, yeast viability drops fast and the starter becomes acid-dominant.
The calculator outputs both the predicted peak time and a usable window. For most bakers, the optimal window is the last 20-30% of the rise (still climbing, near peak) through the first 60-90 minutes after peak.
Open the Peak Calculator →FAQ
Why doesn't my starter peak at the predicted time? Three common causes: (1) temperature in the jar differs from kitchen ambient — measure with a probe thermometer; (2) starter activity has decayed due to fridge storage longer than 5-7 days — refresh 1-2 cycles before relying on predictions; (3) flour was changed — the calculator's multiplier is approximate, not exact for every flour brand.
Can I refrigerate during the rise? Yes. Cold retardation at 4°C slows fermentation roughly 8-12× compared to 24°C. A starter that would peak in 5 hours at 24°C peaks in roughly 40-60 hours at 4°C. Use this to extend the use window for schedule flexibility.
What if my starter never reaches 2× volume? This indicates weak yeast population. Causes include insufficient feedings, contamination by mold, or temperature too low (sub-15°C) for yeast activity. Solution: 2-3 days of 1:1:1 feeds at 24-27°C to restore microbial balance.
Does hydration affect peak time? Yes, but moderately. A 100% hydration starter (equal weight flour and water) is the standard. Stiffer starters (50-60% hydration) ferment slower because lower water activity slightly reduces yeast metabolism. Wetter starters (125%+) ferment slightly faster but plateau lower in volume because gas escapes more easily.
Sources
1. Gänzle, M. G. (2014). Enzymatic and bacterial conversions during sourdough fermentation. Food Microbiology, 37, 2-10.
2. De Vuyst, L., & Neysens, P. (2005). The sourdough microflora: biodiversity and metabolic interactions. Trends in Food Science & Technology, 16(1-3), 43-56.
3. Brandt, M. J. (2007). Sourdough products for convenient use in baking. Food Microbiology, 24(2), 161-164.
4. Corsetti, A., & Settanni, L. (2007). Lactobacilli in sourdough fermentation. Food Research International, 40(5), 539-558.
5. De Vuyst, L., Van Kerrebroeck, S., Harth, H., et al. (2014). Microbial ecology of sourdough fermentations: diverse or uniform? Food Microbiology, 37, 11-29.