Methodology

Editorial rules

Each article opens with a direct answer to the question in its title. We give facts, figures and the names of legal acts only where we can confirm them in a source; when in doubt we write more generally instead of guessing. Sources (one to four) are listed under every article — we prefer primary documents: EUR-Lex, national gazettes, the websites of authorities and institutions, agency reports and publications with a DOI.

We do not publish invented quotations, interviews or customer opinions. All texts are authored by the FeedstockData Editorial Team.

Dates and updates

The date of an article is its publication date and, for events, also the date of the event. The content reflects the state of knowledge on the day of publication. When regulations or data have changed since, we add a highlighted “Update” note dated to the given month and change the modification date in the structured data. The archive thus shows history, and every article shows the current state of affairs.

BMP methodology

BMP (biochemical methane potential) is the amount of methane obtained from a unit mass of organic matter of a substrate in a batch test, converted to normal conditions (0 °C, 101.325 kPa). Results are given in m³ CH₄ per tonne of volatile solids (m³ CH₄/t VS) — numerically the same as the NmL CH₄/g VS (= L/kg VS) used in publications. We report methane, not biogas: the amount of methane does not depend on how much CO₂ stays dissolved in the liquid.

• Records. One record is one result from a publication for a described sample (e.g. a variety, harvest date or pretreatment variant). Every record carries a source identifier from the bibliography.
• Units. Results given in other units are converted, and the original value is kept in the data.
• Weighted mean. Each publication has a total weight of 1, split equally between its results for the given substrate, so a paper testing many varieties does not dominate a paper with a single sample. In inter-laboratory programmes the unit is the laboratory (total weight 1), and within it the weight of a test grows with the number of bottles. We also show the arithmetic mean, the standard deviation and the full range.
• Pretreatment. Results for samples after pretreatment (thermal, chemical, biological) are shown separately and are not included in the mean.
• Nature of the statistics. This is a descriptive compilation of the spread of published results, not a meta-analysis or a guarantee of plant yield.
• Curves. Where a publication gives the course of methane production over time, we draw the measurement. Where it gives only the final result, we draw a first-order kinetic model B(t) = B₀·(1 − e^(−kt)) and label it clearly: only the end point then comes from the publication.
• Interpretation. BMP is a laboratory potential. Yield in a plant depends on retention time, digester loading, temperature, mixing and feed composition.

The data can be downloaded on the open data page.

Methane in biogas — the indicator

For each substrate we show a semicircular indicator of the methane (CH₄) share of the biogas. The more methane, the higher the calorific value of the biogas: 1 m³ of biogas with 50 % CH₄ is about 5.0 kWh and with 65 % about 6.5 kWh (lower heating value of methane 9.97 kWh/m³).

• Measurement (needle). Where BMP studies in our database report the mean CH₄ share of the biogas from the test, we show their weighted mean (with the same weights as BMP) and the range. At present these are 60 studies for 25 substrates: the CROPGEN D4 database (the “methane content” column) and readings from the full texts of publications. Maximum values from single days and figures quoted from other works are left out.
• Estimate from the substrate type (dashed interval). Where there is no measurement, we estimate an interval from the approximate composition of the organic matter and the yields of compound groups according to the FNR guide (2013, Guía sobre el Biogás, the Spanish edition of “Leitfaden Biogas”, table 2.3, p. 30): protein — 700 l biogas/kg VS and 71 % CH₄, fats — 1250 l and 68 %, carbohydrates — 790 l and 50 %. The protein share is calculated from total nitrogen (N × 6.25) when the composition is known; fat and protein richness without composition data is taken from the process classification of the substrate. This is a stoichiometric approximation, not a study result.
• Zones (editorial convention). Red: below 52 % — biogas typical of materials dominated by sugars, starch and cellulose (theoretically 50 % CH₄). Orange: 52–60 % — mixed composition, most plants, silages and manures (e.g. according to FNR: maize silage about 53 %, cattle slurry about 56 %). Green: 60 % and more — a clear share of protein or fat (e.g. pig slurry about 61 %, poultry litter about 64 %, fats 68 %). The typical range for biogas is 50–75 % CH₄ (FNR 2013, table 2.6, p. 31).
• Note. In batch tests the CH₄ share tends to be higher than in a plant because part of the CO₂ stays dissolved in the liquid; a longer digestion time raises and a higher temperature lowers the methane share (FNR 2013, p. 30). The indicator compares substrates — it does not predict the gas composition of a particular digester.

Substrate rating for biogas plants

The “Rating for biogas plants” badge is an editorial rating under an explicit methodology, not a study result or a market valuation. It is meant to help answer quickly whether a material is worth accepting and on what terms. We do not take prices into account — we have no reliable sources for them.

Points are calculated on three axes (details for each substrate are in the expandable “Why?” on its page):

1. Potential per tonne of fresh matter = BMP × DM × VS [m³ CH₄/t]: ≥ 100 → +3, 50–100 → +2, 25–50 → +1, 10–25 → 0, below 10 → −1. With poor degradability of the organic matter (BMP < 200 m³ CH₄/t VS) at most +1, and with BMP < 100 at most 0. Where the dry matter in the data refers to a sample dried for the test and not to the material as delivered (e.g. algae, fresh plants), we do not calculate a rating. For four substrates without a composition in the BMP data (molasses, brewer's spent grain, pressed beet pulp, crude glycerine) we take mean values from the FNR guide (2013, table 4.9, p. 89).
2. Process stability — from the process classification of the substrate (tags, sources in the “Process properties” section): fats (LCFA, flotation), protein and nitrogen (ammonia), inhibitors, salt — 1 risk point each; acidification (rapid degradation 0.5, with an acidic pH 1), foaming and sulphur (H₂S) — 0.5 each. The largest risk counts in full and half of each further one. High alkalinity (manures) +1.5, a source of trace elements +0.5. Range of the axis: from −4 to +2.
3. Handling and costs — animal by-products category 2 (pressure sterilisation; does not apply to manures and gut content) −2, category 3 (pasteurisation 70 °C/60 min) −1 (dairy products −0.5), physical contaminants −1, sand and stones, fibres and floating layer, slow degradation — −0.5 each.

Total points: ≥ 2.5 — very good (also requires +3 for potential, risk no greater than −1 and costs no greater than −1), ≥ 1 — good (with a risk of −2 or more at most neutral), below 1 — neutral. Negative: process risk of −3 or more, or a potential of ≤ 0 points with a total ≤ −2.

The confidence of the rating (high / medium / low) depends on the number of independent BMP sources, the number of dry matter and volatile solids determinations and the confidence of the process data. The editorial team may correct a level manually (e.g. on the basis of plant data) — the correction is then marked on the substrate page together with the reason. The full description of the algorithm and the code: the OCENA.md file in the site data.

Corrections

We correct errors as quickly as possible after they are reported to contact@feedstockdata.com. Substantive corrections are described in an “Update” note on the article; minor language corrections are made without a note. In the BMP data, correcting a record changes the statistics of the substrate at the next publication of the site.