Evidence hub: field research and trial data
Evidence hub: field research and trial data
Key measured reference points
This section previously headlined an annual kilograms-of-nitrogen-per-hectare band for legume cover crops as a class, and a percentage reduction in synthetic nitrogen fertiliser. Neither could be traced to a primary source, and both have been withdrawn rather than replaced. We do not publish a fertiliser-replacement percentage: how much nitrogen a cover contributes depends on how much biomass it actually grows on your site, and applied mineral nitrogen suppresses fixation, so the two interact rather than simply substituting.
Species nitrogen data: what is actually measured
| Species | Nitrogen: the measurement and its source | Biomass (t DM/ha/yr) | Establishment (weeks) | Shade tolerance |
|---|---|---|---|---|
| Pueraria javanica (PJ) | ~250 kg N/ha/yr accumulated in aboveground biomass at 85-93% Ndfa, over four years in one rubber-intercrop trial in northeast Thailand (Clermont-Dauphin et al. 2016, Agriculture, Ecosystems & Environment 217:79-88). A measured upper reference point from a single site, not a design value. The same trial found the cover reduced the drought resilience of the young rubber on water-limited ground. | 8-15 | 8-12 | Moderate while the canopy is open; thins as it closes |
| Calopogonium mucunoides (CM) | ~65 kg N/ha accumulated in leaves, shoots and roots by a 3-month green manure crop in one experiment (PROSEA). That is accumulation in biomass, not measured fixation, and it is not an annual rate. No annual figure for CM exists in any source we can trace. | 5-10 | 6-8 | Moderate; first to be shaded out in a mixture |
| Centrosema pubescens (CP) | 120-270 kg N/ha/yr (PROSEA). An institutional compiled estimate assembled from the literature, not a field measurement, so it should not be planned against at the top of the range. | 5-12 | 10-14 | High: one of the two shade-persistent successors |
| Mucuna bracteata (MB) | 67-84% Ndfa by 15N isotope dilution under oil palm (Cheah, Zaharah & Aminuddin 2010, MPOB Oil Palm Bulletin 60). That is a proportion, not a quantity per hectare; no defensible absolute kg N/ha figure exists, so we publish none. | 10-20 | 10-16 | High: the shade-tolerant species, used under maturing oil palm (~13 t/ha DM at 60-70% shade, MPOB) |
| Calopogonium caeruleum (CC) | Not separately quantified for nitrogen fixation in any source we can trace. We publish no figure for it. | 3-7 | Slow; up to about 20 months to full cover | High: the other shade-persistent successor; productive under 0-25% sunlight |
This table previously carried an annual kilograms-per-hectare fixation range for each species. Those ranges could not be traced to primary sources and have been replaced with the measurement that actually exists in each case, with its method attached. It also ranked Mucuna bracteata below Pueraria javanica for shade tolerance; that was wrong, and has been corrected. MB is the shade-tolerant species of the group, which is why it is used under maturing oil palm, while PJ thins as the canopy closes.
Seeding rates
| Species | Pure stand | Component of a mixture |
|---|---|---|
| Pueraria javanica (PJ) | 4 to 6 kg/ha | 3 to 7.5 kg/ha |
| Calopogonium mucunoides (CM) | 4 to 6 kg/ha | 1 to 3 kg/ha |
| Centrosema pubescens (CP) | 3 to 4.5 kg/ha | 0.5 to 3 kg/ha |
| Calopogonium caeruleum (CC) | 3 to 4.5 kg/ha | 0.6 to 1.5 kg/ha |
| Mucuna bracteata (MB) | Nursery-raised and transplanted at about 320 seedlings/ha, roughly 85 to 100 g of seed per hectare. Never broadcast: a kg/ha rate for MB is wrong by a factor of roughly forty. | |
The best-cited published mixture is MPOB TT No. 501 (Information Series 588, June 2012), Table 1 treatment C2: PJ 4 : CP 3 : CC 1, total 8 kg/ha, with 5 g of Rhizobium (RRI strain) inoculant per kg of seed. Treatment C4, where Mucuna bracteata carries part of the cover, is PJ 3 + CP 1 = 4 kg/ha. Both are institutional best-management-practice recommendations from a peat trial rather than primary field measurements; no mineral-soil rate exists, and MPOB does not state whether the totals are per gross hectare or per interrow hectare actually sown. Published mixture totals run 5 to 15 kg/ha and cluster at 6 to 12, and every published mixture is PJ-dominant.
Weed suppression effectiveness
Research across multiple Southeast Asian plantation environments demonstrates that established legume cover crops reduce weed biomass substantially compared to unmanaged controls. In a 24-month Selangor trial, weed dry weight under cover was 97.3 to 99.9% lower than an unweeded control and weed density 94.8 to 99.7% lower (Samedani et al. 2015). Mucuna bracteata is particularly effective against Imperata cylindrica.
The mechanism is primarily light exclusion: a dense cover crop canopy limits weed seed germination and rhizome regrowth by cutting light penetration to the soil surface. One caveat worth carrying: in the same trial, weed densities at 9 to 12 months were higher in the legume plots before the canopy closed, so weeding still has to be budgeted for during establishment.
Soil health improvements
This is the weakest link in the case for cover crops, and we would rather flag it than oversell it. The most rigorous recent Malaysian trial, running 24 months under mature oil palm, found no significant effect on any soil physicochemical property measured, including pH, organic carbon, nitrogen, CEC, bulk density and aggregate stability (Mohd Noor et al. 2021). A 2025 North Sumatra study found no significant differences either.
The better-supported benefits are nitrogen contribution, weed suppression after canopy closure, and surface protection. Nitrogen fixation by cover crop root nodules is confirmed by 15N isotope dilution studies, and fixation is suppressed where soil nitrate is high.
Erosion control on slopes
On sloping terrain common in Malaysian and Indonesian plantation regions, established cover crops reduce runoff and soil loss substantially: inter-row legume cover in replanted rubber reduced runoff by 88% and soil loss by 98% against bare soil (Perron 2024). On the published Malaysian evidence, however, slope and frond management dominate: contour frond stacking alone cut runoff from 30.8% to 17.9%, and to 10.7% with silt pits (Afandi et al. 2017, Oil Palm Bulletin 75). A cover crop is one contributor alongside terracing, frond stacking and silt pits, not a replacement for them.
SoilBoost EA performance
Nutrient uptake enhancement
SoilBoost EA's humic and fulvic acids are used to support nutrient uptake efficiency in tropical crops. Field observations show improved root development, increased cation exchange capacity (CEC), and more efficient use of applied fertilizers. It is a soil conditioner: it is not a fertiliser substitute and not a disease treatment.
Application at 2-5 kg/ha as a soil drench or foliar spray complements existing fertilizer programs, with growers reporting improved plant vigor and fruit quality in durian, melon, mango, and other tropical fruit crops.
Field trial: SoilBoost EA on banana (FPA EUP No. 3227)
An independent, FPA-supervised efficacy trial (EUP No. 3227, registered to Kudzu Seeds Trading, Chemiseed's sister company) on tissue-cultured Lacatan banana, run by an FPA-accredited researcher at the PCA-Davao Research Center, found that SoilBoost EA improved seedling growth, nutrient uptake and soil quality, with the strongest results when combined with a balanced fertilizer program. All growth differences were statistically significant (p < 0.01).
Read the full SoilBoost EA banana trial: methods, results table, chart and citation →
References and further reading
Key sources cited on this page: Clermont-Dauphin, C. et al. (2016), Agriculture, Ecosystems & Environment 217:79-88; Cheah, S.S., Zaharah, A.R. & Aminuddin, H. (2010), MPOB Oil Palm Bulletin 60; PROSEA (Plant Resources of South-East Asia); MPOB TT No. 501 / Information Series 588 (June 2012); Samedani et al. (2015), International Journal of Agriculture & Biology 17(2); Mohd Noor et al. (2021), Pertanika JTAS 44(1); Afandi et al. (2017), Oil Palm Bulletin 75; Perron (2024). Specific data ranges reflect variability across different soil types, climatic conditions, and management practices, and are given with their method and scope attached.
We continuously update our evidence base as new research becomes available, and we withdraw figures we can no longer trace rather than repeating them. If you are conducting field trials with our products and would like to share your results, please contact our technical team.