A Quarter Less NPK, Same Maize Yield, in One Field for One Season

The honest answer first. A single field trial in Bengkulu, Indonesia found that hybrid maize on an acidic Ultisol yielded the same when 25 percent of the recommended mineral NPK dose was replaced by a fermented liquid fertilizer made from oil palm empty fruit bunches [11]. Replacing half the dose also held yield, but nitrogen uptake fell. Replacing three quarters cut yield. That is the whole of the result, and it comes from one site, one season, four replicate plots per treatment, with no cost analysis [11].

The cost question is not a side note. The paper states its own motivation plainly: hybrid maize on these soils depends on mineral NPK, and the cost of that fertilizer limits smallholder returns [11]. A substitution that works agronomically but costs more to produce, store and apply than the fertilizer it replaces has not solved the problem the study set out to solve. The abstract says the trial lacks a cost analysis [11]. So the central practical question is open.

What was actually tested

The design was a randomized complete block field experiment with four replications. The treatments were an unfertilized control, full NPK at 138 kg N, 54 kg P2O5 and 60 kg K2O per hectare, then 75, 50 and 25 percent of that dose each combined with 3,333 litres per hectare of the fermented liquid, plus the liquid alone [11].

Full NPK produced 7.92 tonnes of grain per hectare. The 25 percent replacement produced 8.15, and the 50 percent replacement produced 7.34. Neither differed from full NPK under Tukey's test at p < 0.05 [11]. The 25 percent figure is numerically higher than full NPK, but the study does not claim an increase, and neither should anyone reading it. The correct reading is that the difference was not detectable at this sample size.

At 75 percent replacement, yield dropped to 6.01 tonnes per hectare [11]. That is the boundary the trial found: somewhere between half and three quarters of the NPK dose, the liquid stops compensating.

Nitrogen uptake tells a slightly different story from yield. It matched full NPK at 25 percent replacement, but was 13 percent lower at 50 percent [11]. So the two endpoints do not agree on where the limit sits. Yield held at 50 percent replacement; nitrogen uptake did not. A plant can produce the same grain with less total nitrogen in its tissue, and that is what appears to have happened here, but the trial does not explain it.

Agronomic efficiency of nitrogen, the kilograms of grain produced per kilogram of N applied, rose from 30.6 to 37.9 at 25 percent replacement [11]. That is consistent with the yield result: less nitrogen applied, similar grain produced. Apparent nitrogen recovery ranged from 49.8 to 62.0 percent, but Tukey's test did not separate the treatments on this measure [11]. That is an inconclusive result, not a positive one.

The mechanism is not established

Fermented oil palm empty fruit bunch liquid is what it sounds like: the fibrous residue left after palm fruit is stripped for oil, fermented into a liquid that carries some plant-available nutrients and organic matter. The abstract describes the material and the rates but does not report its nutrient concentration [11]. That matters. Without knowing how much N, P and K the liquid itself supplied, the substitution cannot be checked on a nutrient-equivalent basis. The trial replaced a share of the NPK dose with a fixed volume of liquid, not with a matched quantity of nutrients.

One measured effect points somewhere interesting. Soil pH after harvest was higher than under full NPK once 50 percent or more of the NPK dose was replaced [11]. Acidic Ultisols constrain maize in part through acidity, and a shift in pH could plausibly affect nutrient availability independently of the nutrients the liquid carries. But the paper does not test that, and the abstract offers no mechanism. Treat the pH result as an observation, not an explanation.

Why anyone is asking

Oil palm empty fruit bunch is described in the paper as an abundant local waste in Bengkulu [11]. That is the appeal: a residue from one crop used to offset an input cost on another. The wider literature treats palm oil biomass as a feedstock for exactly this kind of conversion. An optimisation modelling study of palm oil biomass networks includes fertilizer among its target outputs alongside steam, electricity and transport fuel, and reports that circularity-optimised designs reduced total resource costs relative to linear models [10]. That is a simulation, not a field trial, and it does not test whether the fertilizer works on a crop. Reviews of agricultural waste valorization make the same general case, listing organic fertilizers among the products that can come from crop residues [5]. None of this substitutes for the missing cost analysis in the Bengkulu trial.

Where the evidence runs out

Almost everything a grower would want to know is missing.

The trial ran one season at one site [11]. Soil, rainfall and pest pressure in the second season will differ. A single season cannot show whether the pH shift persists, whether it accumulates, or whether it reverses.

There is no cost analysis [11]. The 3,333 litres per hectare rate is a substantial volume to ferment, store, transport and apply, and none of that is priced in the paper.

The abstract does not report the yield of the unfertilized control or of the liquid-alone treatment [11]. Without those, a reader cannot see how much of the yield came from the soil's own supply, or what the liquid does on its own. The claim that the liquid substituted for NPK rests on the comparison with full NPK, which is fair, but the picture is incomplete.

The nutrient concentration of the liquid is not given [11], so the substitution rate cannot be translated to other materials, other batches, or other sites.

And the result is specific to hybrid maize on an acidic Ultisol in Bengkulu [11]. It does not extend to other crops, other soil types, or other palm residues. The paper says as much: the finding comes from one season at one site and lacks a cost analysis [11].

What the study does show is narrow and real. On this soil, in this season, a quarter of the mineral nitrogen, phosphorus and potassium could be withheld and replaced with fermented palm bunch liquid without a detectable loss of grain or nitrogen uptake. That is a testable starting point. It is not yet a recommendation.