Abstract

Peat soils are a valuable agroecosystem because of their rarity in Africa (< 4%) and their high carbon sequestration capacity (3% C), which is nevertheless disturbed by farming. It is therefore necessary to explore regenerative agricultural practices as part of organic farming. An agronomic trial was set up on peaty soil in Songon (west of Abidjan) to test doses of aqueous Costus afer extracts on the growth and production of flood rice and soil properties. Five (5) doses (0, 250, 500, 750, and 1000 liters/ha) of fermented aqueous extracts (6 months) of Costus afer were applied in a randomized complete block design in three replications with 6 treatments (T0, T1, T2, T3, T4, T5) and the rice variety VA6 was transplanted at a spacing of 20 cm × 20 cm. Rice growth and yield parameters were collected. The results obtained showed a greater number of tillers (324/m²) and greater heights were noted for T1, contrasting with a relatively lower weed biomass for the higher doses (T4 and T5), with a significant effect (P<0.05) at different physiological stage of the rice. No significant difference was observed between yields, although T5 (7.58 t/ha) outperformed the control (T0) by 49%. However, a dose of 649 l.ha-1 is required for better rice yields.

1. Introduction

Mire or Peat soils ¹ are in considerable demand in agriculture due to their organic matter content, which favors chemical restitution, and their high moisture retention capacity ². Their agricultural development requires drainage, which leads to a partial loss of their carbon sequestration function, emitting 6% of anthropogenic CO₂ ³. It is becoming increasingly important to find an agricultural method capable of preserving the ecological balance of these peatlands. Crops adapted to flooded environments, such as rice, could meet this need. However, the impact of chemical fertilizers and other conventional agricultural products on biodiversity and soil functioning ⁴ supports the need for innovation in rice cultivation in peatlands.

The use of biofertilizers ⁵ could be one of the solutions in the context of regenerative agriculture ⁶. In particular, some plants have high concentrations of photosynthesis carbohydrates and high levels of chemical elements that can be assimilated by crops ⁷.

With this in mind, fertilization trials using the aqueous extract of Costus afer were conducted on upland rice in pots, revealing high concentrations of N, P, and K ⁸. The results showed no significant difference between the optimal dose of 500 ml and the use of conventional fertilizers, suggesting some efficacy of this extract. This work aims to confirm this result, which could provide lasting support for the ecological functions of peat soils, by testing several doses of fermented aqueous extracts of Costus afer in the field on rice crops to assess their effects on the growth and yield of irrigated rice compared to other conventional practices ⁹. Ultimately, the study will make it possible to offer rice farmers a low-cost alternative practice that preserves the ecological function of hydromorphic peat soils.

2. Materials and MEthods

The study area is located in the commune of Songon. This commune is bordered to the east by the commune of Yopougon, to the north by the commune of Anyama, to the west by the commune of Dabou, and to the south by the Ebrié lagoon. It covers an area of about 584 km2. The study area is located in a third-order lowland with geographical coordinates of 5°19'32''N and 4°10'17''W (Figure 1). Annual rainfall in the area varies from 1,500 to 2,500 mm, with 90 to 180 days of precipitation, with a minimum of 32 mm in February and August and a maximum of 457 mm in June. The monthly temperature range is 7°C in the dry season and 5°C in the rainy season ^{10 11 12}. The geological formations in the area are mainly Cretaceous sandstones of the terminal continental period. This parent rock has undergone extensive alteration to produce ferralitic soils ¹³, classified by ¹⁴ as acrisols and marginally as basal Petrosols on plateaus between 40 and 60 m in altitude. In swampy areas the soils are hystosols.

The plants used in this work are:

Costus afer: This is a large herbaceous perennial medicinal plant, unbranched, with a creeping rhizome. The organs used for the preparation of aqueous extracts of Costus are the stem and leaves. The plants used in this study are more than one (01) year old ¹⁵.

Rice VA6: This is a variety obtained from a segregating population in the 12^th generation of AfricaRice hybrid (AR 051H). It has a 90-day cycle and a yield of 4 t.ha^-1 in the field. It is a lowland variety.

The technique involved grinding the harvested Costus afer in a mortar. After grinding, the extract was diluted with 1 litre of tap water, then the aqueous solution was stirred manually by mixing the solid residues in the water. Using a clean cloth, the solution was filtered by wringing out the residue in the cloth used as a filter. This operation was repeated until a clear filtrate was obtained, i.e. fourteen (14) times for as many filtrates making up the aqueous extracts of Costus afer. A composite sample of the solutions was conditioned with 500ml of each of the 14 filtrates. The extracts obtained were then collected in canisters before being stored away from the sun under ambient conditions (28 - 30˚C) in order to be used after 4 months of fermentation.

The VA6 rice variety chosen was dried for 24 hours to lift dormancy before being soaked in water for another 24 hours. The seed was then wrung out and dried under cover for 24 hours to pre-germinate. The pre-germinated seeds were lightly buried in the soil under protective mulch. The nursery lasted 11 days and transplanting took place on the 12th day with a spacing of 20 cm × 20 cm between plants.

An area of 357.5 m² was cleared and divided into 18 microplots of 15 m² each, divided into three (03) blocks of 6 microplots. In a randomised complete block design (Figure 2), different doses of fermented aqueous extract were applied. Each treatment was repeated 03 (three) times. A distance of one (01) metre was left between the replicates. Within the same repetition, the microplots were separated by 0.5 metres. The treatments applied are described in Table 1.

Two types of parameters were determined: agro-morphological parameters and rice yield.

Plant height and number of tillers were determined at different stages of rice development: tillering, heading and maturity. The height (H) of the rice plants was measured from the soil surface to the tip of the highest tillers (the tip often being taken as the end of the longest tillers leaf) using a decameter. The number of tillers per 1 m2 (experimental unit) was counted to determine tillering density.

When the rice was ripe, it was harvested by hand with sickles over an area of 8 m², leaving 2 border rows. After drying and threshing, the rice grains were winnowed. The straw and the grains were dried in the sun and weighed separately using an electronic balance with a capacity of 100 kg for the straw and 7000 g for the grains. The moisture content of the rice grains was determined after drying in an oven at 70°C for 24 hours, and the grain yield was calculated with the standard moisture content of 14%. Grain yield (RDG) and straw yield (RDP), total dry matter (TDM), and harvest index (HI) were calculated using the following formulae:

IR= (RDG/ TDM) × 100 (1)

avec RDG = Grain yield et TDM = total dry matter;

TDM (t.ha^-1) = RDG + RDP (2)

avec RDP = straw yield ;

RDG (t.ha^-1) = (dry weight of grain (kg)/ 8(m²)) × (10000/1000) × ((100-H)/86) (3)

RDP (t.ha^-1) = (straw dry weight (kg) / 8(m²)) × (10000/1000) × ((100-H)/86) (4)

With H (p.c. humidity) = [M. initial – M. Final / M. initial] × 100 (5)

avec M= Mass of rice quantity taken from each microparcel.

The data obtained were entered using Excel software. The data on height (HAUT) and number of tillers (TAL) were subjected to an analysis of variance (ANOVA) with a threshold of α = 0.10 using SAS software version V 9 to test the effect of the treatments applied on the vegetative development of the rice plants. Classification of the means using the Newman and Keuls method was used to identify the treatments for which the levels of vegetative development were significantly different. Grain yield (RDG), straw yield (RDP) and harvest index (HI) data were also subjected to ANOVA to test the treatment effect. Classification of the means using the Newman and Keuls method was used to identify the treatments for which the agronomic performances were significantly different.

3. Results

Overall, the highest biomass was observed in T1 (57.89a) and T2 (56.44a). Treatment T4 (39.22a) recorded the lowest amount of biomass.

Treatment T3 (61.76a) recorded the highest value for plant height, while treatment T4 (55.70a) had the lowest height. The treatment had no significant effect on either of these two parameters, with the exception of the number of tillers recorded per square metre, which had a significant probability (P = 0.07) at α = 0.10.

The lowest values for the number of tillers were recorded for T3 (262) and T5 (248) respectively, compared with a higher value (309) for T1. However, the value observed for T1 is statistically different from the values observed for the other treatments.

Figure 3 shows the average values for the number of tillers at different physiological stages, depending on the treatment. Throughout the vegetative phase, there was a significant difference between tillering and bolting. The number of tillers increased steadily until the bolting phase, when T1 treatments had the highest number of tillers. The number of tillers continued to increase right up to heading for the treatments in contrast to the other treatments:

Table 3 shows the average weed biomass values in each treatment as a function of physiological stage. Overall, there was a decrease in weed biomass during the physiological development of the rice. There were significant effects of treatments T4 (P = 0.0067) and T5 (P = 0.0078) on weed biomass, with a significant difference between the mean values observed.

A correlation analysis was carried out between rice height and weed biomass on the one hand, and between this biomass and the number of tillers on the other. The results in Table 4 show significant correlations for treatments T4 and T5. The correlations were very strong negative respectively.

The rice grain and straw yields, as well as the total dry matter and harvest index recorded in Table 5, show that the highest grain (7.63 t.ha^-1) and straw (29.84 t.ha^-1) yields were observed for treatment T5. However, there was no significant treatment effect and no significant difference was observed between the mean values for any of the parameters studied. The response curve shows an optimum dose of 649 l.ha^-1.

The correlation test between grain yield and weed biomass as well as the dry weight of 1000 grains and weed biomass indicates that the treatments are not very significant (Table 6). However, the probability for T2 is significant (0.10) and the correlation between grain yield and weed biomass is strong (0.98). There was a weaker positive correlation (0.40) between weed biomass and the dry weight of 1000 grains.

4. Discussion

The soil studied has a lumpy structure with good porosity due to the presence of earthworms and microorganisms that produce organic molecules, thus contributing to soil cohesion ¹⁶. Although it is well known that organic matter improves soil structure, little is known about the effect of adding an aqueous plant extract. Macroscopic observation shows that fungi are dominant in the aqueous extract of Costus afer. The functions of decomposition of organic matter, physiological processes including hydric and mineral plant nutrients and biostimulant mechanisms are therefore dominant in fungi ¹⁷. This role is reinforced by the presence of bacteria in the aqueous solution of Costus afer, as they feed on carbon and nitrogen ¹⁸, which they return to the soil via their corpses, thus providing structural carbon (carbon saturation). This demonstrates the potential for regenerating peat soils using such a simple practice as the application of an aqueous extract of C. afer.

The results of this work also highlight a new biofertiliser practice that requires less technical skill on the part of farmers and few financial resources: the aqueous extract of Costus is odourless after six months' conservation and is applied in the water that overflows the soil after ploughing, particularly for rice growing. This practice also makes up for the shortcomings of organic farming, which suffers from a lack of raw material.

Although treatment T1 (166.6 l.ha-1) recorded the highest value for the number of tillers (306/m2), no significant difference was observed between the mean values for grain yield (RDG) and those for the other indicators (RDP, MST and IR) of rice production under the effect of the aqueous extract of Costus afer. However, rice grain yield varied from 5.12 (T0) to 7.63 t.ha^-1 (T5) with values equal to or close to 7.00 t.ha^-1 for the other doses. These yields are much higher than those recorded at national level (1.5 - 5.5 t.ha^-1) ¹⁹. The genetic potential of VA6 has therefore improved, probably due to better adaptation to the environment. In any case, since genetic potential represents only part of the production factors (30 - 40%), we should note here the high agricultural value of peat soils, as observed at Songon with a grain yield of 5.12 t.ha^-1 without any input to the soil. The response curve, which was not significant, predicts a grain yield of 5.6 t.ha^-1 without any input and an optimum dose of 649 l.ha^-1, which should be confirmed with further studies.

5. Conclusion

This study highlighted the importance of peatlands for lowland rice production, with grain yields well above the national average in Côte d'Ivoire, using a promising variety, VA6.

Differences in vegetative growth were observed, particularly in tillering, although this did not affect grain yield. With no significant difference, the grain yields were in favour of applying doses of Costus afer. A simple and inexpensive application. However, it is important to reinforce the results obtained by quantifying the carbon stored in the soil in order to assess its effectiveness in the soil's ecological function.

References