Agricultural Research 72(3) July - September 2012

RESEARCH

Bio-Economic and Qualitative Impact of Reduced Herbicide use in Direct Seeded Fine Rice Through Multipurpose Tree Water Extracts

Impacto Bioeconómico y Cualitativo del uso Reducido de Herbicidas en Arroz de Siembra Directa a través de Extractos Acuosos de Árboles Multipropósito

Abdul Khaliq¹*, Amar Matloob¹, and Yasir Riaz¹

¹University of Agriculture, Department of Agronomy, Faisalabad 38040, Pakistan. "Corresponding author (khaliquaf@gmail.com).

Weed control program should be environmentally benign and cost-effective so that reduced herbicide use can help meet these goals. Field trials were conducted to assess the bio-economic and qualitative impact of reduced doses (25 and 50% of label dose) of a postemergence pyrimidinyloxybenzoic acid herbicide (bispyribac-sodium) applied alone or in combination with multipurpose tree (eucalyptus [Eucalyptus camaldulensis Dehnh.], mango [Mangifera indica L.], and mulberry [Morus alba L.]) water extracts in direct seeded rice (Oryza sativa L.) fields. The label dose of bispyribac-sodium and penoxsulam along with weed control were included for comparison. Tank mixing of multipurpose tree water extracts with reduced herbicide doses accounted for > 55% suppression in weed density and > 75% in dry weight; they were quite higher than those recorded for the same herbicide doses used alone. A combination of these extracts with 50% reduced dose of bispyribac-sodium improved rice yield and quality attributes similar to the label dose of this herbicide. Despite the higher net benefits associated with label herbicide dose, the maximum marginal rate of return was achieved with a tank mix of 25% label herbicide dose with multipurpose tree water extracts.

Key words: Reduced dose, weed suppression, yield, kernel quality, Oryza sativa.

Los programas de control de malezas deben ser respetuosos del medio ambiente y rentables, y la reducción del uso de herbicidas puede ayudar a cumplir estas metas. Los ensayos de campo se llevaron a cabo para evaluar el impacto bioeconómico y cualitativo de dosis reducidas (25 y 50% de la dosis etiqueta) de un herbicida de post-emergencia pirimidiniloxi benzoico (bispiribac-sodio) aplicado solo o en combinación con extractos acuosos de árboles multipropósito [eucalipto (Eucalyptus camaldulensis Dehnh.), mango (Mangifera indica L.), morera (Morus alba L.)] en campos de arroz (Oryza sativa L.) de siembra directa. Se incluyó dosis recomendada en la etiqueta de bispiribac-sodio y penoxsulam y un control de malezas para comparación. Mezcla de estanque de los extractos acuosos de árboles multipropósito con dosis de herbicida reducida causó supresión de > 55% en densidad de malezas y > 75% en peso seco, lo que fue superior cuando estas dosis de herbicida se utilizaron solas. La combinación de estos extractos con un 50% de reducción de la dosis de bispiribac-sodio registró granos por panícula, peso 1000 granos y rendimiento de grano similar a la dosis de etiqueta del herbicida superiores a la aplicación de dosis reducida de este herbicida. Este tratamiento también mejoró los atributos de calidad del grano sobre el control similar a la dosis de etiqueta del herbicida. A pesar de los mayores beneficios netos de la dosis de etiqueta del herbicida, la tasa máxima marginal de retorno se obtuvo con mezcla de estanque de 25% de la dosis herbicida sugerida con extractos acuosos vegetales multipropósito.

Palabras clave: dosis reducida, supresión de malezas, rendimiento, calidad de grano, Oryza sativa.

Direct seeding of rice (Oryza sativa L.) has the potential to attain high water productivity and eliminate the edaphic conflict in Punjab's rice-wheat (Triticum aestivum L.) cropping system (Khaliq et al., 2011a). Heavy weed infestation and shifts in weed population are major constraints in the sustainability of direct seeded rice (DSR). An appropriate weed management strategy has always been a major focus and key element for successful DSR. Such a strategy is of the utmost significance to secure yields and the involved production costs, as well as to minimize any negative effect on product quality Good stand and high yield of dry direct seeded rice has a higher correlation with timely and effective weed control. Traditionally, weeds are controlled through cultural and/ or chemical methods. Manual weeding, though effective, is getting increasingly difficult due to labor scarcity, rising wages, and its dependence on weather conditions. Moreover, allowing weeds to reach a sufficient size to be pulled out and the presence of perennial weeds that fragment on pulling are other related concerns (Rao et al. , 2007). Thus, using herbicides is obligatory for weed management in direct seeded rice (Jaya Suria et al., 2011). Several studies (Gitsopoulos and Froud-Williams, 2004; Adigun et al., 2005; Singh et al., 2006; 2008; Mahajan et al., 2009) have concluded that chemical weed control is feasible since it is quick, easy, and economical.

Although efficient, the judicious use of herbicides has also been questioned since it is widely believed that they cause resistance in some previously susceptible weed species; serious environmental concerns also arise due to their high residual effects in the soil (Ahn et al., 2005; Khaliq et al., 2011b). An international survey conducted in 70 000 fields revealed that over 383 biotypes of 208 weed species (122 dicots and 86 monocots) have evolved to be resistant to herbicides (Heap, 2012). Resistance of rice weeds to a number of herbicides has been reported (Rao et al., 2007) since they only target a few molecular sites (Duke, 1990). In fact, 270 herbicides covering the global market represent only 17 modes of action with almost half of them acting as acetolactate synthase (ALS), PS II, and Protox inhibitors (Macías et al, 2007).

A lot of work has recently focused on plant-derived materials as an alternative to herbicides (Kuk et al., 2001). Allelopathy has emerged as a potential tool and as an alternative to herbicides for weed management (Tesio and Ferrero, 2010). Multipurpose tree species with allelopathic potential can be exploited for sustainable weed management (Singh et al., 2003). Suppressive allelopathic effects of eucalyptus (Verdeguer et al., 2009), mango (Javaid et al., 2010), and mulberry (Haq et al., 2010) against weeds have recently been documented. Different plant species contain allelochemicals that vary in type and concentration (Xuan et al., 2004). Furthermore, allelopathic activity is attributed to the combination and interaction of numerous allelochemicals instead of a single compound (Einhellig, 1996). The allelopathic potential of plant species can be exploited in many ways. Using water extracts of these species seems to be one possible tool. Although the use of allelopathic water extracts is economical and environmentally friendly, reduction in weed biomass is less than with herbicides and manual weeding. However, it may be possible to use these allelopathic water extracts in combination or with a reduced dose of herbicides to increase their efficacy and reduce the use of herbicides (Mushtaq et al, 2010). Allelopathic substances in combination with lower doses of herbicides can be an important step in recent times.

In our preliminary studies, water extracts of these trees were found to inhibit rice weed growth (Khaliq et al., 2012). However, the suppressive effects of tank mixing these extracts have not yet been documented against weeds of direct seeded rice fields. Moreover, there is little information regarding the influence of tank mixing these water extracts in combination with a lower herbicide dose on weeds of direct seeded rice. The following study was therefore designed to evaluate the influence of allelopathic plant water extracts alone and in combination with a reduced dose of herbicide on direct seeded rice and its associated weeds.

MATERIALS AND METHODS

Site description
The proposed study was conducted at the Agronomic Research Farm, University of Agriculture Faisalabad (31.25° N, 73.09° E, 184 m a.s.l.). The soil of the experimental site belongs to the Lyallpur soil series (Aridisol-fine-silty, mixed, hyperthermic Ustalfic, Haplargid in USDA classification and Haplic Yermosols in FAO classification). Saturated soil paste pH was 7.7 and total soluble salts were 0.79 dS m^-1. Organic matter, total N, available P and K were 0.74%, 0.055%, 7.1 mg kg^-1, and 183 mg kg^-1, respectively. Due to high evapotranspiration, Faisalabad features an arid climate with a mean annual rainfall of approximately 200 mm.

Experimentation
The experiment was laid out in a randomized complete block design (RCBD) with four replicates during the summers of 2009 and 2010. The net plot size was 6 m x 2.70 m. Seed of the popular rice cv. Super basmati was obtained from the Rice Research Institute, Kala Shah Kaku. In both years, the crop was sown in the first week of July with a single row hand drill at a seed rate of 75 kg ha^-1 and maintaining a distance of 22.5 cm between crop rows. A basal fertilizer dose of 125 kg N, 55 kg P₂O₅, and 40 kg K2O ha^-1 was applied. Fertilizers were urea (46% N), diammonium phosphate (18% N, 46% P₂O₅), and sulfate of potash (50% K2O). All P and K and half of N were applied at sowing. The remaining half of N was top dressed in two splits at tillering and panicle initiation. The first irrigation was applied 10 d after sowing (DAS), and the crop was subsequently irrigated as and when needed; 14 irrigations were applied until crop maturity, when it was manually harvested.

Extract preparation
Eucalyptus, mulberry, and mango leaves were collected from plants available in the surrounding area. These were stored and dried in the shade to avoid possible leaching by rain water. Leaves were chopped separately with an electric fodder cutter into 2-3 cm pieces and soaked in water for 24 h at room temperature (25 ± 5 °C) with a 1:10 ratio (w/v). The soaked material was filtered through 10 and 60-mesh sieves. The respective leachates were boiled at 100 °C to increase the concentration by 20 times for easy handling and application; they were stored for subsequent use. Previous studies revealed that boiling did not affect the nature, relative composition, and efficacy of leachates (Jamil et al., 2009). To exclude the notion of inhibitory effects owing to pH and osmotic potential of the plant water extracts, these two attributes were measured for each water extract. The pH and electrical conductivity (EC) of these extracts was determined with digital pH and conductivity meters (HI 9811-5, Hannah Instruments, Woonsocket, Rhode Island, USA). The osmotic potential (-MPa) of different extract concentrations was determined by multiplying EC (dS m^-1) by -0.036. In this experiment, pH and osmotic potential of the different extracts ranged between 6.2 and 6.8 and -0.05 and -0.09 MPa, respectively. The literature shows that these pH values and osmotic potential were unlikely to prevent plant growth and any growth inhibition was presumably thought to be due to inhibitory compounds present in such extracts (Chon et al., 2003).

Treatment application
Concentrated water extracts of each tree species at 18 L ha^-1 were applied alone and in combination with bispyribac-sodium [2,6-bis(4,6-dimethoxypyrimidin-2-yloxy) benzoic acid] at 15 and 7.5 g ai ha^-1 (50 and 25% of label dose, respectively). The label dose of bispyribac-sodium (30 g ai ha^-1) and penoxsulam (3-(2,2-difluoroethoxy)-N-(5,8-dimethoxy[1,2,4]triazolo[1,5-c]pyrimidin-2-yl)-a,a,a-trifluorotoluene-2-sulfonamide) at 15 g ai ha^-1 was used for comparison. Weed control was also maintained. Herbicides and their combinations with allelopathic plant water extracts were applied at 15 DAS the rice. Spraying was done with a Knapsack sprayer fitted with a flat fan nozzle and spray volume (300 L ha^-1) was determined by water calibration.

Data collection
Data of weed attributes (density and dry weight) were recorded for each experimental unit from two randomly selected quadrants (50 x 50 cm) at 15 and 30 d after treatment (DAT). Weeds were counted and clipped from the ground surface, dried in an oven at 70 °C for 48 h, and dry weight was recorded. Data of rice yield attributes were recorded from 15 randomly selected plants from each plot and the mean was calculated. Productive tillers (m^-2) were counted from two randomly selected sites from each plot and averaged. The crop was harvested, tied into bundles in their respective plots, and biological yield of the sun-dried samples was recorded. Each experimental plot was manually threshed and grain yield recorded (t ha^-1). A random sample of rice kernels was taken from the production of each plot. One thousand kernels were counted manually and weighed with an electric balance.

Kernel length and width were determined with a digital Vernier caliper. Abortive, chalky, opaque, and normal kernels were separated by positioning the panicle in front of a common electric bulb fitted with a flexible stand as a light source; this is expressed as the total percentage. Kernel N was determined by micro-Kjeldhal digestion followed by ammonia distillation; it was transformed into kernel protein by a multiplication factor of 5.95 (AOAC, 1990). Kernel amylose contents were determined as per Juliano (1971). The intensity of blue color was read out in a UV visible spectrophotometer meter (UV-4000, ORI, Germany) at 620 nm. The water absorption ratio (WAR) of kernels was calculated as suggested by Juliano et al. (1965):

Statistical and economic analyses
The data collected were subjected to Fisher's ANOVA technique with the "MSTATC" statistical package (Freed and Scott, 1986); the least significance difference test at P ≤ 0.05 probability was applied to compare the differences among treatment means. Year effect was found to be non-significant so the 2-yr mean will be presented and discussed in its respective section.

Economic and marginal analyses based on variable costs and prevailing herbicide and rice market prices were carried out. Gross income and net benefit (the value of yield benefit as a result of any practice less the cost of such a practice) were calculated as described by CIMMYT (1988). The marginal rate of return (MRR) was calculated as follows:

RESULTS AND DISCUSSION

Weed growth
Field observations revealed that weed flora of the experimental site was comprised of horse purslane (Trianthemaportulacastrum L.), jungle rice (Echinochloa colona [L.] Link), crowfoot grass (Dactyloctenium aegyptium L. Willd.), and barnyard grass (Echinochloa crus-galli [L.] P Beauv.) The local climate and experimental site soil moisture regimes promoted weed diversity as well as their growth primarily because the fields were not immersed under water. The weeds that are otherwise non-native to rice fields and those normally controlled by flooding in transplanted rice were abundant in the experimental plots. Rao et al. (2007) attributed greater weed pressure in dry seeded rice to the absence of flooding, dry tillage, and alternate wetting and drying during crop establishment.

Total weed density (both narrow and broad-leaved) was significantly influenced (P ≤ 0.05) by different treatments (Table 1); the difference was more pronounced for either one whether an herbicide dose was used alone or in combination with allelopathic plant water extracts. Tank mixing bispyribac-sodium at 15 and 7.5 g ai ha^-1 with allelopathic plant water extracts suppressed horse purslane density by 89 and 83% and 63 and 62% at 15 and 30 DAT, respectively (Table 1). The corresponding inhibition in barnyard grass and jungle rice density ranged from 43-49% and 51-62% and 57-63% and 52-60%, respectively, at these times. However, these treatments provided a small reduction in crow foot grass density. Bispyribac-sodium at 15 and 7.5 g ai ha^-1 tank mixed with allelopathic plant water extracts also recorded similar (P ≤ 0.05) total weed density. When applied alone at reduced rates, this herbicide recorded a lesser reduction in weed density in most cases. The label dose of bispyribac-sodium suppressed horse purslane (85-92%), barnyard grass (49-88%), and jungle rice (68-74%) density.

Table 1. Influence of multipurpose tree water extracts applied alone and in combination with reduced herbicide doses on weed density in direct seeded fine rice (mean of 2009 and 2010).



DAT: Days after treatment; ai: active ingredient. Means with different letters differ significantly at 5% probability level according to LSD test.

However, this herbicide poorly controlled (41-46%) crow foot grass. This confirms our previous findings (Khaliq et al. , 2011a) where bispyribac-sodium failed to achieve a desirable level of crow foot grass suppression. Maximum suppression in individual and total weed density at 30 DAT was recorded for penoxsulam when used at its label dose.

Weed dry weight was reduced to a significant level (P ≤ 0.05) by different weed control treatments (Table 2). Allelopathic plant water extracts alone recorded a weed dry weight reduction of 48 and 33% at 15 and 30 DAT, respectively. Tank mixing these water extracts with a reduced dose of bispyribac-sodium suppressed weed density by > 75% but was lower when the same doses were applied alone. Moreover, reduced herbicide doses applied either alone or as a tank mix with allelopathic plant water extracts were more effective against broad-leaf weed horse purslane (61-80%) than barnyard grass (9-38%), jungle rice (26-42%), and crow foot grass (39-62%) than the control. The label dose of bispyribac-sodium exhibited a dry weight reduction of > 80% at 15 and 30 DAT. Penoxsulam was higher with 91 and 87% reduction in weed dry weight, respectively, which can be attributed to the suppression of both narrow and broad-leaved weeds.

Table 2. Influence of multipurpose tree water extracts applied alone and in combination with reduced herbicide doses on weed dry weight in direct seeded fine rice (mean of 2009 and 2010).

DAT: Days after treatment; ai: active ingredient. Means with different letters differ significantly at 5% probability level according to LSD test.

Data (Tables 1 and 2) revealed a pronounced negative influence of the eucalyptus+mango+mulberry water extracts mix against density and dry weight of different weed species. Such an inhibition can be attributed to the suppressive effect of phytotoxins present in these extracts. Nevertheless, using allelopathic plant water extracts as 'nature's own herbicide' is eye-catching as an eco-friendly weed management approach; this is contrary to bioassays carried out under controlled environments where the suppression achieved under field conditions is far below the level that can be referred to as acceptable for economical crop production. Limited activity and selectivity are still major drawbacks when working with natural herbicides (Duke et al, 2001). Bhowmik and Inderjit (2003) proposed that interaction between allelochemicals and herbicides can be exploited for weed management because of their complementary action.

This can lead to reducing the use of herbicides as farm chemicals and simultaneously help enhance the efficacy of allelopathic agents (Khaliq et al, 2012c). The present experiments demonstrated a 75% reduction in weed dry weight with the tank mix of different allelopathic plant water extracts and 50% of the herbicide label dose. Suppression of weed germination and growth is believed to arise from the simultaneous and additive action of numerous allelochemicals (Einhellig, 1996). Duke and Lydon (1993) proposed that a synergistic interaction occurs between various allelochemicals since they can replace each other on the basis of their biological exchange rate and relative potency; thus, this adds to each one's herbicidal potential when mixed (Gerig and Blum, 1991). Khanh et al. (2005) further reported that the magnitude of suppression in the allelopathic interaction is directly proportional to the dose of the allelopathic product.

Both bispyribac-sodium and penoxsulam suppressed weeds at their label dose. Bispyribac-sodium acts as an acetolactate synthase (ALS) inhibitor that retards synthesis of branch chain amino acids, such as leucine, isoleucine, and valine (Darren and Stephen, 2006). Penoxsulam retards root growth by reducing photosynthate movement from leaves to root by interfering with an ALS enzyme (Devine, 1989). Effectiveness of these herbicides against rice weeds has been reported elsewhere (Mahajan et al, 2009). Our results also demonstrated that the magnitude of weed suppression is proportional to the herbicide dose whether it was applied alone or in combination with allelopathic plant water extracts, which was altered by the type of weed flora present. Broad-leaved weeds were susceptible to even the lowest herbicide dose so their density and dry weight were reduced to a greater extent than grasses when compared with the control. However, for grassy weeds, ≤ 75% of the label dose was required to keep weed count within an acceptable limit. Other authors (Iqbal and Cheema, 2007; Mushtaq et al, 2010) reported similar findings while working with tank mixes of a reduced herbicide dose with allelopathic plant water extracts in some other field crops under similar environments. The reduction in density and dry weight was presumably due to the decline in density of horse purslane, a dominant broad-leaved weed of the experimental site. Control plots were initially taken over by this weed, and after it completed its life cycle in approximately one month it showed senescence and a decline in density. Such a decline can also be in part because of interspecific competition among weeds at later stages.

Rice yield and yield components
Weed management treatments significantly promoted rice grain yield components as compared to the control (Table 3). The highest number of productive tillers (345 m^-2) was recorded when penoxsulam was applied, while the lowest (122 m^-2) was found in the control plots. Tank mixing multipurpose tree water extracts with 50% reduced dose of bispyribac-sodium recorded a similar (P ≤ 0.05) number of kernels per panicle, 1000-kernel weight, and grain yield to the results recorded when applying the label dose of this herbicide. Tank mixing allelopathic plant water extracts with a reduced (25 and 50%) dose of bispyribac-sodium achieved a significantly higher grain yield as compared to the sole application of these herbicide doses. However, the highest rice grain yield (2.80 t ha^-1) was achieved with penoxsulam.

Table 3. Influence of multipurpose tree water extracts applied alone and in combination with reduced herbicide doses on yield and yield components of direct seeded fine rice (mean of 2009 and 2010).