Spatial variability of the competitive effect of Barley (Hordeum vulgare L.) on Lolium rigidum L.
Variabilidad espacial del efecto competitivo de la cebada (Hordeum vulgare L.) en Lolium rigidum L.
Jordi Izquierdo1 * , César Fernández–Quintanilla 2
1 Departamento de Ingeniería Agrolimentaria y Biotecnología. Universitat Politécnica de Catalunya. Campus Baix Llobregat. Edifici D4. Av. Canal Olímpic s/n. 08860 Castelldefels, Catalunya, Spain, *Author for correspondence: (jordi.izquierdo@upc.edu).
2 Centro de Ciencias Medioambientales. CCMA–CSIC. C/ Serrano, 115. 28006 Madrid, Spain, (cesar@ccma.csic.es).
]]>Received: September, 2008.
Approved: October, 2009.
Abstract
Lolium rigidum is a major grass weed of winter cereals in the Mediterranean area, in spite of the continuous use of herbicides in these crops. New management approaches focus on the reduction of the seed banks by enhancing crop competitiveness and, consequently, minimizing weed seed rain. However, the spatial heterogeneity that exists within fields results in differences in the growth and the competitiveness of crops and weeds. In order to determine if the competitive interactions between barley and L. rigidum are site–specific biomass and seed production of this weed, growing in monoculture (plots with L. rigidum) and in mixed culture (plots with L. rigidum+barley), were studied at three sites (in upland, mid–slope and lowland positions) within barley fields. In each site were determined weed populations, and in soil separates, nutrient content, organic matter, slope and orientation were determined for each site. Crop presence significantly reduced weed biomass between 5 and 79 % and seeds per spike between 10 and 48 %, depending on the site. The competitive effect of the crop was greater in the more fertile sites (with higher N, P and organic matter content). In these sites, differences in plant biomass accumulation between the weed in monoculture and the weed in mixed culture started to be significant after stem elongation. Regardless the reduction in the number of seeds per spike observed in the most fertile sites, seed rain (measured as seeds m–2 ) could still be very important if weed density of the site is high. The differences in the competitive interactions between barley and L. rigidum observed within the fields suggest that adequate crop husbandry practices addressed site–specifically to enhance crop competitiveness can play an important role as a mechanism to reduce L. rigidum populations over the long term.
Keywords: Hordeum vulgare, crop competition, seed rain, site–specific, topography, weed biomass.
Resumen
Lolium rigidum es una de las principales malezas de los cereales de invierno en la región del Mediterráneo, pese al uso continuo de herbicidas en estos cultivos. Los nuevos enfoques para su manejo se centran en disminuir los bancos de semillas mediante el aumento de la competitividad de los cultivos y, en consecuencia, la reducción de la lluvia de semillas de malezas. Sin embargo, la heterogeneidad espacial que existe dentro de los campos ocasiona diferencias en el crecimiento y la competitividad de los cultivos y las malezas. Para determinar si las interacciones competitivas entre cebada y L. rigidum son sitio–específicas, la biomasa y la producción de semillas de esta maleza, cultivada en monocultivo (parcelas con L. rigidum) y en cultivo mixto (parcelas con L. rigidum + cebada), se estudiaron en tres sitios (altiplano, laderas medias y tierras bajas) dentro de campos de cebada. En cada sitio se determinaron las poblaciones de maleza y el contenido de nutrientes, materia orgánica, inclinación y orientación en las fracciones del suelo. La presencia del cultivo disminuyó significativamente la biomasa de la maleza entre 5 y 79 % y el número de semillas por espiga entre 10 y 48 %, dependiendo del sitio. El efecto competitivo del cultivo fue mayor en los sitios más fértiles (con mayor contenido de N, P y materia orgánica). En estos sitios, las diferencias en la acumulación de biomasa entre la maleza en monocultivo y en cultivo mixto empezaron a ser significativas después de la elongación del tallo. Aun cuando se observó una reducción en el número de semillas por espiga en los sitios más fértiles, la lluvia de semillas (medida como semillas m–2 ) todavía sería muy importante si la densidad de la maleza del sitio es elevada. Las diferencias en las interacciones competitivas entre cebada y L. rigidum observadas dentro de los campos sugieren que las buenas prácticas agrícolas de manejo sitio–específico para aumentar la competitividad del cultivo pueden desempeñar una función importante como mecanismo para reducir las poblaciones de L. rigidum a largo plazo.
]]> Palabras clave: Hordeum vulgare, competencia de cultivo, lluvia de semillas, sitio–específico, topografía, biomasa de la maleza.
INTRODUCTION
Lolium rigidum L. is a widespread and troublesome grass weed in cereal fields of the Mediterranean region (Gill, 1996, Recasens et al, 1996). According to Gill (1996), ecological features such as high genetic variability, plasticity, fecundity and seed survival have contributed to its success as a major grass weed. In Catalonia, northeastern Spain, L. rigidum is present in more than 50 % of the cereal fields, occurring in a wide range of soils and environmental conditions and causing important yield losses to the crops (Izquierdo et al, 2003). L. rigidum resistance to various herbicide groups (ACCase inhibitors, ureas, photosystem II inhibitors, glycines) has been reported in several Spanish locations, threatening the sustainability of the existing herbicide–dependent cropping systems (Heap, 2008).
Most studies about crop–weed competition were carried out under homogeneous soil conditions (Izquierdo et al, 2003; Lemerle et al, 2004) and, consequently, did not take into account the underlying variability within a field. Topographic features (Shafii et al., 2003; Burton et al., 2005) and soil properties (Walter et al., 2002; Nordmeyer and Häusler, 2004) are some of the factors that may act at the local scale and influence the response of weeds to competition. Characterizing spatial variability is essential for agricultural planning and site–specific management.
Weed populations have been found to be spatially heterogeneous within agricultural fields, in patches of various sizes and shapes (Bianco–Moreno et al., 2004; Ruiz et al., 2006 a). The spatial distribution of some weed species has been associated with various site properties such as elevation, exposure, slope angle and aspect, soil–water accumulation, soil texture and fertility (Burton et al., 2005; Dieleman et al, 2000; Ruiz et al, 2006 b). In this regard, the spatial heterogeneity of L. rigidum populations frequently found in fields with a rolling landscape is probably associated with the differential crop–weed interactions in different landscape positions. In order to gain a population dynamics perspective to develop sustainable and integrated site–specific L. rigidum management programs, knowledge of the variability of the effect of the crop on the biomass and reproductive fitness of the weed is required.
The aim of this study was to determine the effect of barley on L. rigidum survival, biomass and seed production in different sites of the same field, as well as the consequences of field heterogeneity on crop–weed competition.
MATERIALS AND METODS
Locations and experimental design
]]> Experiments were conducted in commercial barley fields with dense L. rigidum populations located at Calaf and Moiá in Catalonia, in the northeastern cereal growing region of Spain. Both fields had southern exposure and were irregularly shaped, with differences in elevation up to 10 m and an average slope of 10.5 %. Fields were managed with agronomic practices typical of the region. Seedbeds were prepared with one pass of harrow before planting at Calaf, while the Moiá location was directly drilled. Fields were sown with winter barley cv. Dobla, at 180 kg seed ha–1 with the farmer's own seeding equipment. At Calaf, a granular application of NPK at 33–49–49 kg ha–1 was added annually before sowing. In addition, the upper part of this field was fertilized by the farmer with 25 t ha–1 of organic manure prior to seedbed preparation. At Zadocks stages Z13–15, a liquid application of SN32 (urea–ammonium nitrate) at 90 kg N ha–1 was added. At Moiá, liquid manure was applied twice at Z13–15 and Z21–22 at the rate of 35 000 L ha–1. Fields were sprayed with diclofop (Iloxan, 360 g ai L–1, EC, Bayer Cropscience SL) and tribenuron (Granstar, 750 g ai L–1, WG, DuPont Ibérica SL) at the label–recommended rate of 2.5 L ha–1 and 30 g ha–1 for grass and broadleaf control. During herbicide application, experimental plots were covered with plastic to avoid any damage to the L. rigidum plants.The irregular shape of the fields allowed identify three sites, upland, mid–slope and lowland, within each field and a trial was set up in each of them. According to the farmers, crop yields in these sites were different during the last years and differences were attributed to different soil composition and fertility. To characterize the sites, soils were analyzed at the beginning of the growing seasons for the 0 to 20 cm depth, and soil separates, nutrient content and organic matter were determined. Slope and orientation (degrees from north) were calculated for each site.
The experimental layout in each field was a split–plot design with site as the main plots (3.25 m × 2.25 m). Site had three levels (upland, mid–slope and lowland) and four replicates. Each plot was split in two (3.25 m × 1.12 m), establishing a L. rigidum monoculture in one half (removing all the barley seedlings by hand as they emerged) and maintaining in the other half a mixed population of barley and L. rigidum. Positions of subplots were at random. Plots were placed perpendicularly to barley rows (15 cm between rows) and separated by 0.5 m wide borders, accounting for the variability of the site. Broad–leaved weeds that emerged in the plots were removed by hand.
Measurements
In order to monitor plant biomass accumulation, one sample (0.5 m × 0.5 m) was taken in each treatment plot at four crop stages: initial tillering (Z21), stem elongation (Z30), early boot stage (Z45) and crop maturity (Z92). In each sample, barley and L. rigidum densities were assessed and plants were clipped at the soil surface, dried at 60 °C for 48 h and weighed. At the sampling times, four soil cores (20 cm deep by 4 cm wide) were collected from each site to determine soil moisture content gravimetrically. Seed production per plant was estimated at crop maturity, collecting 10 L. rigidum plants at random within each sample and counting the number of spikes and the number of seeds from 15 randomly chosen tillers of these plants. Seed rain (measured as seeds m–2) was calculated multiplying seed production by the estimated density of weeds. Seed size was estimated by weighting 1000 grains selected at random from these plants. L. rigidum losses were assessed by relating the L. rigidum data estimated in monoculture plots with the data estimated in the mixed culture plots. L. rigidum survival in each site was assessed by marking 80 plants at random at the beginning of the season (20 in each plot) and recording their survival at crop maturity. Marked plants were located at one end of the plots to avoid being collected during the sampling times.
Statistics
L. rigidum data sets from the two fields were analyzed separately using analysis of variance. For each field, a model with two factors (crop and site) was used to compare the effect of barley on L. rigidum. Crop (monoculture and mixed culture) and site (upland, mid–slope and lowland) were considered fixed factors. Calculations were made using the GLM procedure (SAS version 8.2, SAS Institute, Cary, NC, USA) followed by Tukeys test (p<0.05). Because data from several variables were not normally distributed, they were transformed prior to statistical analysis (Table 1). To test the differences in biomass accumulation of L. rigidum among sites during the growing season, separate analysis of variance were done for each sampling time with the GLM procedure, and Tukey's test (p<0.05).
]]> RESULTS AND DISCUSSION
Soil characteristics of the sites
Calaf
The upland site had significant higher N, P, organic matter and organic carbon contents than the mid–slope and lowland sites (Table 2). Clay content was also higher at the upland site than at the two other sites, leading to higher soil moisture content throughout the growing season (Figure 1). Rainfall during the growing season was 390 mm.
Moiá
Concentration of P and Kwere significantly higher at the upland site than at the two other sites (Table 2). The values of the remaining soil variables were similar among sites, with a slightly (not significant) higher sand and organic C content at the upland site. In this location, no significant differences were found in soil moisture at the three sites throughout the growing season (Figure 1). Rainfall during the growing season was 620 mm.
Weed density and survival
Calaf
Densities of L. rigidum were very high (average of 2,480 plants m–2). According to the farmers, L. rigidum control was accomplished historically with herbicides that had the same mode of action (diclofop, tralkoxidim). Repeated treatments may have resulted in the appearance of herbicide resistant–genotypes in this location. The mid–slope and lowland sites had three times higher weed densities (average 3744 and 2632 plants m–2) than the upland site (average 1066 plants m–2). This fact was possibly related to the water runoff of the seeds from the uppermost area of the field (Figure 2). L. rigidum survival recorded at the end of the season was above 85 % in all plots except at the upland site in mixed culture, where survival was significantly lower (28 %; Figure 3). A significantly higher barley biomass was observed in this site (Figure 4), which also showed the highest nitrogen content (Table 2).
Moiá
]]> Average densities of L. rigidum were 407 plants m–2, with values ranging from 206 plants m–2 at the mid–slope site to 593 plants m–2 at the upland site (Figure 2), with not statistically significant differences. According to the farmers, agronomic practices such as crop and herbicide rotations and no tillage were regularly carried out and may have contributed to maintain lower levels of L. rigidum populations. L. rigidum survival at the end of the season was greater than 90 % in all plots, with no significant differences among sites (Figure 3).Weed biomass and reproduction
Calaf
In monoculture, the biomass accumulation of L. rigidum was similar in the three sites up to stem elongation (20 weeks after sowing). From this on, weed biomass in the upland site increased faster and, at the end of the season, it was almost four times higher than in the other sites (Figure 5A). L. rigidum spike production was close to 1.5 spikes per plant in all sites, with or without crop (Figure 6A). The number of seeds per spike was higher in the upland site (Figure 6B), but seed rain was 40 to 50 % lower than at the two other sites (Figure 6C). Apparently, the lower plant densities in the upland site compensated for the greater seed production of each individual plant.
In mixed culture, L. rigidum biomass accumulation and final plant biomass were similar at the three sites (Figure 5B). Significant competitive effect of barley could be observed from stem elongation in the upland site leading a L. rigidum biomass reduction per plant at the end of the season of 68 %. No differences were found in plant biomass between monoculture and mixed culture in the other sites (Figure 5C). Additionally, in the upland site the presence of barley significantly reduced (42 %) the number of seeds per spike (Figure 6B). L rigidum seed size was neither affected by position nor presence of the crop (data not shown).
Moia
No significant differences among sites were detected on biomass accumulation of individual plants growing in monoculture or in mixed culture (Figure 5D and E). Significant and similar competitive effect of barley was observed in mixed culture in all sites from stem elongation. At the end of the season, L. rigidum biomass per plant was reduced 79 % at the upland, 74 % at the lowland and 61 % at the mid–slope sites (Figure 5F). Significant and similar reductions in the numbers of spikes per plant (64–73 %), seeds per spike (44–48 %) and seed rain (74–89 %) of L. rigidum in the three sites were observed in presence of barley (Figure 6D, E and F). Seed size was neither affected by position nor presence of the crop (data not shown).
The competitive effect of barley on L. rigidum was observed in both fields. Crop presence significantly reduced weed biomass and seeds per spike in Calaf and, additionally, number of spikes in Moiá. This effect was not significantly different among sites in Moiá, but varied among sites in Calaf. In this location, crop competitiveness was significantly lower in the mid–land and lowland site, where nitrogen and phosphorus content were significantly lower (0.18 % in both sites versus 0.30 % in the upland site). Apparently, the competitive effect of barley on L. rigidum growth and reproduction was more important in the more fertile areas. Similar results were reported by Ruiz et al. (2008) studying the competitiveness of barley on Avena sterilis. The increased growth and seed production of that weed in the more fertile areas was counterbalanced by the increased suppressive effect of the crop. Izquierdo et al. (2003) also reported that barley showed greater competitiveness against L. rigidum in environments with not limiting water supply. Barley is considered a crop with a high potential to suppress L. rigidum populations due to its great efficiency in nitrogen uptake (González–Ponce, 1998), great initial growth rate (Cousens, 1996) and more extended canopy. However, the effect of barley on L. rigidum survival was limited, indicating that once weed seedlings are established they are likely to complete their development.
No evidence of relationship between the topographic position (orientation and slope) and the competitive relationship between barley and L. rigidum can be suggested from our results. In the Mediterranean areas, light is not considered a limiting factor. Consequently, topographic factors such as orientation, slope and position within a field are not as important for the growth of crops as other factors (soil moisture or nutrient content of the soil). However, the little importance of the topographic location of the site per se can not be generalized. As pointed out by Wright et al. (1990), steep slope sites tend to have lower soil fertility due to nutrient runoff and erosion processes that alter the distribution of soil chemical and physical properties. Under such conditions, competitiveness of the crop will be diminished.
Increasing crop competitiveness can be a useful technique for weed management in organic or low input farming systems (prevalent in the Mediterranean area) or when herbicide resistance develops in weeds. Increased crop competitiveness can be achieved by either using adequate seeding rates (Medd et al, 1985; Lemerle et al, 2004) or adequate fertiliser application. Current weed management approaches focus more on the long–term reduction of the weed seed bank obtained by interfering with the reproduction of the weed than on the alleviation of the competition with the crop obtained by eliminating individuals (Jones and Medd, 2005). Reducing seed return will help to prevent weed spread and reduce weed populations. Crop husbandry practices addressed site–specifically in order to account for the spatial variability of the soil and leading to enhance crop competition, will improve L. rigidum control and possibly substitute, at least in part, for herbicides.
]]> The main drivers of L. rigidum success in the two fields studied were weed density and crop competition. Emerged seedling populations are consequence of the seed bank present in the soil and the weather conditions during the germination period. Historical crop and weed management of the field are likely to regulate the seed bank size because agronomic practices determine weed seedling survival, plant fecundity and seed dispersal. Any mismanagement that results in increased L. rigidum seed production will result in a rapid increase in population size.Further studies should be carried out in order to confirm and quantify the relationship between soil fertility and barley competitiveness and the influence of environmental conditions (such as water availability) on barley – L. rigidum interactions. Our study is restricted to the description of the consequences of field heterogeneity on crop–weed competition. The recognition of this variability is a pre–requisite for site–specific weed management practices leading to a better control of this weed.
CONCLUSIONS
The competitive effect of barley on L. rigidum survival, biomass and seed production was not uniform within a field. Greater competitive effect of the crop was observed in sites with higher nitrogen, phosphorus and organic matter content. In these sites, L. rigidum plant survival, final biomass and number of seeds per spike were reduced up to 67 %, 79 % and 48 %. However, weed density determined the final seed rain in each site. Crop husbandry practices should be addressed site–specifically in order to enhance crop competitiveness throughout the field, as a mechanism to reduce L. rigidum populations over the long term.
ACKNOWLEDGEMENTS
We are very grateful to Dr. Mortensen and the Weed Science Lab at the Pennsylvania State University for their valuable help and comments on the manuscript. The present research was funded by the Spanish Commission for Science and Technology (project AGL2002–04468–C03–03).
LITERATURE CITED
]]>Blanco–Moreno, J. M., L. Chamorro, R. M. Masalles, J. Recasens, and F. X. Sans. 2004. Spatial distribution of Lolium rigidum seedlings following seed dispersal by combine harvesters. Weed Res. 44: 375–387. [ Links ]
Burton, M. G., D. A. Mortensen, and D. B. Marx. 2005. Environmental characteristics affecting Helianthus annum distribution in a maize production system. Agrie. Ecosystems and Environ. 111: 30–40. [ Links ]
Cousens, R. 1996. Comparative growth of wheat, barley and annual ryegrass (Lolium rigidum) in monoculture and mixture. Austr. J. Agrie. Res. 47: 449–464. [ Links ]
Dieleman, J. A., D. A. Mortensen, D. D. Buhler, C. A. Cambardella. and T. B. Moorman. 2000. Identifying associations among site properties and weed species abundance. I. Multivariate analysis. Weed Sci. 48: 567–575. [ Links ]
Gill, G.S. 1996. Ecology of annual ryegrass. Plant Protection Quart. 11: 195–198. [ Links ]
González–Ponce, R. 1998. Competition between barley and Lolium rigidum for nitrate. Weed Res. 38: 453–460. [ Links ]
Heap, I. 2008. The International Survey of Herbicide Resistant Weeds (on line). Available in http://weedscience.com (12 May, 2008). [ Links ]
Izquierdo, J., J. Recasens, C. Fernández–Quintanilla, and G. Gill. 2003. Effects of crop and weed densities on the interactions between barley and Lolium rigidum in several Mediterranean locations. Agronomie 23: 529–536. [ Links ]
Jones, R. E., and R. W. Medd. 2005. Methodology for evaluating risk and efficacy of weed management technologies. Weed Sci. 53: 505–514. [ Links ]
Lemerle, D., R. D. Cousens, G. S. Gill, and S. J. Peltzer. 2004. Reliability of higher seeding rates of wheat for increased competitiveness with weeds in low rainfall environments. J. Agrie. Sci. 142: 395–409. [ Links ]
Medd, R. W, B. A. Auld, D. R. Kemp, and R. D. Murison. 1985. The influence of wheat density and spatial arrangement on annual ryegrass, Lolium rigidum Gaudin, competition. Austr. J. Agrie. Res. 36: 361–371. [ Links ]
Nordmeyer, H., and A. Häusler. 2004. Impact of soil properties on weed distribution within agricultural fields. J. Plant Nutr. Soil Sci. 167:328–336. [ Links ]
Recasens, J., F. Riba, J. Izquierdo, R. Forn, and A. Taberner. 1996. Grass weeds growing in winter cereals of Catalonia. ITEA: Producción Vegetal 2: 116–130. [ Links ]
Ruiz, D., C. Escribano, and C. Fernández–Quintanilla. 2006a. Assessing the opportunity for site–specific management of Avena sterilis in winter barley fields in Spain. Weed Res. 46: 379–387. [ Links ]
Ruiz, D., C. Escribano, and C. Fernández–Quintanilla. 2006b. Identifying associations among sterile oat (Avena sterilis) infestation level, landscape characteristics, and crop yields. Weed Sci. 54: 1113–1121. [ Links ]
Ruiz, D., J. Barroso, P. Hernaiz, and C. Fernández–Quintanilla. 2008. The competitive interactions between winter barley and Avena sterilis L. are site–specific. Weed Res. 48: 38–47. [ Links ]
Shafii, B., W. J. Price, T. S. Prather. L. W. Lass, and D. C. Thill. 2003. Predicting the likelihood of yellow starthistle (Centaurea solstitialis) occurrence using landscape characteristics. Weed Sci. 51: 748–751. [ Links ]
Walter, A. M., S. Christensen, and S. E. Simmelsgard. 2002. Spatial correlation between weed species densities and soil properties. Weed Res. 42: 26–38. [ Links ]
Wright, R. J., D. G. Boyer, W. N. Winant, and H. D. Perry. 1990. The influence of soil factors on yield differences among landscape positions in an Appalachian cornfield. Soil Sci. 149: 375–382. [ Links ] ]]>