Stubble management and microclimate, yield and water use efficiency of canola grown in the semiarid Canadian prairie

Transcription

1 Stubble management and microclimate, yield and water use efficiency of canola grown in the semiarid Canadian prairie H. W. Cutforth 1, S. V. Angadi 2, and B. G. McConkey 1 1 Semiarid Prairie Agricultural Research Centre, Agricultural and Agri-Food Canada, P.O. Box 1030, Swift Current, Saskatchewan, Canada S9H3X2; and 2 New Mexico State University, Agricultural Science Center, Clovis, NM , USA. Received 14 April 2005, accepted 15 August Cutforth, H. W., Angadi, S. V. and McConkey, B. G Stubble management and microclimate, yield and water use efficiency of canola grown in the semiarid Canadian prairie. Can. J. Plant Sci. 86: Standing stubble traps snow and creates a favorable microclimate, which increases yields in wheat (Triticum aestivum L.) and pulses [chickpea (Cicer arietinum L.), field pea (Pisum sativum L.) and lentil (Lens culinaris L.)]. Generally, the taller the stubble the greater is the effect on microclimate and yield. A field study using farm-scale seeding and harvesting equipment was conducted over four seasons (1999 to 2002) to assess the effect of stubble management on the microclimate, water use and seed yield of argentine canola (Brassica napus L. Arrow ) in the semiarid prairie surrounding Swift Current. Tall (30 cm), short (15 cm) and cultivated stubble treatments were deployed in fall and in spring. An additional tall stubble treatment with extra fertilizer N (application rate recommended for the Black soil zone in the subhumid prairie) was included to assess the role of fertilizer in canola response to stubble management practices. The differences in wind velocity, soil temperature and solar radiation reaching the soil surface indicated significant modification of the microclimate by tall compared with cultivated stubble. Yields were highest from the tall stubble receiving extra fertilizer. Further research is needed to determine optimum fertilizer rates to maximize canola yields in the semiarid prairie. For treatments receiving equivalent rates of fertilizer, tall stubble increased seed yield of canola by about 24% and water use efficiency (WUE) by about 19% compared with stubble cultivated in the fall. Comparing between stubble treatments deployed on fields that overwintered as tall stubble and which received equivalent rates of fertilizer, tall stubble increased canola yield by about 16% and WUE by about 11% compared with cultivated stubble. Crop water use was not affected by stubble management so the increased grain production was due to increased WUE. Key words: Stubble height, microclimate, canola, yield, water use efficiency Cutforth, H. W., Angadi, S. V. et McConkey, B. G Gestion du chaume et microclimat, rendement et efficacité de l utilisation de l eau chez le canola cultivé dans les zones semi-arides des Prairies canadiennes. Can. J. Plant Sci. 86: Non fauché, le chaume retient la neige et engendre un microclimat favorable qui accroît le rendement du blé (Triticum aestivum L.) et des légumineuses [pois chiche (Cicer arietinum L.), pois de grande culture (Pisum sativum L.) et lentille (Lens culinaris L.)]. En général, plus haut est le chaume, meilleurs sont ses effets sur le microclimat et le rendement. Les auteurs ont procédé à une étude sur le terrain en recourant à de l équipement commercial pour les semis et la récolte; l étude a duré quatre saisons (de 1999 à 2002) et devait servir à évaluer l incidence de la gestion du chaume sur le microclimat, l utilisation de l eau et le rendement grainier du canola argentin (Brassica napus L. cv Arrow) dans la région semi-aride des Prairies qui entoure Swift Current. À cette fin, les auteurs ont recouru à du chaume élevé (30 cm), du chaume court (15 cm) et du chaume cultivé à l automne et au printemps. Ils ont aussi examiné un autre traitement consistant en du chaume élevé avec application d engrais N supplémentaire (au taux recommandé pour les sols noirs dans la zone sub-humide des Prairies) de manière à évaluer le rôle de l engrais dans la réaction du canola aux pratiques de gestion du chaume. Les variations au niveau de la vitesse du vent, de la température du sol et du réchauffement du sol par le soleil indiquent que le chaume élevé modifie significativement le microclimat comparativement au chaume cultivé. Le traitement chaume élevé et engrais donne le meilleur rendement. Il faudrait entreprendre d autres recherches pour établir le taux de fertilisation qui maximiserait le rendement du canola dans la zone semi-aride des Prairies. Lorsque les taux de fertilisation sont identiques, le chaume élevé accroît le rendement grainier du canola d environ 24 % et l efficacité de l utilisation de l eau d environ 19 % comparativement au chaume cultivé en automne. Quand on compare les différents traitements employés pour conditionner les champs laissés sur chaume élevé en hiver à des taux de fertilisation équivalents, on se rend compte que le chaume élevé augmente le rendement du canola d environ 16 % et l efficacité de l utilisation de l eau d environ 11 % par rapport au chaume cultivé. La gestion du chaume ne modifie pas la quantité d eau utilisée par la culture, si bien que la hausse de la production grainière résulte d une utilisation plus efficace de l eau. Mots clés: Hauteur du chaume, microclimat, canola, rendement, efficacité de l utilisation de l eau Low water availability is the greatest limitation to crop production in the Canadian semiarid prairie. Potential evaporative demand in the Brown soil zone is the highest of any region on the Prairies (Cutforth et al. 1993). From long-term weather data, average daily precipitation begins to decline 3 4 wk before average daily maximum temperatures have 99 reached their highest values in late July (Angadi et al. 2004). Therefore, the moisture deficit for the semiarid prairie typically increases as the growing season progresses. The transpirational component of evapotranspiration can be increased by increasing the water supply and/or by decreasing evaporative water losses. Low-disturbance direct seed-

2 100 CANADIAN JOURNAL OF PLANT SCIENCE ing into standing stubble is a very effective practice for increasing the supply and for reducing evaporation of water. Infiltration water from snow trapped by the standing stubble can increase soil water reserves (Campbell et al. 1992; Lafond et al. 1992). The amount of snow trapped is directly proportional to stubble height (Aase and Siddoway 1980; McConkey et al. 1997; Steppuhn 1994). Standing cereal stubble also reduces evaporation of soil water (Caprio et al. 1985; Cutforth and McConkey 1997; Cutforth et al. 2002a) effectively making more water available to the crop. Further, standing stubble creates a more favorable microclimate for crops by reducing wind, solar radiation, and evaporative demand for water (Aase and Siddoway 1980; Cutforth and McConkey 1997; Cutforth et al. 2002a). Previously, we found that growing season evapotranspiration was not affected by stubble management. However, in these studies, wheat (Cutforth and McConkey 1997) and pulses (Cutforth et al. 2002a) seeded into tall stubble had more than 10% higher grain yield and WUE compared with wheat or pulses seeded into cultivated stubble. Yield and WUE for wheat or pulses seeded into short stubble were intermediate to yields from the tall and cultivated stubble treatments. Standing stubble alters the microclimate, with the magnitude of the alterations dependent upon stubble height (Cutforth and McConkey 1997; Cutforth et al. 2002a). Whereas information is available on the benefit of tall stubble on yields of spring wheat and pulses, the same is not true for canola. Therefore, our objective was to determine the effects of standing wheat stubble of various heights on the in-crop microclimate and on the growth and yield of canola. MATERIALS AND METHODS A field study was conducted on a Swinton loam soil (Orthic Brown Chernozem) (Ayres et al. 1985) at the Agriculture and Agri-Food Canada, Semiarid Prairie Agricultural Research Centre (SPARC), Swift Current, for four seasons ( , beginning with late fall seeding in 1998). Glyphosate-tolerant Brassica napus L. Arrow was seeded with a Flexicoil 5000 Air seeder with Flexi-Coil Stealth knives (Flexi-Coil Ltd, Saskatoon, SK S7K 3S5) at 23-cm row spacing. A relatively high seeding rate of 9.5 kg ha 1 was used to attain an acceptable stand of both fall and spring seeded crops. Fertilizer rates recommended for the Brown soil zone were 84 kg N ha 1, 22 kg P ha 1 and 22 kg S ha 1, of which 78 kg N and all of S were broadcasted (as ammonium sulphate) in the spring and the remaining N and all P were mid-row banded at seeding. Post-emergent glyphosate applications kept the experimental area weed free. Stubble treatments were imposed on spring wheat grown on fallow (i.e., a fallow-wheat-canola crop rotation); the wheat was harvested using a header equipped combine to leave stubble taller than 30 cm. The six stubble management treatments were: (1) stubble cut at 30 cm tall in fall and direct seeded in spring; (2) stubble cut at 30 cm, then cut at 15 cm tall in fall and direct seeded in spring; (3) stubble cut at 30 cm, then cultivated in fall, and seeded in spring; (4) stubble cut at 30 cm tall in fall and direct seeded in spring, but fertilized with an additional 34 kg ha 1 of N (as ammonium nitrate); (5) stubble cut at 30 cm tall in fall and cut at 15 cm in spring, and direct seeded in spring; (6) stubble cut at 30 cm tall in fall then cultivated in spring, and seeded. The fall and spring cultivated plots were slowly (<5 km h 1 ) worked with a tandem disc followed by harrow-packers. About one-half of the standing stubble was buried with the remaining residue lying flat on the soil surface. The short stubble (15 cm high) plots in fall and spring were deployed by cutting the tall stubble with a haybine (sicklebar cutter) without windrowing. For one of the treatments, the tall stubble plots received an additional 34 kg N ha 1 in the spring in order to attain application rates of fertilizer N equivalent to recommended rates for the Black soil zone. Previously, we studied the effect of seeding date on canola yield and water use (Angadi et al. 2004), whereas, in this study, seeding dates were used to maximize the number of different environments within a given year experienced by the treatments. Seeding dates were late fall (Oct. 27 to Nov. 06; just before freeze-up), early spring (Apr. 24 to 25) and late spring (May 19 to 25). The main (stubble management) and subplot (seeding date) sizes were 45 m 45 m and 15 m 45 m, respectively. From 1999 to 2002, and within a randomly selected block, microclimate measurements were taken from three early spring seeded plots of tall, short (spring) and cultivated (spring) stubble. Because of limited resource availability, microclimate measurements were conducted on only one replicate. Soil temperatures were measured with thermistors at and 0.30-m depths. Wind velocities at 0.15, 0.50, 1.00 ( ) and 2.00 ( ) m above the soil surface were measured with anemometers (Wind Sentry and Gill 3 Cup anemometers, RM Young Company, Traverse City, MI). Solar radiation above the canopy and solar radiation reaching the soil surface (approximately 7 cm above the ground) were measured using 1-m-long tube solarimeters ( Monteith pattern tube solarimeter, Delta-T Devices Ltd. Cambridge, UK). Soil temperatures were measured at three to four locations within a plot. Wind velocity and solar radiation measurements were recorded at one location per plot. For each plot, soil water was measured gravimetrically just before the early seeding date and after harvest to a depth of 1.2 m. Consumptive water use (CWU) was defined as: CWU = (soil water at seeding soil water at harvest) + growing season precipitation. Water use efficiencies (WUE) were calculated: WUE = grain yield/cwu. Before harvest, one 1.8-m row length from each quarter of a given plot (i.e., four rows per plot) were hand harvested to measure biomass production and harvest index. Seed yields were measured with a MF 550 combine (AGCO Corporation, 4205 River Green Parkway, Duluth, GA 30096) after windrowing; the middle swath in each plot (5.5 m 18 m) was used for seed yield estimation. Error variances were homogeneous across years for water use, biomass and WUE (P > 0.10), and for seed yield and

3 CUTFORTH ET AL. STUBBLE MANAGEMENT AND MICROCLIMATE, YIELD AND WUE 101 harvest index (P > ), according to Bartlett s test (Gomez and Gomez 1984). Data were analyzed using the PROC MIXED procedure of the SAS Institute Inc. (Littel et al. 1996), with years considered as a random effect and with stubble management as a fixed effect. As well, the estimation method was set to ML (maximum likelihood) and the degrees of freedom method was set to Satterthwaite. The seeding dates were combined for analyses. Least significant differences (LSD 0.05 ) were determined using LSMEANS. Treatment effects were declared significant at α = RESULTS AND DISCUSSION Weather Precipitation from October 1998 through April 1999 was above normal (Table 1). Average monthly air temperatures during the 1999 May-June-July-August (MJJA) growing season were cooler than normal until August. The MJJA growing season precipitation was 20% above normal with May, June and July receiving above-normal rainfall. August was very dry. Precipitation from October 1999 through April 2000 was near normal. Average monthly air temperatures during the 2000 growing season were near normal. MJJA precipitation was 29% above normal. Overall, 2001 was warm and extremely dry representing a severe drought; the 2nd driest and 5th warmest year on record for Swift Current (Judiesch and Cutforth 2002). Precipitation was 52% of normal. October through April precipitation was well below normal. Air temperatures during May, July, and August were warmer than normal. May and June received less than half their normal precipitation whereas precipitation during July was near normal mainly because of two significant rainfall events. August had only a trace of rain. The dry weather continued throughout 2001 until June Air temperatures during May and August were well below normal but slightly above normal for June and July. Whereas October through May was extremely dry, the rest of 2002 was wet, especially June through August, which was very wet. June and August each received at least twice their normal rainfall, and the rainfall for July was 30% above normal. The 1999 growing season was cool and moist. The 2000 growing season was wet. Both 1999 and 2000 experienced brief late season dry spells. The 2001 growing season was warm and extremely dry throughout representing a severe drought. In 2002, the initial portion of the growing season was very dry (an early season drought) whereas the later two thirds was wet. Microclimate Microclimate before and after flowering was significantly affected by stubble management (Table 2). Similar to previous results, (Cutforth and McConkey 1997; Cutforth et al. 2002a), the microclimate effects of short stubble were intermediate to those for cultivated and tall stubble and thus, for clarity, were not presented here. Before and after flowering, the average daily wind speeds were greater over the cultivated stubble than over the tall stubble (200 cm wind speeds before flowering were different at P = 0.058). After flowering, the average daily wind speed at 15 cm was near zero for both cultivated and tall stubble treatments. Compared with cultivated stubble, tall stubble reduced wind speeds by about 77% at 15 cm (before flowering only), 14% at 100 cm, and 7% at 200 cm heights. Both before and after flowering, the average daily soil temperatures at the 5-cm depth were greater under cultivated than tall stubble. Average daily total incoming solar radiation measured at about 7.5 cm above the soil surface was lowest for the tall stubble (about 21% lower before flowering; about 38% lower after flowering) compared with the cultivated stubble. These results were similar to those we and others have reported previously, although the reduction in solar radiation by the canolastanding stubble treatments was twice as large as the reduction by the spring wheat-standing stubble treatments (Aase and Siddoway 1980; Caprio et al. 1985; Cutforth and McConkey 1997; Cutforth et al. 2002a). Canola seedlings and plants are much larger than wheat. Leaf orientation of canola is more horizontal, whereas wheat leaves appear more vertical. Therefore, the shading potential for canola is probably much larger than for wheat. The effects of stubble management on microclimate are similar to those we reported previously (Cutforth and McConkey 1997; Cutforth et al. 2002a). When spring wheat was in the seedling stage, tall stubble reduced wind speed by approximately 70 and 10% at 15 cm and 100 cm above ground. Tall stubble also reduced average daily soil temperatures at 5 cm and 30 cm soil depths, reduced incoming (by about 11%) and reflected (by about 19%) solar radiation near the soil surface compared with the cultivated treatment (Cutforth and McConkey 1997). Throughout the growth cycle of spring wheat, tall stubble reduced evaporation rate at the soil surface by about 25% compared with the cultivated treatment. Before flowering of pulses, tall stubble reduced wind speed by about 70 and 8% at 15 cm and 100 cm above ground, and reduced evaporation rate at the soil surface by about 26% compared with the cultivated stubble (Cutforth et al. 2002a). Both before and after flowering, soil temperatures at 5 cm and 30 cm depths were reduced by the tall stubble compared with the cultivated stubble. Depending on the height of the stubble, stubble management changed the microclimate near the soil surface by reducing wind speed, solar radiation, and soil temperatures throughout the life cycle of canola. Prior to flowering (about Julian day 180), wind speeds at 15 cm above the soil surface were always substantially lower in the tall compared with cultivated stubble (Fig. 1). After flowering, very little wind occurred near the soil surface for either tall or cultivated stubble. At 1-m and 2-m heights, differences between tall and cultivated stubble were minimal at lower wind speeds; but at higher wind speeds, winds over the cultivated stubble were greater than those over tall stubble. Throughout the growing season, solar radiation near the soil surface in the tall stubble was equal to or less than solar radiation near the soil surface of the cultivated stubble (Fig. 2). Before stem extension (bolting), solar radiation near the soil surface of the cultivated stubble was equivalent to incom-

4 102 CANADIAN JOURNAL OF PLANT SCIENCE Table 1. Average mean air temperatures and precipitation for the winter months beginning the previous October to March, and for the months of April through August from 1999 to 2002, as well as the long-term ( ) mean monthly air temperature and precipitation total Year Month Mean temperature ( C) Precipitation (mm) Oct. Mar. z Apr May Jun Jul Aug May Aug z Previous October to March Table 2. From 1999 to 2002 (1999, 2000 for wind at 200 cm height above the soil surface), daily average wind speed, soil temperature and incoming solar energy (approximately 7.5 cm above the soil surface) before and after flowering of canola grown in cultivated and tall stubble at Swift Current, SK Wind (m s 1 ) Soil temperature ( C) Growth stage Stubble 15 cm height 100 cm height 200 cm height 5 cm depth 30 cm depth Solar energy (MJ d 1 ) Before flowering Cultivated Tall LSD After flowering Cultivated Tall LSD ing solar radiation at 2 m whereas solar radiation near the soil surface in tall stubble was generally lower. Once bolting began, solar radiation near the soil surface of tall and cultivated stubble steadily decreased relative to solar radiation at 2 m with the decrease generally starting earlier in tall stubble. However, because of extreme drought (Table 1), exceptions occurred in the latter half of the 2001 and in the spring portion of the 2002 growing seasons. Early in the season, when the canola seedlings were small and close to the soil surface, the entire difference in solar radiation penetration to about 7.5 cm above the soil surface (sensor height) was due to the differences in standing stubble. As the seedlings grew and extended above sensor height, they shaded the sensor and contributed to treatment differences for solar radiation measured near the soil surface. Visual observations showed that with good moisture conditions, the canola seedlings grew taller more quickly, with larger leaves in the tall compared with cultivated stubble (Cutforth et al. 2002b). In 2001, because of the extreme drought, differences in seedling size initiated early in the growing season by stubble management disappeared during the latter half of the growing season. Because of the extreme water stress, canola plants were not able to take advantage of the microclimate differences due to stubble management. Thus, there were no differences between stubble managements in reducing solar radiation penetration to the soil surface. Seedlings growing in 2002 were severely water stressed and were unable to respond to the microclimate differences brought about by the differences in stubble management. Rain in early June relieved stress, at which time the canola plants were able to respond to the differences in microclimate. Seedling size, especially leaf size, increased for canola growing in tall compared with cultivated stubble (Cutforth et al. 2002b). The increased shading of the soil surface by the larger, leafier canola in the tall stubble is reflected in the earlier, greater reduction in solar radiation measured near the soil surface from about bolting to near maturity. Soil temperatures at the 5-cm depth were generally lower under tall compared with cultivated stubble and these differences were maintained well after flowering when plants were at least double the height of the tall stubble (Fig. 3). However, early in the growing season differences in the average daily temperatures between stubble treatments were often minimal. With cooler springs (Table 1: April May), treatment differences took longer to manifest (compare 2002 with 1999). As well, the extreme water stress due to the drought of 2001 and which extended to early June 2002, affected the growth of canola seedlings in Because of the extreme stress, the seedlings were unable to respond to the microclimate differences relative to stubble management and, therefore, plant and leaf sizes were similar for tall and cultivated stubble. The severe stress minimized shading differences between stubble treatments that would have occurred under more favorable moisture conditions. Thus, similar amounts of solar energy probably penetrated the canopy to the soil surface resulting in minimal differences in soil temperatures in the surface soil layer. Once moisture conditions improved, plants were able to respond to the microclimate differences with leaf and plant sizes increasing for canola growing in tall compared with cultivated stubble. The increased shading of the soil surface would probably result in lower temperatures within the surface soil layer for tall compared with cultivated stubble treatments. Effect on Crop Growth and Yield Error variances were homogeneous for water use, dry matter and seed yield, harvest index (HI), and WUE, and there-

5 CUTFORTH ET AL. STUBBLE MANAGEMENT AND MICROCLIMATE, YIELD AND WUE 103 Fig. 1. For each year, the relationship of wind speed at 15 cm, 100 cm (2001, 2002), and 200 cm (1999, 2000) above the soil surface with Julian day for the spring deployed cultivated and tall stubble treatments.

6 104 CANADIAN JOURNAL OF PLANT SCIENCE Fig. 2. For each year, the relationship of incoming solar energy approximately 7 cm above the soil surface with Julian day for the spring deployed cultivated and tall stubble treatments, and incoming solar energy at 2 m above the soil surface

7 CUTFORTH ET AL. STUBBLE MANAGEMENT AND MICROCLIMATE, YIELD AND WUE 105 Fig. 3. For a given year, the relationship of soil temperature at the 5 cm depth with Julian day for the spring deployed cultivated and tall stubble treatments. Also identified for each year are the Julian day for 50% flowering.

8 106 CANADIAN JOURNAL OF PLANT SCIENCE fore data were pooled across years for analyses. The effect of stubble management on canola growth and yield was consistent across years. Therefore, the effect of stubble management on canola growth was independent of within year and between year variations in weather. The highest biomass, grain yield and WUE occurred for canola growing in tall stubble with extra fertilizer (fertilizer N applied at rates recommended in the Black Soil Zone) (Table 3). Tall stubble with extra fertilizer increased canola dry matter, grain yield and WUE by about 15% compared with canola in tall stubble. This suggests that to achieve maximum yields in the semiarid prairie, canola seeded into tall standing stubble should be fertilized with rates at least equivalent to those used in the Black soil zone. However, further research on the interaction of fertilizer N rate and stubble management is needed to revise fertilizer recommendations for canola grown in the semiarid prairie. Canola and mustard grown in the semiarid prairie can respond to higher levels of N under favorable moisture conditions (Miller et al. 2003). Canola grown in tall stubble with extra fertilizer tended to use the most water and had the highest HI, and, indeed, used more water and had a higher HI than canola grown in stubble cultivated the previous fall [cultivated (fall) stubble]. Generally, however, canola growing in tall stubble with extra fertilizer used similar amounts of water and had similar HI as canola growing in the other stubble management treatments, especially those treatments that overwintered as tall stubble [(i.e., tall, short (spring), cultivated (spring) stubble]. Water use and HI were independent of stubble management, especially for those treatments receiving the recommended fertilizer rates for the semiarid prairie. Comparing among treatments not receiving extra fertilizer, there was a tendency for canola grown in tall stubble to produce more dry matter than canola in short stubble, followed by canola in cultivated stubble (Table 3). However, the only significant difference in dry matter production occurred between canola growing in tall stubble compared with cultivated (fall) stubble. Thus, canola dry matter yields were generally independent of stubble management, especially for those stubble treatments deployed in the spring. [The lack of significant differences may have resulted because of our methodology using small sample sizes (four 1.8 m long rows) which have inherently large variability. This conclusion could change upon further research employing much larger sample sizes, i.e., dry matter yield may indeed be dependent upon stubble management.] Canola growing in tall stubble produced more grain and had higher WUE than canola growing in cultivated (spring or fall) stubble (Table 3). Compared with cultivated (spring) stubble, tall stubble increased canola yield by about 16% and WUE by about 11%. As well, canola in tall stubble produced more grain than canola in short (fall) stubble. Further, although not significant, compared with treatments that overwintered as tall stubble [short (spring) and cultivated (spring)], the fall-deployed treatments tended to use slightly less water, to produce slightly less dry matter, and tended to have lower grain yields, lower HI, and lower WUE partly because of increased snow trapping and/or reduced evaporative water loss from the treatments that overwintered as tall stubble (McConkey et al. 1997). However, generally, while using similar amounts of water, canola dry matter production was independent of stubble management whereas canola grown in tall stubble produced more grain and used water more efficiently than canola grown in cultivated stubble. Yield and WUE for canola seeded into short stubble were intermediate to the other stubble treatments. In general, the response of canola to standing stubble was similar to that of spring wheat (Cutforth and McConkey 1997) and pulses (Cutforth et al. 2002a). The overall effect of tall stubble was to consistently increase grain production and WUE of spring wheat and pulses relative to a cultivated seedbed; however, the benefits were generally neither large nor significant (P < 0.05) in any given year. In both studies, growing season evapotranspiration was independent of stubble management. On average, seeding wheat into tall stubble increased grain yield and WUE by about 12% compared with wheat seeded into cultivated stubble. Compared with cultivated stubble, tall stubble increased the average yield of pulses by about 13% and the average WUE by about 16%. For spring-deployed stubble treatments receiving equivalent rates of fertilizer, tall stubble increased the average yield of canola by about 16% and the average WUE by about 11% compared with cultivated stubble. The higher grain yields and WUE for the tall compared with cultivated stubble may be attributed to a decrease in potential evaporation, an increase in the proportion of evapotranspiration that was transpired by the crop, an increased net efficiency of carbon assimilation, and/or an increased capture of photosynthetically active radiation (Cutforth and McConkey 1997). However, some researchers argue that the overall benefit to the crop by the creation of a favorable microclimate by the tall stubble could be very small. For example, Grace (2003) contends that the reduced mechanical stimulation by a windbreak or shelterbelt is likely the more important consideration towards improved crop productivity. Grace (2003) suggests that mechanical stimulation by wind is more often than not a stress. Plants alter their allocation patterns of photosynthates from fruit (grain) production to cope with mechanical forces associated with wind, which could include the reinforcement of stems or the replacement of damaged tissues. In our analysis, we included the reduction in mechanical stimulation as part of the overall improvement of the crop microclimate by the taller stubble. Further research is needed to determine the relative importance of reduced mechanical stimulation to the creation of a favorable microclimate towards increasing crop yield. SUMMARY Compared with cultivated stubble, tall stubble positively modified the microclimate within the canola canopy. Seasonal trends in soil temperature at the 5 cm depth, wind velocity at 15 cm, 100 cm and 200 cm above the soil surface, and solar radiation interception near the soil surface were significantly modified by stubble management. Canola grown in tall stubble with extra fertilizer yielded more grain and used water more efficiently than canola in any of the

9 CUTFORTH ET AL. STUBBLE MANAGEMENT AND MICROCLIMATE, YIELD AND WUE 107 Table 3. Mean water use, dry matter, seed yield, harvest index and water use efficiency (WUE) of canola as affected by stubble management in the semiarid prairie Stubble Water use Dry matter Seed yield Harvest WUE (mm) (kg ha 1 ) (kg ha 1 ) index (kg ha 1 mm 1 ) Tall + fertilizer Tall Short (spring) Short (fall) Cultivated (spring) Cultivated (fall) LSD other treatments. Further research is needed to determine optimum fertilizer rates to maximize canola yields in the semiarid prairie. Comparing between stubble treatments deployed on fields that overwintered as tall stubble and which received the recommended rates of fertilizer for the semiarid prairie, tall stubble increased canola yield by about 16% and WUE by about 11% compared with cultivated stubble. Yield and WUE for canola seeded into short stubble were intermediate to the other stubble treatments. Crop water use was not affected by stubble management so the increased grain production was due to the increased WUE. ACKNOWLEDGMENTS The research was financed by Saskatchewan Canola Development Commission, Saskatchewan Agriculture Development Fund and AAFC-Matching Investment Initiative. We also thank Doug Judiesch, Don Sluth, Evan Powell, Jason Newton, Ken Deobold and several summer students for their technical help. Aase, J. K. and Siddoway, F. H Stubble height effects on seasonal microclimate, water balance, and plant development of no-till winter wheat. Agric. Meteorol. 21: Angadi, S. V., Cutforth, H. W., McConkey, B. G. and Gan, Y Early seeding improves the sustainability of canola and mustard production on the Canadian semiarid prairie. Can. J. Plant Sci. 84: Ayers, K. W, Acton, D. F. and Ellis, J. G The soils of the Swift Current map area 72J Saskatchewan. Saskatchewan Institute of Pedology. Publication S6. University of Saskatchewan, Saskatoon, SK. 226 pp. Campbell, C. A., McConkery, B. G., Zentner, R. P., Selles, F. and Dyck, F. B Benefits of wheat stubble strips for conserving snow in southwestern Saskatchewan. J. Soil Water Cons. 47: Caprio, J. M., Grunwald, G. K. and Snyder, R. D Effect of standing stubble on soil water loss by evaporation. Agric. For. Meteorol. 34: Cutforth, H. W., Jones, K. and Lang, T-A Agroclimate of the Brown soil zone of southwestern Saskatchewan. Research Branch, Agriculture Canada. Publ. No. 379MOO88. Cutforth, H. W. and McConkey, B. G Stubble height effects on microclimate, yield and water use efficiency of spring wheat grown in a semiarid climate on the Canadian Prairies. Can. J. Plant Sci. 77: Cutforth, H. W., McConkey, B. G., Ulrich, D., Miller, P. R. and Angadi, S. V. 2002a. Yield and water use efficiency of pulses seeded directly into standing stubble in the semiarid Canadian Prairie. Can. J. Plant Sci. 82: Cutforth, H. W., Angadi, S. V. and McConkey, B. G. 2002b. Seeding management to increase and stabilize canola production in the semiarid prairie. Final Report. Saskatchewan Canola Development Commission. Saskatoon, SK. [ scdc@scdc.sk.ca]. Gomez K. A. and Gomez, A. A Statistical procedure for agricultural research. 2nd ed. John Wiley and Sons, New York, NY. Grace, J Mechanical stress and wind damage. Encyclopedia of applied plant sciences. Book Series. Academic Press, London, UK. pp Judiesch D. and Cutforth, H : A drought odyssey. Res. News Letter, Semiarid Prairie Agricultural Research Centre (SPARC), Swift Current, SK Jan. 07. Lafond, G. P., Loeppky, H. and Derksen, D. A The effects of tillage systems and crop rotations on soil water conservation, seedling establishment and crop yield. Can. J. Plant Sci. 72: Littel, R. C., Milliken, G. A., Stroup, W. W. and Wolfinger, R. D SAS system for mixed models. SAS Institute, Inc. Cary, NC. 656 pp. Miller, P. R., Angadi, S. V., Androsoff, G. L., McConkey, B. G., McDonald, C. L., Brandt, S. A., Cutforth, H. W., Entz, M. H. and Volkmar, K. M Comparing Brassica oilseed crop productivity under contrasting N fertility regimes in the semiarid northern Great Plains. Can. J. Plant Sci. 83: McConkey, B. G., Ulrich, D. J. and Dyck, F. B Snow management and deep tillage for increasing crop yields on a rolling landscape. Can. J. Soil Sci. 77: Steppuhn, H Snowcover retention capacities for direct combined wheat and barley stubble in windy environments. Can. Agric. Eng. 36:

10