Grapevine and Soil Water Relations with Nodding Needlegrass (Nassella cernua), a California Native Grass, as a Cover Crop

Nodding needlegrass [Nassella cernua (Stebbins & R.M. Love) Barkworth], a California native perennial grass, was tested for its effects on grapevine and soil–water relations in a drip-irrigated vineyard in Parlier, CA. Vine water status and in-row and between-row soil moisture (at 0.3 m, 0.6 m, 0.9 m, 1.2 m, and 1.5 m) were monitored semiweekly from June to September. There was no overall significant difference in leaf water potential between treatments. In-row soil moisture was lowest at depths of 0.6 m to 0.9 m within the nodding needlegrass treatment but was lowest from 0.3 m to 0.9 m within the clean cultivation treatment. Compared with clean cultivation, nodding needlegrass in-row soil moisture was significantly higher at depths of 0.3 m and 0.6. m and did not differ at depths of 0.9 m and 1.2 m. In contrast, in-row soil moisture was significantly higher under clean cultivation compared with nodding needlegrass at 1.5 m. Betweenrow soil moisture was significantly higher under clean cultivation compared with nodding needlegrass at every depth. Combining in-row and between-row data, overall vineyard soil moisture was slightly lower, by 1.2% points, in the nodding needlegrass treatment compared with clean cultivation. There was no interaction between treatment and depth for between-row soil moisture, indicating that the vines used little water from the between-row area. The lack of difference between treatments in the rate of soil moisture depletion over the season indicates that nodding needlegrass used little water during the summer. Based on these results, nodding needlegrass appears to be suitable as a permanent cover crop in California drip-irrigated vineyards where competition for summer water is a concern. Cover cropping in California vineyards is system, between-row floor vegetation is typ­ recognized as having multiple management ically managed during the growing season challenges, chief among them water use through cultivation or mowing (McGourty (Ingels et al., 1998). California has a Mediand Christensen, 1998). The negative conse­ terranean climate with clearly defined rainy quences of frequent cultivation are that it and dry seasons. Average annual rainfall in disturbs floor vegetation root channels and the San Joaquin Valley city of Fresno is ;300 accelerates the rate of organic matter decom­ mm with 82% falling in the off-season (1 position, which collectively can lead to poor Nov. to 31 Mar.), and the vast majority of water penetration (Gulick et al., 1994). Mainvineyards in the San Joaquin Valley are taining and managing floor vegetation through irrigated. The use of drip irrigation has mowing is much less disruptive to the soil and become increasingly common, and under this therefore can promote soil colloid aggrega­ tion, increase soil pore size, reduce compac­ tion, and improve water infiltration (Celette Received for publication 2 Dec. 2009. Accepted et al., 2005; Goulet et al., 2004; Klik et al., for publication 22 Jan. 2010. 1998; McGourty and Christensen, 1998). I thank the University of California Sustainable However, there is concern about the amount Agriculture Research and Education Program, of water used by the floor vegetation and how which provided funding for this study. it might affect vine growth and/or yield. Thanks also to laboratory assistants Jose Cantu, In regions with a Mediterranean climate, Abebe Gebreheiwet, Kimberly Miyasaki, Juliet Schwartz, and Jonathan Wroble for their data annual cover crops are typically winter ancollection efforts; to Ron Brase of California Ag nuals, which are planted in fall and senesce Quest for assistance with the neutron probe readby late spring. Because their growth is ings; and to Scott Stewart of Conservaseed (Rio primarily in the off-season, and their water Vista, CA) for project inspiration and for providing source is primarily rainfall, direct competi­ the native grass seed. I am also grateful for the tion with the grapevines for water is miniassistance of UC Cooperative Extension advisors mized. However, after senescence in spring, Chuck Ingels, Tim Prather, Kurt Hembree, and Dan weeds will take the place of the cover crop Munk. and must be managed to prevent excessive Current address: Horticulture and Crop Science Department, Cal Poly State University, 1 Grand competition. An alternative is the mainteAvenue, San Luis Obispo, CA 93407. nance of a perennial cover crop, which proe-mail [email protected]. vides similar benefits as an annual cover crop while avoiding the drawbacks of invasive weeds. However, all of the nonnative peren­ nials that have been tested as cover crops in California vineyards or orchards either require summer water or are competitive enough with the vines to decrease vigor or yield (Gulick et al., 1994; Ingels et al., 2005; Prichard et al., 1989; Wolpert et al., 1993). Two studies have evaluated California native grasses in vineyards (Baumgartner et al., 2008; Ingels et al., 2005), and neither found a negative effect on grape yield. Several studies have looked at how vine­ yard cover crops affect soil– and plant–water relations. Celette et al. (2005), working in a nonirrigated vineyard in Languedoc-Roussillon, France, found that soil water content with a tall fescue (Festuca arundinacea Shreb.) cover crop was higher at a depth of 0.75 m from spring to midsummer, but at other depths, soil water was equivalent to clean cultivation. Gulick et al. (1994) looked at continuous floor vegetation (Bromus hordea­ ceus L. subsp. hordeaceus in the winter fol­ lowed by mowed weedy vegetation in the summer) in a San Joaquin Valley, furrowirrigated vineyard. They found that the cover increased soil water infiltration by more than twofold, but also increased between-row soil water depletion compared with cultivation. King and Berry (2005), working in a dripirrigated vineyard, found higher betweenrow soil moisture with a Trifolium fragiferum L. cover crop versus a blend of California native grasses: Elymus glaucus Buckley, Hordeum brachyantherum Nevski, and Bro­ mus carinatus Hook. & Arn. However, this study limited soil moisture measurements to the top 10 cm of soil and did not compare the cover crops to clean cultivation. The intent of the present study was to evaluate the potential for California native grasses as vineyard cover crops, and further­ more, how one species, nodding needlegrass, affects soil– and vine–water relations. Vine vigor and yield data of nodding needlegrass and other cover crops in this study are presented in another paper (Costello, 2010). The native range of nodding needlegrass extends from the Sacramento and San Joa­ quin Valleys through the coastal regions of central and southern California (Beetle, 1947). Little biological information is avail­ able for this species, and most of what exists pertains to characteristics such as seed, leaf, and inflorescence morphology, and ecologi­ cal adaptations such as drought tolerance or geographical distribution (Amme, 2003; Barkworth and Torres, 2001). Observations suggest that it has a high degree of summer dormancy, remaining dormant midsummer even in the presence of available soil mois­ ture. If this is the case, then it should provide the advantages of a perennial cover crop without the disadvantage of excessive com­ petition with the vines for water. Clary (2006) found that nodding needlegrass exhibited a low summer cuticular transpiration rate, suggesting a drought tolerance mechanism. This study evaluated summer soil moisture and vine leaf water potential patterns with HORTSCIENCE VOL. 45(4) APRIL 2010 621 Fig. 1. Mean in-row soil moisture, 1998 season. Percent volumetric soil moisture content ± SE estimated from neutron probe readings taken weekly from June to September, nodding needlegrass versus clean cultivation treatments. Fig. 2. Mean in-row soil moisture 1999 season. Percent volumetric soil moisture content ± SE estimated from neutron probe readings taken weekly from June to September, nodding needlegrass versus clean cultivation treatments. nodding needlegrass as a cover crop com­ pared with clean cultivation. Materials and Methods The experimental site was at the University of California Kearney Agricultural Center in Parlier, in the San Joaquin Valley, in a warm climatic region, classified as a Viticultural Region V (Winkler et al., 1974). Mean high and low temperatures in August are 35 and 16 °C, respectively. The vineyard was a 0.4-ha drip-irrigated block, cv. Barbera, planted in 1989, with 3 m between rows and 2.1 m between vines within the row. Soil type at the site was a Hanford series fine sandy loam. This study’s focus on nodding needlegrass and its effect on water relations was part of a larger experiment on vineyard cover crops (Costello, 2010). Cover crop treatments were established in Nov. 1996, and nodding needlegrass was planted at a rate of 13.2 kg·ha equivalent. Plot size was five rows by six vines (189 m), and the cover crop treatments and a clean cultivated control were replicated three times in a randomized complete block design. The between-row width planted to the cover crops was 2 m, leaving a 1-m band in-row treated with herbicide. Neutron probe tubes made of poly­ vinyl chloride were placed in each nodding needlegrass and control plots, one within the row and one between rows (i.e., a total of three in-row and three between-row neutron probe tubes for each treatment). The in-row probes were placed midway between drip emitters. The vineyard was irrigated daily at 80% of estimated crop evapotranspiration (ETc) from 1 May to 1 Nov. of each year, which optimizes yield according to Williams (2000). Daily reference evapotranspiration figures were taken from the California Irri­ gation Management Information System weather station located on-site and monthly crop coefficient values from Williams et al. (2003). Rainfall outside of the study season (1 Oct. to 31 May) was 432 mm for 1997– 1998 and 202 mm for 1998–1999. Estimated total irrigation water applied was 503 mm (1998) and 538 mm (1999). The only other water inputs during the study periods were ;15 mm of rain (between 6 and 12 June 1998), and an accidental flood irrigation, which provided the entire block, with ;12 mm of water (6 Aug. 1999). Vine water status was estimated biweekly from these same treatments using a pressure bomb (PMS Instruments, Corvallis, OR), taking five readings per plot between the hours of 1100 and 1400 HR. Leaves selected for measurement were mature and in full sun. For each leaf, the petiole was cut with a razor blade, the entire leaf was placed into a plastic bag, and placed into the chamber within 30 s. Williams and Araujo (2002) found that bag­ ging after cutting the petiole resulted in lower readings than if the leaf were bagged before cutting (;12% lower at an irrigation regime of 100% ETc). Soil moisture status was estimated using a neutron probe (Model 503DR; Campbell HORTSCIENCE VOL. 45(4) APRIL 2010 622 Fig. 3. Mean between-row soil moisture, 1998 season. Percent volumetric soil moisture content ± SE estimated from neutron probe readings taken weekly from June to September, nodding needlegrass versus clean cultivation treatments. Fig. 4. Mean between-row soil moisture, 1999 season. Percent volumetric soil moisture content ± SE estimated from neutron probe readings taken weekly from June to September, nodding needlegrass versus clean cultivation treatments. Pacific Nuclear, Martinez, CA). Readings tubes within the vine row and between rows were taken every 2 weeks from May through in each plot. Volumetric soil moisture conSeptember at depths of 0.3 m, 0.6 m, 0.9 m, tent was determined from soil core samples 1.2 m, and 1.5 m from the neutron probe taken at the beginning of the study from each of the probe locations. In all, 18 core samples were taken, and neutron probe readings were taken at the same time. The neutron probe was read as counts during 30 s. The soil cores were weighed, oven-dried at 100 °C for 48 h, and then weighed again. These values were regressed against actual neutron probe read­ ings at each location to get an estimate of percent volumetric soil moisture content. The regression equation was y = –3.0171 + 0.0016x (r = 0.73, P < 0.01), in which y = volumetric soil moisture content (% w/v) and x = neutron probe reading. Cover crop management consisted of one mowing to a height of 15 cm in the middle of June of each year, and between-row weed control in the clean cultivated treatment was undertaken every 2 weeks by use of a tractordrawn rototiller. For the entire study site, inrow weed control was accomplished by a band-spray application of glyphosate to the 1 m in-row area (2.6 kg glyphosate acid equivalent/ha) in March and May. Soil moisture and leaf water potential were analyzed by repeated-measures analysis of variance (ANOVA) using year or date as the repeated-measures (‘‘within-subjects’’) vari­ able using orthogonal contrasts for mean sep­ aration (PROC GLM; SAS Institute, 2003). Differences between means were considered significant with P < 0.05. Regression analysis was performed by regressing soil moisture readings (pooling dates and years) against soil depth for in-row and between-row locations (PROC REG; SAS Institute, 2003). The between-row data were best fitted to a secondorder polynomial, whereas the in-row data were best fitted to a third-order polynomial. Linear regression analysis was performed on the between-row data only to estimate the rate of change in soil moisture over the season (PROC REG; SAS Institute, 2003), and t tests were calculated to compare the slopes between treatments at each soil depth.