Experiment Station. Department of Soil and Crop Sciences

Size: px

Start display at page:

Download "Experiment Station. Department of Soil and Crop Sciences"

Jemima Shaw
6 years ago
Views:

1 Tehnial Bulletin TB06-03 August 2006 Ag riultural Experiment Station College of Agriultural Sienes Department of Soil and Crop Sienes Western Colorado Researh Center Organi Alternatives for Weed Control and Ground Cover Management: Effets on Tree Fruit Growth, Development and Produtivity

2 Organi Alternatives for Weed Control and Ground Cover Management: Effets on Tree Fruit Growth, Development and Produtivity Ron Godin 1, Steve Ela 2, Shane Max 3, Kim Shultz 4, and Jim Rohde 4 A Cooperative Projet of Colorado State University Agriultural Experiment Station Western Colorado Researh Center Rogers Mesa Hothkiss, Colorado and Silver Sprue Orhard Hothkiss, Colorado 1 Researh Sientist, Colorado State University, Western Colorado Researh Center at Rogers Mesa, Hothkiss, Colorado 2 Farmer/Owner, Silver Sprue Orhard, Hothkiss, Colorado 3 Former Manager, Western Colorado Researh Center, Grand Juntion, Colorado 4 Researh Assoiate, Western Colorado Researh Center at Rogers Mesa Aknowledgements The authors would like to thank the Organi Farming Researh Foundation of Santa Cruz, California, for supporting this researh. Mention of a trademark or proprietary produt does not onstitute endorsement by the Colorado Agriultural Experiment Station. Colorado State University is an equal opportunity/affirmative ation institution and omplies with all Federal and Colorado State laws, regulations, and exeutive orders regarding affirmative ation requirements in all programs. The Offie of Equal Opportunity is loated in 101 Student Servies. In order to assist Colorado State University in meeting its affirmative ation responsibilities, ethni minorities, women, and other proteted lass members are enouraged to apply and to so identify themselves.

3 Abstrat Prodution of organi fruit in western U.S. has been expanding for the past deade or more. Fresh market sales aount for a majority of produer s inome. In order to gain market aeptane and reeive prie premiums for their organi fruit, produers must grow large, flavorful, high quality fruit. Weeds ompete strongly with young trees for water and nutrients, and there has been muh researh on the negative effets of weed pressure on young trees. However, little researh has been done on the effets of weeds on older, mature orhards. The objetives of this researh were to determine what, if any, effets weeds have on mature orhards and are one or more weed ontrol methods more onduive to quality fruit prodution than others. This on-farm researh was onduted at Silver Sprue Orhards near Hothkiss, CO, on a nine-year-old Gala apple orhard with miro-sprinkler irrigation. This study inluded seven treatments with eight repliations. The seven treatments were: 1) mowing (M), (2) propane weed flamer (F), 3) a weed barrier landsape fabri (LF), 4) shredded paper mulh (P), 5) mowing with material thrown into the tree row (M&T), 6) shredded bark mulh (B), and 7) farmer s favorite (FF), where no weed treatments were imposed and weeds were allowed to grow throughout the season. Results show that the mulh treatments (P and B) did suppress weeds and improve yield over the FF (ontrol). However, three years of drought during the study may have skewed results towards treatments that benefited soil moisture retention, namely the mulhing treatments, rather than as a diret effet of weed suppression only. The P treatment trees also retained higher fruit numbers until harvest and had the highest yields in two of the three years of the study. However, the M treatment only showed signifiantly lower yields in the third year and is muh less labor intensive than manual appliation of the P treatment. Although the FF treatment showed signifiant impats on yield due to unheked weed pressure the M treatment appears to be a viable alternative over the P treatment. The M treatment yields were omparable to the P treatment yields in two of three years and do not inur the high labor ost of manual appliation of the P treatment. Introdution Organi fruit prodution in the US, espeially the western regions, is expanding. The inrease is ourring for both eonomi and eologial reasons. Current market onditions ditate that organi apple growers produe large, flavorful, high quality fruit. Large, high quality fruit reeive prie premiums and market aeptane whereas small fruit an be diffiult to sell, even at lower pries. To grow large fruit, trees must be unstressed and provided with adequate water and nutrition. Weeds an ompete with fruit trees for both water and nutrients. Researh has demonstrated that weed ompetition in young fruit trees redues tree growth and nutrient and water use effiieny, and therefore dereases fruit prodution and fruit size (Merwin and Stiles, 1994). Over time, redued tree growth redues tree volume and potential prodution. Thus, a standard orhard pratie is to ontrol weeds during the establishment and early growth of an orhard. However, the effet of weed ompetition on prodution and fruit size of mature fruit trees has not been studied. Most experiments are onduted on young trees beause stress on young trees redues yearly prodution potential for the life of the tree, whih ould ost growers tens of thousands of dollars in inome over the year lifespan of the tree. Most

4 ommerial fruit, inluding both peahes and apples, are produed on mature trees. Thus it is important to understand the effets or lak of effets that weeds may have on a mature tree. This information ould have signifiant impat on how orhards are managed along with the possibility of signifiantly reduing prodution ost. Currently, organi growers spend onsiderable time and money ontrolling or removing weeds from their orhards based primarily on the researh trials in young orhards. If weeds have only a minor effet on fruit size in mature trees, this time and money ould be redireted to other parts of the operation. If weeds do have an effet, then the grower needs to know if one means of weed ontrol is more effetive than another. This study investigated the effets of several different weed ontrol methods on fruit yield and size in mature apple trees. The information generated will give organi growers better knowledge as to how to manage weeds while produing large, marketable fruit. Materials and Methods This on-farm researh was onduted at Silver Sprue Orhard (SSO), near Hothkiss, Colorado. The researh was onduted on a ommerial, ertified organi blok of nine-year old Gala apples on EMLA 26 rootstok on an Aqua Fria lay loam soil [fine-loamy, mixed, mesi, Typi Haplargid]. The experimental design is a randomized omplete blok with seven treatments and eight repliations; the experimental blok had a border row of apple trees on either side. Plots onsisted of five onseutive trees where treatments were applied. Within eah plot, the three enter trees were used for data olletion with the two outside trees in eah plot serving as buffer trees. The seven different treatments were applied in the tree row. The tree row onsisted of a six foot wide strip, three feet on either side of the tree trunk. The trees are planted six feet apart within the row and there are 13 feet between rows. The weed ontrol treatments were: 1) mowing (M), (2) propane weed flamer (F), 3) a weed barrier landsape fabri (LF), 4) shredded paper mulh (P), 5) mowing with material thrown into the tree row (M&T), 6) shredded bark mulh (B), and 7) farmer s favorite (FF), where no weed treatments were imposed and weeds were allowed to grow throughout the season. The F and both mowing treatments were applied approximately every two weeks to one month as needed during the growing season. Mulhes were renewed or replenished eah spring in the tree row to a depth of approximately six inhes. The P mulh onsisted of shredded paper reyled from a loal bank. The B mulh onsisted of oarse bark from a loal lumber mill. For the LF treatment, the fabri was removed, fertilizer applied, and the fabri replaed. The experimental plots were established during the summer of 2000 with data olletion from 2001 through Commerial organi fertilizer (12-0-0, derived from feather meal) was applied eah spring at the rate of 25 lbs of nitrogen (N) a -1. In the treatments where mulhes were applied (P and B) what remained of the mulhes in the seond and third springs was raked aside prior to fertilizer appliation. Following fertilizer appliation the old mulh was then raked bak into the tree row over the organi fertilizer and new mulhes applied. Trees were pruned eah winter by professional orhardists. Approximately two weeks after bloom, fruit were thinned to an approximate equivalent number of fruit per tree to establish a onsistent starting rop load aross all treatments. The orhards were 2

5 miro-sprinkler irrigated every five to ten days as needed during the growing season. Data was olleted for weed density, fruit yield and quality and tree growth. Weed density, or the perentage of the tree row overed by weeds, was estimated on all plots prior to F and mowing treatments during the growing season, May through August. The weed density was then averaged over the growing season for eah treatment. Fruit yield and quality was determined by ounting and weighing the fruit from eah of the three data trees within eah plot at harvest. Average fruit size was alulated from total fruit weight and fruit number, as a measure of fruit quality. Tree growth was determined by measuring the irumferene of eah of the three data trees in eah plot and alulating the total trunk ross-setional area (TCA) at six inhes above ground level. Tree growth measurements were taken prior to the initiation of the study and following the onset of dormany eah fall. Tree growth was evaluated on inrease in TCA between the initial measurement and final measurement following the 2003 harvest. Data was analyzed using the general linear model with a least signifiant differene level of 0.05 (SAS Institute, 2001). A permanent weather station is loated at the Rogers Mesa Researh Station (RM) (approximately one mile south SSO) where data is downloaded to a omputer daily. A summary of average in-season limati data an be found in Table 1. For evaluation of the weather data, the growing season was defined as April 1 st to August 31 st. All three years of the study were onsidered drought years due to well below average in-season preipitation (Table 1) and below average annual preipitation (data not shown). However, of the three drought years, 2002 was onsidered a severe drought year not only beause of the very low preipitation amounts but also beause of the extreme dryness (Min relative humidity (RH)) and extremely high maximum temperatures (Table 1). Average in-season preipitation at RM over the past 20 years is 4.85 inhes and average maximum high temperature is approximately 80 F (Table 1). In-season preipitation for both 2002 and 2003 was ½ inh or less and average maximum temperatures were above 85 F and 83 F, respetively. Table 1. Average in-season limate data Year MaxTemp ( F) Min Temp ( F) Vapor Pres. (mb) Preip. (in) Min RH (%) GDD Ref ET (in) yr avg GDD = Growing degree days, based on the averaging method of alulating GDD with a maximum temperature utoff of 97F and a minimum utoff temperature of 41F, standard for apples. Ref ET = Referene evapotranspiration, omputed using an alfalfa referene equation know as the 1982 Kimberly-Penman method, Wright, The average oeffiient for apples from full bloom to harvest is 72% of Ref ET (U.S. Bureau of Relamation website - AgriMet Crop Coeffiients: Apples at 3

6 Results & Disussion Results Tree Growth Treatments showed no signifiant differenes in tree size at the start of the study in the spring of 2001, average TCA was 29.4 in 2. Over the three years of the study there were no signifiant differene in tree growth among any of the treatments and the average TCA in the fall of 2003 was 48.0 in 2. These data show that the treatments applied had no influene on tree growth. The lak of response may be due to equal amounts of N fertilizer appliations aross all treatments, whih may have the most influene on tree growth and the mature stage of the trees. Weed Density The weed density data shows a signifiant treatment by year interation; therefore, data was analyzed separately by year (Fig.1). Weed density was also affeted by the severe drought in 2002 as an be seen in the FF treatment where weed density was near 100% in 2001 and However, weed density dereased to 65% in 2002 (Fig. 1). In 2001, the M, FF and M & T treatments had signifiantly higher weed density than the other treatments, with the LF treatment having the fewest weeds. This would be expeted, as the LF is a weed barrier with weeds only growing near tree trunks were the LF has slight openings (Fig. 1). In 2002, although weed density was muh lower than the previous year for all treatments, the M and FF were signifiantly higher and the LF signifiantly lower than the other treatments (Fig. 1). In 2003, the weed density in the FF treatment was signifiantly higher than in all other treatments and the LF had a signifiantly lower weed density. In all treatments exept FF, weed density was less eah subsequent year, probably due to the umulative effet of the treatments over the three years of the study. Although plots were irrigated on a regular basis, the extremely low relative humidity and very high maximum temperatures probably ontributed to the redued weed density in Overall the LF, P and B treatments showed the best weed suppression. The M, F and M&T treatments showed definite improvement from the first to third years of the study. The F, LF, P, and B treatments are more labor intensive than the other treatments and if yield or fruit quality does not inrease to off-set the labor osts of these treatments, ommerial fruit growers will not adapt the pratie. 4

7 Figure 1. Weed Density, Weed Density a* a b b b d d e Mow Flamer Land Fab Paper F Fav M & T Bark Treatment a a a a b d * Letter followed by a different letter in eah year indiates signifiant differenes between means at P < Fruit Number Fruit number showed a signifiant treatment-by-year interation; therefore, fruit numbers were analyzed individually by year (Table 2). Following bloom, fruit was thinned to approximately the same number of fruit per tree, however, the trees in some treatments had dropped signifiant numbers of fruit by harvest. This may be due to rapid soil moisture depletion following irrigation. In 2001, the trees in the P treatment had a signifiantly higher fruit number than the M, LF, FF and M & T treatments but were not signifiantly higher than the F or B treatments. The trees in the mulh treatments are retaining more fruit until harvest, possibly due to better soil moisture retention than the other treatments. The retention of higher numbers of fruit in the F treatment may be due to redued weed ompetition for water and nutrients. However, these results appear to be an anomaly, as the results are not onsistent with results for the remainder of the study (Table 2). Although the LF virtually eliminates weed pressure on the trees, the LF is blak and is possibly ausing higher soil temperatures thereby reduing available soil moisture. In 2002, there were no signifiant differenes in fruit number among any treatments. This is probably due to limati stresses mentioned above. In 2003, the P treatment had signifiantly higher fruit numbers than any other treatments. The B treatment had the seond highest number of fruit although not signifiantly higher than the other treatments. As in 2001, the P treatment retained the highest number of fruit and exept for the LF had the lowest weed densities. In drought years, weed suppression ombined with improved soil moisture retention afford by the P treatment may be the best way to retain fruit on the trees until harvest. 5

8 Table 2. Fruit number per tree, Fruit number fruit tree Treatment Mow 62 b* 68 a 72 b Flamer 79 ab 61 a 59 b Landsape Fabri 55 d 75 a 67 b Paper Mulh 86 a 56 a 96 a Farmer s Favorite 44 d 65 a 66 b Mow & Throw 34 d 56 a 60 b Bark Mulh 65 ab 52 a 74 b * Letter followed by a different letter in eah year indiates signifiant differenes between means at P < Fruit Weight For this study we used fruit weight as an indiator of fruit quality, with the assumption that heavier fruit are larger and more marketable. The data indiate that there was a signifiant treatment-by-year interation; therefore, treatments were analyzed by year. This data shows that no treatment onsistently produed signifiantly higher fruit weight than any other treatment over the three years of the study (Table 3). However, fruit weight was signifiantly different eah year. The highest fruit weights were in 2001 and the least in the severe drought year of Presumably, the limate and growing onditions in eah partiular year had a more signifiant effet on fruit weight than any of the treatments imposed (Table 3). In 2001, the LF treatment had signifiantly higher fruit weight than the F treatment, while all other treatments were not signifiantly different than either treatment. In 2002, the M treatment had signifiantly higher fruit weight than the FF treatment. In 2003, there were no signifiant differenes in fruit weight. Table 3. Fruit weight, Fruit Weight g fruit Treatment Mow 165 ab* 137 a 133 a Flamer 149 b 121 ab 133 a Landsape Fabri 166 a 117 ab 135 a Paper Mulh 155 ab 116 ab 129 a Farmer s Favorite 155 ab 111 b 126 a Mow & Throw 156 ab 117 ab 123 a Bark Mulh 161 ab 121 ab 123 a * Letter followed by a different letter in eah year indiates signifiant differenes between means at P <

9 Fruit Yield Fruit yields also showed a year-by-treatment interation and varied by year (Table 4). In 2001, the P treatment yielded more than all other treatments but not signifiantly higher than the M, F or B treatments. This is due to the ombination of a higher number of fruit retained on the trees until harvest and a high fruit weight for these four treatments. Although yields of the M, F and B treatments were not signifiantly lower than the P treatment statistially, the F treatment yielded 1868 lbs a -1 less than the P treatment. This differene may not be statistially signifiant, but it is eonomially signifiant. If the grower s inome from the additional 1868 lbs is approximately $0.50 lb -1, the grower would realize approximately $1000 more inome per are. In 2002, there were no signifiant yield differenes in any of the treatments (Table 4). This is probably due to the severe drought onditions that stressed the trees in general. The RH was very low and maximum temperatures were very high during the growing season in 2002, whih probably had a major effet on yield regardless of irrigation water applied. In 2003, the P treatment yielded signifiantly higher than all other treatments (Table 4). Fruit weights were not signifiantly different in the P treatment in 2003 but fruit number were signifiantly higher, aounting for the signifiantly higher yields. Table 4. Fruit yield, Yield lbs a Treatment Mow ab* a b Flamer ab 9090 a b Landsape Fabri bd a b Paper Mulh a 7926 a a Farmer s Favorite 8453 d 8969 a b Mow & Throw 6556 d 8157 a 9101 b Bark Mulh ab 7759 a b * Letter followed by a different letter within eah year indiates signifiant differenes between means at P < Disussion The M treatment yields were not signifiantly less than the P treatment exept in the third year of the study. Mowing is a standard pratie in most organi orhards and not as labor intensive as appliation of the P treatment and the labor needed is spread over the ourse of the growing season for the M treatment. The appliation of the P treatment requires large amounts of labor early in the season, as it must be hand applied when other orhard tasks are a high priority, suh as fruit thinning. The F treatment also did not yield signifiantly less than the P treatment in the first two years of the study. However, the rising osts of propane in the past year or so has made the F treatment ost prohibitive for most growers. The LF treatment yielded signifiantly less than the P treatment in two of the three years of the study. The lower yields oupled with the labor osts of removing the LF, applying fertilizer and reapplying the LF make this treatment ost prohibitive. It has sine been learned that if the LF is not removed in the fall the LF makes for good winter 7

10 habitat for mie, that tend to girdle fruit trees as a winter food soure in the LF treatments (Steve Ela, et al. personal ommuniation). The FF treatment had onsistently low yields throughout the study, indiating that unmanaged weeds do have an impat on mature trees likely by ompeting for water and nutrients. The M & T treatment did not yield well in any year; the reason for this is not understood at this time. The B treatment yielded well in the first and third years of the study but ost of material and labor for bark appliation, without a orresponding signifiant inrease in yield does not make this treatment ost effetive. Conlusions The results from this study are not definitive, probably due to drought onditions, and may also suggest that a three-year study may be too short in duration to onlusively determine meaningful outomes from imposed treatments for organi perennial systems. The mitigating fator of limate appears to play a larger role in fruit tree prodution in drought years than imposed treatments. The data do show that the P mulh treatment redued weed density and hene, weed pressure on the orhard, whih likely led to the higher yields although the data are not onsistent for the B mulh treatment. In the third year of the study, the P treatment produed signifiantly higher yields but this is by no means onlusive. However, one onlusion that an be made is that organi perennial agriultural systems are highly buffered and very resistant to large hanges over the short term, whih is also the strength of organi systems ompared to onventional systems. Sine the M treatment is the standard pratie for organi orhards and in light of the results of this study, growers will likely ontinue with this pratie. Referenes Merwin, I.A., and W.C. Stiles Orhard groundover management impats on apple tree growth and yield, and nutrient availability and uptake. J. Am. So. Hort. Si., 119(2): Wright, J.L New evapotranspiration rop oeffiients. J. of the Irrig. and Drain. Div., ASCE, 108(IR2): SAS Institute SAS/STAT user s guide. Release 8.2. SAS Institute, Cary, N.C. 8

Camber Variability in Prestressed Concrete Bridge Beams

CONCRETE BRIDGE TECHNOLOGY Camber Variability in Prestressed Conrete Bridge Beams by Dr. Maher Tadros, econstrut Beams ast with extra amber in storage yard at Conrete Tehnology Corporation; amber shown