Azolla (Azolla filiculoides) is a floating fern that maintains a symbiotic relationship with an N-fixing blue-green algae. In many parts of Asia, azolla is used as a green manure in flooded rice cultivation. Taro (Colocasia esculenta) grown under flooded conditions is used to produce a traditional Hawaiian staple, poi. Azolla has been present in Hawaii for many years, but is not used in a controlled way for either nutrient augmentation of production sites or weed suppression. In this experiment, azolla was removed from a stream on the island of Kauai and multiplied in a nursery pond. Phosphoric acid was added to the nursery pond as a nutrient (P = 5 ppm) at 5-day intervals to accelerate azolla growth. Azolla was moved from the nursery pond and added to taro production plots at a seeding rate of 488 kg·m–2. Phosphoric acid was used in production plots to hasten coverage of the water surface by azolla. Ten days after azolla inoculation, production plots were covered and taro seed pieces were planted. Weed dry weights from conventional and azolla covered plots were recorded 91 days after taro planting. Taro corms were harvested 315 days after planting. Weed dry weight in azolla plots was 86% less than conventional plots. Azolla delayed taro maturity, causing a 41% reduction in marketable corm yield.
DeFrank and Charles R. Clement
Pejibaye (Bactris gasipaes, Palmae) is being evaluated for palm heart production in Hawaii. Counts of parasitic nematodes and yields at 18 months and weed control were evaluated in response to: Arachis pintoi, Cassia rotundifolia cv. Wynn, Desmodium ovalifolium, Chloris gayana, and woven black polypropylene mat. Four open-pollinated progenies from the Benjamin Constant population of the Putumayo landrace were used as replications. Twenty five percent of the plants were harvested, with means of 5, 20, 15, 15, and 70%, respectively. Individual heart weights did not vary significantly among treatments (mean = 169 g). Actual yields were significantly different, with means of 31, 125, 92, 99, and 440 kg/ha, respectively. All vegetative ground covers competed with pejibaye for nutrients, which explains the harvest percentages and yields. D. ovalifolium and C. gayana provided acceptable weed control. A. pintoi provided good ground cover, but reduced weed control.
James Leary and Joe DeFrank
An important aspect of organic farming is to minimize the detrimental impact of human intervention to the surrounding environment by adopting a natural protocol in system management. Traditionally, organic farming has focused on the elimination of synthetic fertilizers and pesticides and a reliance on biological cycles that contribute to improving soil health in terms of fertility and pest management. Organic production systems are ecologically and economically sustainable when practices designed to build soil organic matter, fertility, and structure also mitigate soil erosion and nutrient runoff. We found no research conducted under traditional organic farming conditions, comparing bareground monoculture systems to systems incorporating the use of living mulches. We will be focusing on living mulch studies conducted under conventional methodology that can be extrapolated to beneficial uses in an organic system. This article discusses how organic farmers can use living mulches to reduce erosion, runoff, and leaching and also demonstrate the potential of living mulch systems as comprehensive integrated pest management plans that allow for an overall reduction in pesticide applications. The pesticide reducing potential of the living mulch system is examined to gain insight on application within organic agriculture.
Kelvin Sewake and Joseph DeFrank
Anthurium bacterial blight is a disease that has devastated much of the Hawaii anthurium industry in recent years. In response to the crisis, the Cooperative Extension Service produced a video entitled “Strategies for Anthurium Blight Control - A Growers' Discussion.” The video format was selected to allow the extension service to utilize experience of four successful large scale commercial growers and deliver blight control information to other growers with a graphic demonstration on control procedures for disease management. Production of the video eliminated the need for farm tours, eliminated the risk of disease spread, condensed information, and allowed growers to borrow copies. This video proved effective in prompting growers to adopt and implement recommended blight control procedures.
Osamu Kawabata and Joseph DeFrank
A modified power function, y = (A + B·x)–C, was developed for determining the relationship between plant growth and growth retardant treatment. This function accounts for the plant response characteristics by incorporating three coefficients: A, growth level of the nontreated plants; B, the degree of growth reduction; and C, the smallest effective dose of the growth inhibitor. The function accounted for 97% of the variation in purple nutsedge (Cyperus rotundus L.) leaf length as a function of the amount of a growth retardant applied. The procedure resulted in a smaller error sum of squares than several common nonlinear functions because of its greater shape flexibility.
O. Kawabata and J. DeFrank
Purple nutsedge is a difficult weed to eradicate due to extensive underground growth. One eradication strategy is to inhibit tuber formation for preventing reproduction, In developing this strategy, soil applications of paclobutrazol (PA) were conducted to suppress rhizome and tuber development in Hawaii. Factors examined were 4 PA levels at 0, 0.5, 2 and 8 mg ai·liter-1 (medium) and 4 application types: a) soil drench, fresh tuber, b) soil incorporation, fresh tuber, c) soil drench, stored tuber for 1 month at 4C, and d) drench to a synthetic medium (vermiculite: perlite = 2: 1 by volume), fresh tuber. Treatment design was 4×4 factorial with 5 replications, and experimental unit was a tuber planted 2 cm deep in a 1-liter plastic container. Two months after planting tuber numbers and rhizome lengths were recorded and analyzed for regression and orthogonal comparisons. Control means were 11.1 tubers and 1178 mm total rhizome length. PA reduced both measurements as no tuber was formed at 2 or 8 mg ai·liter-1 and total rhizome lengths were limited to 8 and 3 mm, respectively. PA was less effective in the synthetic medium than in the soil, while application methods and tuber storage did not result in significant differences.
Osamu Kawabata and Joseph DeFrank
Hector R. Valenzuela and Joseph DeFrank
Living mulches offer a low-input alternative to achieve weed control while minimizing herbicide applications, decreased fertilizer leaching, insect and nematode management. and improved soil texture. A study was conducted to evaluate the effect of a Rhodes Grass (Chloris gayana cv. Katambora) living mulch on the growth and productivity of ten eggplant. Solanum melongena, cultivars grown under fertigation. The living sod was established at the Univ. Hawaii Waimanalo Experiment Station in June 1992. Soil analysis was taken before experiment initiation. Ten eggplant cultivars were transplanted on both living-mulch and control (woven-polyethelene mulch) plots on 4 March 1993. Weekly or bi-weekly harvests were conducted for six months. beginning on 19 May 1993. In addition plant height and canopy dimensions were determined on 16 April. and 10 Nov. Plant growth was monitored throughout the experiment. Soil samples were taken from the eggplant rhizosphere, hare-ground and in Rhodes grass monoculture, for nematode count determinations. Soil samples were also taken for nutrient determination after completion of the experiment. Overall yields were greater in the polyethelene mulch than in the living mulch plots. A differential response was observed on the response of cultivars to cropping system. However the most vigorous cultivars performed well in both systems. The living mulch system showed potential for nematode management in eggplant agroecosystems.
Joseph DeFrank and Charles R. Clement
Pejibaye (Bactris gasipaes, Palmae) is being evaluated for heart of palm production in Hawaii. Yields and weed control were evaluated in response to: oryzalin (4.5 and 9.0 kg ai/ha), oxyfluorfen (0.6 and 1.2 kg ai/ha), paraquat (1.2 and 2.4 kg ai/ha) and woven black polypropylene mat. Four open-pollinated progenies from the Benjamin Constant population of the Putumayo landrace were used as replications. Paraquat was sprayed at 50 day intervals, while the preemergence herbicides were sprayed at 90 day intervals. Harvest started at 18 months after planting out, 17 months after treatment initiation. The polypropylene mat yielded the highest percent harvest (80%), followed by Oxyfluorfen (50%), paraquat (20%), oryzalin (12.5%). There were replication (genotype) effects that suggest varying tolerance to paraquat and oryzalin. Estimated palm heart yields (3731 plants/ha), corrected for % harvest, were highest with polypropylene mat (490 kg/ha), followed by oxyfluorfen 1.2 ai (425 kg/ha) and 0.6 ai (330 kg/ha). Paraquat severely inhibited growth of the suckers that assure future harvests. The performance rating of these weed control treatments was: mat = oxyfluorfen > oryzalin > paraquat.
Joseph DeFrank and Charles R. Clement
Pejibaye (Bactris gasipaes Kunth, Palmae) is being evaluated for production of fresh heart of palm in Hawaii. Precocity, yields, and weed control were evaluated in response to woven black polypropylene mat (control), oryzalin, oxyfluorfen, and paraquat. Control plots attained 100% of plants harvested by 26 months, followed by oxyfluorfen (97.5%), oryzalin (77.5%), and paraquat (60%). Estimated heart of palm yields (3731 plants/ha) were similar with oxyfluorfen 1.2 kg a.i./ha (707 kg·ha–1), polypropylene mat (612 kg·ha–1), oxyfluorfen 0.6 kg a.i./ha (600 kg·ha–1), and oryzalin 4.5 kg a.i./ha (478 kg·ha–1). Based on precocity, yields, and weed control efficiency, the performance rating of these weed control treatments was mat ≈ oxyfluorfen > oryzalin > paraquat. Chemical names used: 4-(dipropylamino)-3,5-dinitrobenzenesulfonamide (oryzalin); 2-chloro-1-(3-ethoxy-4-nitrophenoxy)-4-(trifluoromethyl)benzene (oxyfluorfen); 1,1′-dimethyl-4-4′-bibyridinium ion (paraquat).