Use of Cover Crops with Medicinal Herbs in North Hawaii

Final Report for FW10-056

Project Type: Farmer/Rancher
Funds awarded in 2010: $20,117.00
Projected End Date: 12/31/2012
Region: Western
State: Hawaii
Principal Investigator:
Dr. Katherine Pomeroy
Kohala Medicinal Herb Farm
Expand All

Project Information


Three farms in the district of North Kohala, Hawaii Island planted test plots of five medicinal plants over a year long period as an on-farm trial of their objectives:

1) determining optimum stress conditions,
2) providing on-farm income, and
3) training local youth.

Extensive soil and plant data was collected and interpreted. Soil nutrient deficiencies were found in all the farms. Soil nutrient deficits are common in North Kohala due to the impacts of many decades as a sugar plantation. Curcumin, a medicinal constituent in turmeric, increased in the cover crop treatments in two of the three farms. This may be due to soil mining by the cover crop, by deficit irrigation conditions created by the cover crop and weeds or by another undetermined variable.

Over a dozen youth were engaged in the process, with three of those involved throughout. Community outreach was conducted via events and local media, community groups, a farmer’s market day and the field day. While the project did not result in on-farm income, it did provide some ground-work for future development of income streams, such as the incorporation of health/agricultural tourism. This project also gave insight into what is involved with working cooperatively, with good results in terms of social relationships. The project was successful in raising community awareness regarding growing and using medicinal herbs.


Three small farms (15-20 acres) located in North Kohala on the Island of Hawaii (20.1319 N, 155.7939 W) were planted with the same treatment plots. While the trials were set up to be as uniform as possible, a great deal of variation was inevitable. Although the farms are within five miles of each other, there is still a great deal of variation in rainfall and wind. The general area spans the northern tip of Hawaii Island, from windward to leeward, and all of the farms are located within two miles of the ocean at 400-600 feet elevation (see Figure 1).

Our research question was based on the well-documented fact that wild-harvested plants are often more potent than cultivated plants in terms of medicinal chemical constituents. The reason for the higher potency found in wild-harvested plants is believed to be due in part to the more stressful growing conditions found in the wild than under commercial cultivation, and to the relatively larger amount of medicinal constituents often found in wild or traditional species (Biodiversity and the Ecosystem Approach in Agriculture, Forestry, and the fisheries Conference. Impacts of Cultivation and Gathering of Medicinal Plants on Biodiversity, Oct 2002, Rome). The project also builds on the concept of improving crop quality and chemical constituent levels by stressing plants during their growth cycle. Research has shown that deficit-irrigation treatments might be effective to increase the chemical constituency and/or quality of herbs and horticultural crops (Khalida et al., 2010; Lopes et al., 2011; Masmoudia et al., 2010; Oh et al., 2010; Palesea et al., 2010; Patan; et al., 2011; Pernice et al., 2010; Wang et al., 2010)

Figure 1: Map of Hawaii Island and District of North Kohala

Project Objectives:

The objectives of this project were to:

1) Determine the optimum stress conditions for medicinal plant growth on five tropical medicinal plants (moringa, lemongrass, ashwaganda, turmeric and galangal ginger).

2) Provide additional on-farm income for small, tropical family farms through the marketing of tropical medicinals.

3) Provide local youth with inspiration and knowledge of sustainable agricultural practices through paid farm internships.


Click linked name(s) to expand
  • Karin Cooke
  • Dr. Richard Liebmann
  • Dr. Katherine Pomeroy


Materials and methods:
Field Trials Background

We artificially introduced stress by using a cover crop to compete with the medicinal plants for nutrients and especially water. When planted next to the medicinal plants, the cover crop, grown as an intercrop or living much, was predicted to compete for water and nutrients. The cover crops used were annual rye grass and, later in the trial, buckwheat. For both cover crops, the seed sprouted and grew but was quickly overtaken by local weed and pasture grass species. Our cover crop treatments thus became a mixture of cover crop and weedy species, and it was assumed that the cover crop/weedy mixture continued to compete for water and nutrients, as compared to the weed-free controls. We collected data on plant growth, soil analysis, tissue analysi, and curcumin constituent levels for all the farms. The results are outlined and interpreted below.

On each farm, three adjacent 20’ x 20’ test plots were established, planted with five medicinal plants: moringa (Moringa oleifera), lemongrass (Cymbopogon citratusis), ashwaganda (Withania somnifera), turmeric (Curcuma longa) and galangal ginger (Alpinia galangal). Sugar cane was used as a wind break. Plot 1 was the control, did not have a cover crop and was maintained as weed-free as possible. Plot 2 received the cover crop/weed/water deficit treatment (called the competitive treatment) for a period of four – six months. Plot 3 received the competitive treatment for eight – ten months. Thus there were three test plots per farm and a total of nine test plots for the entire field research project.

Uniformity in the field trials was implemented by using the same root and seed stock for all of the nine test plots (three per farm), installing irrigation, using the same cover crop seed, using the same treatments, the same timing for field treatments, the same youth labor pool and collection of the same data. Yet within these parameters it was impossible to avoid some variation as is typical for on-farm field trials (Valenzuela, H. pers. comm.). For instance, the irrigation water was sourced differently for all the farms, and Kokolulu’s system completely failed due to the wind. Also, there was a two month variation in the timing of treatments due to the schedules of the participants. Weed control also varied among the farms. It was particularly important to eliminate weed competition for the control plot 1 (no cover crop). Overall, the general uniformity of the plots among farms was reflected in the growth of the plants (Table 1). All three farms showed a clear gradation in plant growth with Plot 1 having the largest plants, Plot 2 the next largest and Plot 3 the smallest plants. We believe that this pattern shows that even with all the variables, we were able to achieve an acceptable degree of consistency in our methods.

Plant Growth Data

Plant height for the medicinal plants was similar or was increased in response to cover crop treatments, as compared to the weed and cover crop-free treatments. While the height of Moringa increased by an average of 20% with the eight – ten month cover crop treatment, the growth of the other medicinals was slightly decreased, as was the case with galangal, or slightly increased as was the case with lemon grass and ashwaganda (Table 1). The growth data indicates that these medicinal plants compete well with an intercrop, even under conditions of greater water competition.

Soil Analysis Data

The soil pH increased slightly in all farms, probably as a result of the organic amendment applications, applied prior to planting (Table 2). The average phosphorus, potassium, calcium, magnesium, organic matter (C) and total Nitrogen soil content increased in the cover crop treatment plots, as compared to the weed-free controls. The greater soil nutrient levels observed in the cover crop plots may be due the nutrient contributions provided by the cover crops, by mining nutrients from the soil, by creating a richer nutrient environment in the rhizosphere or by better utilizing the organic amendments that were applied prior to planting (Table 2). In general phosphorus levels were quite low in all farms. Soil phosphorus levels remained low in all treatment plots at the Kohala Medicinal Herb Farm, probably due to the low P levels in the amendments used at that site.

Tissue Analysis Data

Nutrient tissue analysis from leaf samples taken at the completion of the experiment were conducted for ashwaganda. Overall the weed-free control treatments had greater levels of Nitrogen, Potassium, Copper, Iron (on two farms) and Boron (only on one farm). Treatments that included competition with cover crops for eight – ten months (Plot 3, Table 3) had greater levels of Phosphorus (on two farms), Calcium (two farms), Magnesium (two farms), Iron (only on one farm), Manganese (two farms), Zinc (one farm) and Boron (two farms). Competition with a cover crop for four – six months (Plot 2) resulted in greater nutrient levels for Phosphorus (one farm), Calcium (one farm), Magnesium (one farm), Zinc (one farm), Copper (one farm) and Boron (one farm).

Overall, the cover crop treatments (four – ten months duration) had greater levels of Phosphorus, Calcium, Magnesium, Iron, Manganese, Zinc, Copper and Boron in one or more of the sites, as compared to the weed-free treatments (Table 3).

Curcumin Analysis Data

Curcumin constituent levels were determined for turmeric. For all locations pooled together, the weed-free controls (Plot 1) had the highest curcumin levels. However the data was affected by outlier results obtained from the KMHF Farm. When the results were separated by farm, curcumin levels were greatest for the control plots (Plot 1) only for the KMHF farm (Table 4). The higher levels of curcumin observed in the KMHF farm can be explained because ONE of the curcumin samples from the KMHF plants were mistakenly taken from the ‘mother’ root, while samples for the other farms were taken from the ‘daughter’ roots. Research conducted in India also observed higher curcumin levels in mature rhizomes compared to the younger ones (Nirajan et al 2003).

Curcumin samples taken from the Lokahi and Kokolulu farms showed that the highest curcumin levels were obtained from the eight – ten month cover crop treatments (Plot 3, Table 4), as compared to the four – six month cover crop treatments (Plot 2) or to the bare-ground/weed free controls (Plot 1, Table 4).

Data Interpretation

Based on the growth data and on field observations, all of the medicinal species evaluated appeared to be rather hardy and were able to grow well despite the relatively dry and windy growing conditions and despite the competition from the four to six or eight to ten month cover crops or living mulch treatments. The growth of the medicinal plants was either little affected, decreased or actually increased when interplanted with cover crops, as compared to the weed-free/bare ground controls. The ability to grow relatively well despite the dry and windy weather and competition was expected, given the hardy nature of these medicinal plants.

In terms of the soil fertility, nutrient levels seemed to have actually increased in the cover crop treatments, as compared to the weed-free controls. The increased fertility under the cover crops may be explained by a better mining of the soil by the four – ten month cover crops. The organic amendments applied on all farms and to all treatments may have decomposed better under the cover crops, and the cover crops may have been able to better uptake and recycle these nutrients, compared to the amendments applied to the bare-ground plots. In terms of differences between farms, the Phosphorus levels were 2.5 times lower in the KMHF Farm, as compared to the other two farms (average of 22.5 ppm P in KMHF farm vs. average of 56.5 ppm in the other farms). The considerable lower P levels in the KMHF farm may help to explain the differential level of constituents found between farms at the end of the experiment.

In terms of nutrient tissue levels, the control plots showed greater levels of Nitrogen and Potassium, mobile elements that may have been more available in the plots where there was no competition from the cover crops. In contrast several other less mobile nutrients such as Phosphorus, Calcium, Magnesium, and micronutrients such as Iron, Manganese, Zinc and Boron were found at greater levels in the cover crop treatments than in the weed-free control plots. The greater level of Phosphorus and micronutrients in the cover crop treatments may help to explain the trend toward higher medicinal constituent levels found in two of the three farms, in the cover crop treatments. Research in India and in Japan also attributed higher Phosphorus levels to increased Curcumin medicinal constituent levels in turmeric (Akamine et al., 2007; Niranjan et al., 2003), and in ashwagandha (Kubsad et al., 2009).

In two of the three farms, greater medicinal curcumin constituent levels were found in the eight – ten month cover crop treatments, as compared to the four – six month cover crops or to the weed-free controls. A mistake was made with one of the samples when collecting root tissue samples from KMHF farm for analysis of curcumin levels. When this outlier data is taken out (the first row in Table 4), the average curcumin levels in the Lokahi and Kokolulu farms were 4x as great as the ones found in the KMHF farm. The lower curcumin levels observed in the KMHF farm can be in part attributed to the 2.5x lower P levels found in the KMHF farm than in the two other farms (Table 2). Pooled soil nutrient data from the Lokahi and Kokolulu farms showed overall greater soil fertility compared to the KMHF farm, with overall 2.5x greater P, almost 3x greater K, 27% greater Ca and about 20% greater Mg than the levels found in the KMHF farm. The greater soil fertility in Lokahi and Kokolulu may help to explain the greater curcumin levels found in these farms as compared to those observed in the KMHF farm (Table 2). In both the Lokahi and Kokolulu farms, curcumin levels also tended to be greater in the cover crop than in the weed-free control treatments (Table 4).

The greater curcumin levels found in the cover crop treatments in two of the three farms may be attributed to improved growing conditions provided by the companion cover crops, by a greater concentration of medicinal constituents promoted by a greater water stress and competition provided by the cover crops or by another yet undetermined variable.

Research results and discussion:
Reactions from Producers

The market day and field day showed a lot of interest in growing, tasting and using medicinal plants. At the market day we gave away seed and root stalks of the five plants. We also allowed people to try samples of teas and fresh and dried plants. Many people already had personal use experience with the plants, and we learned from them about their experience with growing and using these products. At the field day, the same trend continued. While many people want to grow the plants for their personal use, almost no one expressed interest in commercial venture growing medicinal herbs in the near future. Furthermore, we would not recommend it beyond the cottage-industry level. This is due to the limited potential for a return on investment. However, small-scale production for personal use and sharing with others is considered an important value and activity in this community. We were pleased by the level of interest in medicinal plants generated by this project.

The most interesting feedback was from a woman in her early 60s, native to Hawaii, who recognized ashwaganda by smell and taste. She said she used it as a child for chronic severe kidney problems. Her grandmother and mother told her to take it as a tea, and it was her only treatment for the unspecified ailment. Ashwaganda is not traditionally used in Hawaii, or even available, so this was unusual. She could have been mistaken, but she was certain it was the same plant, and she has never had any kidney problems since adopting the use of ashwaganda tea for treatment.

Moringa is traditionally used by the Filipino community, and this was evident at the market day. One non-Filipino woman reported a project in Jamaica that her friend is doing to increase infant nutrition. The dried moringa powder is used as a formula supplement (not to replace breast milk). Her friend is growing a plantation of moringa in Jamaica for this purpose. In general, our project raised awareness about the potential to grown medicinal herbs for home-garden or small-scale market production and about alternative ways to grow the crops and to possibly increase their chemical constituents.

Producer Adoption

As explained above in relation to our second objective, we have laid the groundwork for future producer adoption when and if fuel prices increase and locally grown items can compete in the price point with imports. Alternatively, the whole-farm experience that incorporates locally-made “cottage-industry” medicinal products has immediate potential for small farms in this area. The three farms that participated may continue to experiment and research the production, marketing and development of value-added products from these herbs. Members of the community became more aware of the values of cover cropping and with the concept of deficit-irrigation as potential alternatives to increase the medicinal chemical constituents of horticultural crops or medicinal herbs.

Participation Summary

Educational & Outreach Activities

Participation Summary

Education/outreach description:

1. We publicized paid internship positions for youth via flyers, a local newspaper article and discussion at a community action plan committee meeting. School children did not lend themselves to this activity do to school hours. We expanded the age range from 15-21 year old to 15-24 in order to increase the pool of applicants.

2. Farmer Market Day booth display: 4/28/12, Hawaii, 50-60 stopped by.

3. Field Day: 5/6/12, Kapa’au, 20 people (see attachments below)

4. CTAHR Sustain Ag Newsletter article forthcoming. eg Kathie Pomeroy et al. Growing medicinal herbs with cover crops under deficit irrigation CTAHR Sustainable Ag Newsletter. Univ. Hawaii Coop. Extension Service. (forthcoming)

5. Kohala Mountain News: September 2012 article forthcoming in Oct/Nov isse. Prior article in 2011.

6. UH CoP: poster and abstract at 2011 Phytochemical Society of North America Annual Conference

7. A presentation was made by the technical advisor at the annual state MidPac Horticultural Conference held in Hilo on July 26, 2012, and discussed the on-farm Western SARE project in Kohala, as part of an on-farm research project involving medicinal herbs, cover crops and deficit-irrigation treatments.

8. Survey/email follow-up email survey: omitted

Project Outcomes

Project outcomes:
On-Farm Income from Medicinals: Farmers Market Booth Demonstrations

Further exploration is needed to understand the full potential of our second objective. This pilot project was found to be too small to generate income. First of all, it was found that a tea combining all the herbs would taste very bitter and this would make it difficult to market in the short-term. Secondly, success using the farmers’ market venue requires consistent participation for weeks and months, not just one or two times. Finally, generating income from medicinal herbs or food crops is a difficult undertaking under the current food system economy. Hawaii currently imports 85% of its food. This is mainly due to the policies and infrastructure that imitate mainland needs rather than being based on Hawaii’s needs. Many Hawaii residents expect this to change as fuel prices increase and as local markets adapt to better compete in terms of the price of imported products.

In light of this knowledge, we adapted the project by not charging for the products at the Farmer Market booth. Our tea samples and root and seed stock were all made available free to the public. This was well-received by the community, and we believe it generated more interest in medicinals than if we had charged for it. At the booth we provided samples for tasting and smelling, along with information about how to grow and use the medicinals, and consequently we had a lot of traffic. As for the tea, we offered two teas. One was a lemongrass-basil-peppermint combination served cold and the other consisted of ashwaganda (very bitter) steeped with milk and sugar, and served warm. Turmeric samples were made available to taste in the fresh and dried form, as was moringa. Galangal ginger, used in Asian cuisine, is usually cooked and then discarded like a bay leaf. We did not serve any food preparations for this purpose but did provide seedling-starts for galangal and all the other herbs, which were popular items and collected by almost everyone who stopped at our table (about 50-60 people).

Under these circumstances, we felt that it was not appropriate for us to survey the public about starting a commercial herb farm. In fact, doing so prematurely seemed potentially detrimental since a “half-baked” business concept would not establish credibility. By introducing the community to medicinals from a grant-funded educational standpoint, we felt we were laying the groundwork for future efforts.

One model that has been financially successful in Hawaii incorporates Ag and Health Tourism. Both Kokolulu Farm and Lokahi Farm already have programs that use this approach but without any herb products per se. When herbs are introduced in the context of a full-experience that includes the farm and related lifestyle, herb products can be marketed more effectively and on a much smaller-scale than a cooperative or other models that are based on achieving an economy of scale. This method has its own challenges, however, due to county and state regulations that impede Ag and Health Tourism infrastructure and operational needs. Except for the limitations imposed by such regulations, all the farms in the project are in a good position to promote and sell medicinal herbs as part of a whole-farm experience.

Youth Internship

Our goal was to involve at least six youth interns in the project from start to finish. The age range we targeted was 15-24 year old participants. We engaged over a dozen interns in the project, however, only three were involved in the entire process. Of those, only one did so in a fashion that was consistent with our original intentions. Some of the obstacles were that labor was needed intermittently; transportation was difficult for some; and that most local youth have little experience with the physical demands of farm labor and even less of a desire to actually do it.

Of the three interns who stayed, the one outstanding and successful “poster child” was a young man (age 24) who already had a high level of ag skill, experience and enthusiasm when he joined the project. We believe that our youth participation results were consistent with many similar opportunities offered in this area. There is generally low “early adoption” among youth of any novel idea, or anything new even if it is not unusual. It usually takes several years to create a presence, to become a “known entity” and to establish a good reputation within the community in order to attract steady, uninterrupted participation in a project. We have seen this even in the case of generously funded outdoor adventure and youth entrepreneurial projects that are not nearly as demanding in terms of physical labor. Considering these factors, we are satisfied that we were able to expose a dozen youth to the project and significantly impact about six of those. “Significant impact” means that these youth got exposure to what is needed for an ag endeavor, even if they decided not to follow through with it at this point in time.


Potential Contributions

Benefits for Agriculture

This project increased community awareness of medicinal plants for food, medicine and livestock feed. It also emphasized using sustainable methods of agriculture. The results of this project were shared with many in the community, including at the farmers market and at the field day, and people became more aware about on-farm research and about the potential of niche medicinal herb markets in Hawaii.

Future Recommendations

Recommendations or New Hypotheses

The natural variability observed among the three farms makes it difficult to make any final or definite conclusions. Additional research is recommended, perhaps focused on one crop at a time to further narrow down the potential of using cover crops and deficit irrigation to improve the constituent content and quality of horticultural crops and herbs in Hawaii.


Akamine, Hikaru; Hossain, Md. Amzad; Ishimine, Yukio; Yogi, Kenichi; Hokama, Kazuo; Iraha, Yukikazu; Aniya, Yoko. 2007. Effects of Application of N, P and K Alone or in Combination on Growth, Yield and Curcumin Content of Turmeric (Curcuma longa L.). Plant Production Science (Japan), 10(1): 151-154

Biodiversity and the Ecosystem Approach in Agricuture, Forestry, and the fisheries Conference. Impacts of Cultivation and Gathering of Medicinal Plants on Biodiversity, Oct 2002, Rome

Khalida, K.A., Jaime A. Teixeira da Silva, and Weiming Cai. 2010. Water deficit and polyethylene glycol 6000 affects morphological and biochemical characters of Pelargonium odoratissimum (L.). Scientia Horticulturae 125:159–166.

Kubsad, V. S.; Palled, Y. B.; Mansur, C. P. 2009. Productivity, quality and economics of rainfed ashwagandha (Withania somnifera) as influenced by spacing and fertilizer levels. Indian Journal of Agronomy (India) 54(4):449-453. ISSN 0537-197X

Lopes, C.M., Tiago P. Santos, Ana Monteiro, M. Lucília Rodrigues, Joaquim M. Costa, and M. Manuela Chaves. 2011. Combining cover cropping with deficit irrigation in a Mediterranean low vigor vineyard. Scientia Horticulturae. 129:603–612.

Masmoudia, C.C., Mouna Mezghani Ayachi, Mohamed Gouia, Foued Laabidi, Saloua Ben Reguaya, Abderrazzek Oueled Amor, and Mohamed Bousnina. 2010. Water relations of olive trees cultivated under deficit irrigation regimes. Scientia Horticulturae. 125:573–578.

Niranjan, A.; Dhan Prakash; Tewari, S. K.; Pande, A.; Pushpangadan, P. 2003. Chemistry of Curcuma species, cultivated on sodic soil. Journal of Medicinal and Aromatic Plant Sciences (India). 25(1)69-75. ISSN 0253-7125

Oh, M.M., Edward E. Carey and C.B. Rajashekar. 2010. Regulated Water Deficits Improve Phytochemical Concentration in Lettuce J. AMER. SOC. HORT. SCI. 135(3):223–229.

Palesea, A.M., Vitale Nuzzo, Fabio Favati, Angiolina Pietrafesa, Giuseppe Celano, and Cristos Xiloyannis. 2010. Effects of water deficit on the vegetative response, yield and oil quality of olive trees (Olea europaea L., cv Coratina) grown under intensive cultivation Scientia Horticulturae. 125:222–229.

Patanè, C., Simona Tringali, and Orazio Sortino. 2011. Effects of deficit irrigation on biomass, yield, water productivity and fruit quality of processing tomato under semi-arid Mediterranean climate conditions. Scientia Horticulturae. 129:590–596.

Pernice, R., Mario Parisi, Italo Giordano, Alfonso Pentangelo, Giulia Graziani, Monica Gallo, Vincenzo Fogliano, and Alberto Ritieni. 2010. Antioxidants profile of small tomato fruits: Effect of irrigation and industrial process. Scientia Horticulturae. 126:156–163.

Wang, Y., Fulai Liu, Andreas de Neergaard, Lars S. Jensen, Jesper Luxhøi and Christian R. Jensen. 2010. Alternate partial root-zone irrigation induced dry/wet cycles of soils stimulate N mineralization and improve N nutrition in tomatoes. Plant and Soil. 337:167-177. DOI 10.1007/s11104-010-0513-0.

Technical Advisor Statement

This project consisted of on-farm research involving three organic farms in the rural district of North Kohala on the Island of Hawaii, Hawaii. This was the first time that these growers experienced a formal on-farm research project, and it was a great learning experience. All of the farmers were highly committed and flexible to accommodate the needs and timing of their fellow researchers. Because most information in the community occurs from word of mouth, it is likely that many members of the community learned about the project, about cover crops and about medicinal plants, throughout its duration. Despite the fact that all farmers followed very (relatively) consistent production practices, in terms of timing and plot management, including all using the same source of cover crops seeds and planting materials for the herbs that were used, the natural variation between farms makes it difficult to make definite conclusions to explain the results (i.e. too many confounding factors, which is also an inherent feature of agroecology research). Overall this was a great learning experience for the farmers involved, for the youth that participated and for the community members that got exposed to the project. This project also was a step forward in familiarizing people in the area with the concept of on-farm research. In Hawaii on-farm research is still a relatively new concept, and it will take several years and many projects to refine the procedures, to obtain a buy-in from farmers and to accept on-farm research as an ongoing process to continually improve their farming operations. – Hector Valenzuela, Ph.D.

Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.