Hōkūnui is 100% water self-sufficient. The project has a well for residential use and has designed an innovative pond capture system for agricultural water needs. 

The Project: Water Security for Agriculture—Capturing and Utilizing On-Farm Surface Water

Act 172, approved by the Governor on June 30, 2016, established a two-year pilot program for a water security advisory group to enable public-private partnerships that increase water security by providing matching funds for projects. The State objectives of program were to:

  • Increase the recharge of groundwater resources
  • Encourage the reuse of water and reduce the use of potable water for landscaping
  • irrigation; and
  • Improve the efficiency of potable and agricultural water use

Hōkūnui was awarded a grant by the Water Security Advisory Group (WSAG) for its project entitled: Water Security for Agriculture—Capturing and Utilizing On-Farm Surface Water. During the project timeframe we accomplished the following:

  • This project constructed a 1 million gallon detention basin and farm road that collects water from a 25-acre area. 
  • A filtration and distribution system were designed in order to treat and transport the water to end uses. 
  • An app and monitoring process created to track rainfall and water capture.
  • Educational outreach—print, website, social media and a video.

    Hōkūnui hopes to demonstrate to other land developers, ranchers and agricultural operations that it is possible to efficiently capture surface water in order to reduce reliance on water that is diverted from natural stream flows and municipal water supplies. 


    Most developments are designed to direct water away from the property—into storm drains, streams and the ocean. Hōkūnui is designed so that surface rainwater is captured and retained on the property in ponds for agricultural use. Two regenerative design principles are in play here:

    1. The project design incorporates seven detention basins at keypoints in order to collect and store runoff water for agricultural use. 
    2. The Native Forestry Corridor is planned within the same natural drainage area as the seven drainage basins, which makes use of the water for establishing native reforestation and agroforestry. In turn, the native plants and trees return the favor through their wise use of water, as opposed to water thirsty invasive species, which sap the water resources of an area.

    Plant life at Hōkūnui Farms begins at Kapūʻao (the womb) Plant Nursery, which nurtures about 30 different native and canoe plant species before they are planted in the ground. Hōkūnui also works with native plant experts in our community to source plant material and learn from their expertise.

    Keyline Design and Keypoints

    Hōkūnui’s overall design philosophy was informed by Darren Doherty and his Regrarians platform, which derives heavily from the work of P.A. Yeomans who, in the 1950’s in Australia, developed what is called Keyline Design.

    “[Keyline Design] A comprehensive design strategy for agricultural and urban development based on fundamental, repeating land shapes that have been created by water.” -Abe Collins[i]

    In Keyline Design water capture begins at the “keypoint,” the natural point of water collection on the land. “Keylines,” the other concept in Keyline Design, are contour lines cut by a Yeomans Plow (loosens, without turning over soil) that are parallel to the keypoint and help distribute water to the naturally dryer parts of the landscape.

    According to the California Ag Water Stewardship Initiative:

    “Keyline systems capture significant quantities of water that would otherwise run off, and store it in the soil. Keyline systems also build soil fertility, which further improves moisture-holding capacity. Ultimately, while no research has quantified the reductions in applied water associated with this system, it is clear that the benefits are substantial. For instance, we know that for each 1% increase in soil organic matter, which can increase water storage by 16,000 gallons per acre-foot of applied water.

    Keyline plans typically employs water storages (usually ponds) as a component of the overall plan. Small ponds of surplus runoff water can be placed at the natural intersection of a ridge and a valley, or convex and concave slopes, known as a key point. This stored water can provide gravity-fed irrigation later in the season for pastureland or crops. The spillway or primary outlet channel from ponds is managed in a way to maximize the distribution of water to irrigate the land below.”[ii]

    Using the natural keypoints of the land, Hōkūnui has been successful at designing retention basins to capture water for agricultural use.

    [i] KEYLINE DESIGN Mark IV ‘Soil, Water & Carbon for Every Farm’ Building Soils, Harvesting Rainwater, Storing Carbon Abe Collins & Darren J. Doherty

    [ii] California Ag Water Stewardship Initiative:

    Stormwater Capture and Use

    Taking advantage of land shaped by plantation agriculture and natural land features, is demonstrating how to use regenerative principles to construct a water capture system that makes efficient use of surface water for agriculture, livestock, forestry and landscaping.  With the support of Roth Ecological Design Int. (REDI), the drainage areas to the right (noted as D6 and D1A & D1B) to the Ponds were delineated and stormwater modelling was conducted as part of a Drainage Assessment to estimate the potential for stormwater capture.

    The areas above marked D1A and D1B represent an approximately 25-acre drainage area that diverts storm water into Pond #1.  Existing roads were engineered to direct the water into the pond. 

    The stormwater modelling incorporates the soils information, slopes and land types and uses in each drainage area to calculate stormwater runoff volumes during specified rain events.

    (REDI)s Drainage Assessment and  stormwater modeling results can be found here:


    Table 1 below summarizes the stormwater runoff volumes determined by the model. According to the model, significant stormwater capture in Pond 1 would occur when rainfall events areis equal to or greater than 1” over 24hrs.   Importantly for rain events exceeding 4’’over 24hrs, the stormwater runoff volume exceeds Pond 1’s storage capacity. Therefore, the maximum stormwater runoff volume that could be captured in any given ran event is 1 million gallons.

    Table 1:  Stormwater Runoff Volumes for Capture in Pond 1


    REDI then also analyzed 3-years of rain data collected from a neighboring station.  The table below shows the number of rainfall events, highlighted in green are the events that are greater than or equal to 1” and will therefore result in stormwater capture into Pond 1 and provide irrigation/agricultural water supply (estimated to be 6000 gpd).


    Based on the rainfall data over the last 3-years, there was a total of 55 events that would create significant volumes of stormwater.  This was further broken down into the frequency of certain storm intensity events to estimate the potential for stormwater runoff capture in Pond 1.  During the 3-year period the model results estimate ~12.6MG gallons of stormwater runoff that could be potentially captured in Pond 1.  

    Although rainfall can vary significantly from year to year, we used this 3-year rainfall analysis to estimate stormwater runoff capture in Pond 1 to be ~ 4 MG per year, which assumes stormwater capture for storms greater than or equal to 1’’over 24hrs.   Table 2 summarizes the frequency of rainfall events equal to or greater than 1’’-24hrs over the 3-years of data collected.  Where there was the one storm that exceeded 4’’-24 hrs. in the data set, it was assumed that Pond 1 would only be able to capture volume of its storage capacity (1 MG) and the remaining would overflow into the gulch.  

    Table 2:  Potential for Stormwater Capture in Pond 1 Using a 3-yr Historical Rainfall Data Set


    With an intention to capture the maximum amount of stormwater and keep it on the property (not letting it flow into the gulch), planning the irrigation/usage schedule is important. The water in the ponds must be used consistently, in order to allow enough empty space to capture more stormwater during significant rain events. When the two existing ponds (Ponds 1 and 6) are full, they could store 50-80% of the estimated 6,000 gallons per day needed for agricultural use.  

    During the monitoring period (March-October 2019), there was very little opportunity for stormwater capture. Only four rainfall events were over 1”, and those were scantly over 1”.

    Table 3: Daily Rainfall Data at Hōkūnui (Weather Underground, Weather Station ID: KHIHAWAI9)


    Pond Measurements

    Another simple method Hōkūnui  explored to measure actual volumes of stormwater captures was to use pond water height measurements.  These measurements occurred during the month of April 2019.


    • There are two flags placed at the edge of the pond—one in a permanent location and one at the upper point of the water line. 
    • Staff takes the measurement (in inches) between the flags.
    • The data is recorded in an app.
    • The height gain is recorded and the difference in height in the pond is calculated into total stormwater captures during that period of time
    • REDI prepared a stormwater calculator for Hōkūnui  that is designed for Hōkūnui staff to input the difference in the height of the pond water and output the stormwater volume captured.  The modelling was based on the calculations shown in the Stormwater Storage Calcs figure below.
    Download the Stormwater Calculator here.

    April 2019 Stormwater Capture

    Using the Stormwater Calculator for Pond 1 provided by REDI, Hōkūnui was able to calculate the stormwater captured in April 2019.  There were two events in April where the Pond water height rose one-foot, both on April 22nd and April 25th.  The Tables below show pond water heights taken in April and the results of the stormwater captured.  In April, according to the model, Pond 1 collected 197,662 gallons of stormwater runoff.

    Pond Height Data Collected in April 2019


    Stormwater Capture Calculator Results for Data Collected in April 2019




    The cost of the pond and road system was approximately $380,000. Whether or not this system is financially feasible for other landowners depends on the engineering design of the pond. County supplied agricultural water is cost effective. However, given the increased frequency of droughts, the limited quantity of water meters, areas where County supplied water is not available, or where a landowner is relying upon contentious and scarce water sources, this is viable long-term solution that can be employed.

    The pond at Hōkūnui is lined with a geosynthetic clay liner called Bentoliner®.

    Utilizing the Water

    Capturing the water is one thing, and distributing it is another. Based on conversations with Hōkūnui staff and an analysis of the topography, water quality, usage, need for solar power, and pond designs, Roth Ecological Design International (REDI) recommended the following outline of the general design and equipment required for:

    • 100 micron filtration for 8gpm.
    • Solar pump to provide 4gpm at 60psi to water throughs above pond #1 location.
    • Review of submersible pump selection from pond #6 to pond #1


    The existing pond #1 required filtration to provide 8 gpm for agricultural uses at various locations on the site. This recommendation does not cover the design of piping from the pond to points of use, only filtration of 8gpm and pressurization of 4gpm@ 60PSI to water troughs. The sketch at the end of this document lists two 2” Amiad Super T 100 Micron scan-away filters in parallel that filter pond water from Pond #1 (Elevation = 2,050ft) at a minimum flow of 8 gallons per minute. Operators will manually backwash filters on a weekly basis. If filters are clean on a weekly basis then frequency of backflushing can be increased to every other week and so on as necessary.

    Downstream of screen filters, a standpipe will supply 4GPM to a continuous duty diaphragm pump operating off direct solar to fill water troughs above pond. Pump has internal pressure switch to shut off automatically when pressure reaches 60PSI. Pump can also run dry without damage if screen filters aren’t cleaned appropriately. Pump model is: Sea flow Series 51 24VDC pump model# SFDP2-040-060-51. See cutsheets for pump, solar controller, solar panel and

    installation sketch at end of document. It is recommended that the pump and filters be covered from the elements to prevent damage from uv and weather.

    REDI’s Engineer also has reviewed the solar pump information by Hōkūnui. The pump that was selected by Hōkūnui was to pump from pond #6 (Elevation = 1,781ft) to Pond #1 (Elevation = 2,050ft) The selected SR-12 is acceptable (See reviewed cutsheet) flow rates to be determined by Hōkūnui by number of panels selected to power pump.

    Pump specs can be found here.

    Floating Wetlands

    Hōkūnui is considering the addition of floating wetlands or the introduction of other aquatic plants to the pond. One consideration is Azolla, a plant that can be useful as animal feed or feed for hydroponic fish systems.

    Click here to receive info about the Aquatic Resources for Chicken Feed.

    Resources that we are utilizing to evaluate the floating wetlands design is the Floating Wetland Case Study for Kā‘anapali Lagoon and a study conducted for Hōkūnui by REDI of various plants to grow in the ponds for poultry feed.


    In summary, we expect the stormwater pond to capture 4 million gallons of stormwater runoff per year.  Hokunui plans to continue to take Pond measurements using the flag system as a method to track stormwater captured.  We are committed to sustainable water management and are available to share our experience capturing and using stormwater runoff as a resource.