INTRODUCTION

The most critical issue facing Hawaii’s livestock segment of our agricultural industry is developing and implementing cost effective pollution prevention technology. Unlike their U.S. mainland counterparts, Hawaii’s farmers are faced with higher land costs, higher cost of materials and supplies due to transportation expenses and highly sensitive and unique ecological zones of the islands. Our neighboring Pacific-Basin island nations face the similar concerns in their insular island environment.

A national revolution to improve the quality of our environment is upon us. This stimulus is driving a revolution in livestock waste management. With increasing concerns and public pressures relating to the Clean Water Act, concentrated livestock operations have come under increased scrutiny by environmental groups and regulators. Traditional waste disposal methods, primarily as effluent lagoons, are nearing capacity levels and have resulted in environmental violations due to effluent discharge into streams. Economical, yet practical alternatives or best management practices, need to be developed in order to improve the sustainability of confined feeding operations in these pristine island economies.

Manure, a waste by-product of animal production, can be considered in two ways; most commonly as a liability that created odors, builds up nuisance pest populations, and increases nonpoint source pollution from run-off and seepage; and secondly, as a resource. By shifting paradigms, animal manure can be processed and developed as a marketable organic soil amendment for the agricultural, garden, and landscape industries. The interest in and movement towards organic products is tied to this "environmental revolution" and is creating opportunities for innovators of nutrient management, rather than, waste management. The dry litter technology was developed and practiced in environmentally sensitive and land limited countries such as the Netherlands, Japan, and Taiwan. However, problems of excessive heat generation, disease and parasite build-up particularly in enclosed buildings were of primary concern. Therefore, the intent of the program was to develop a modified cost-effective and environmentally suitable version a modified dry litter system for Hawaii and the tropics.

This project sought to resolve those problems, while addressing other concerns such as the high cost of labor and development of a nutrient product that would be acceptable to the agricultural and landscaping industries. Its main objective was to dramatically eliminate the use of water in the cleanup, transport, and collection phases of the traditional waste management system; virtually eliminating the potential of animal waste as a nonpoint source pollutant. Project is entitled "Animal waste management program in Hawaii: Implementation and evaluation of the modified dry litter waste management system under a tropical agricultural ecosystem".

BACKGROUND

The Modified Dry Litter (MDL) Waste Management System is designed to reduce or prevent non-point source pollution by eliminating the use of water to clean hog production facilities. Elimination of water in the system, remove the possibility of pollution from various components of a typical confined feeding operation waste management system. The traditional water-based system is currently the standard for the nations hog industry and over ninety percent of the hog industry in all tropical nations in the world. The on-farm demonstration research project developed and investigated various aspects of the Modified Dry Litter Waste Management System.

Project Location and Facility The cooperator farm is located in the village of Kealia in the Kona Soil and Water Conservation District on the western side of the Island of Hawaii. Kealia Farm, constructed the first prototype of the MDL system shown in figure 1. Kealia Farm is a commercial farrow-to-finish hog operation integrated with other crops such as coffee, avocado, citrus and taro. The farmer financed the construction of the facility. An engineered greenhouse structure provided cover and protection for the livestock and the solid and liquid storage components. A 6-mil polyethylene sheet and an 80% shade cloth were installed as the roofing material. Gutters were installed to divert rainwater away from the facility. The pre-compost holding trench, approximately 6 feet wide and 4 feet deep ran across the width of all pens, was constructed wide enough to accommodate a small front-end loader, typical of the type of equipment normally available on most livestock farms. At one end of the holding trench, a liquid storage facility was incorporated to collect and store drinking water and rainwater runoff.

Pen Slopes The Kealia facility consisted of a bank of six production pens, 12 feet wide by 20 feet in length. Four different pen slopes were integrated in the design; the slopes were: 8:1, 10:1, 13:1, and 20:1 (two pens). A 40:1 slope was incorporated into the trials by the addition of a batter board at the lower end of the pen. Tested various slopes to evaluate the movement of the selected dry litter materials and the performance and behavior of the test animals. See Table 1 for pen slope descriptions.

Carbon Materials Various types of carbon materials were used for litter materials and observation noted for adaptability in the MDL system. Materials tested included macadamia (Macadamia integrifolia) nut husks, wood chips from commercial tree trimmers, and guinea grass (Panicum maximum) hay. Bulk density determinations, total carbon added, discharged and residual litter in the pens were recorded.

Animal Performance Nursery pigs, ranging in age from 6-9 weeks old, were evaluated in the growth performance trials. Live animal weights at the start and end of the 49-day trial period and total feed consumption data were recorded. An all grain diet and water was available ad libitum. Crossbred feeders (Yorkshire, Landrace, Chesterwhite, Hampshire and Duroc breed crosses) were used for the trials and were stocked at a density of 10 ft2 per head. Average daily gains and feed conversion ratios were calculated. In trial 3a, feeder pigs were selected from the previous trial, stocked at 12 ft2 per head and continued for an additional 28-days.

Odor Hydrogen sulfide gas was used as the indicator for odor. A gas meter (Jerome Analyzer, Model 631-X) used for state regulatory purposes was used for the trial. A comparison between a conventional water-based waste management system and the MDL system was evaluated. Gas measurements were taken at five selected points within the pens and at the various components (storage, transport) of both systems.

Composting Process Approximately one yd3 of pen discharge material was sub-sampled and composted. Three bins were constructed to keep compost batches separated. The static pile composting method was utilized. During the initial 2 month-period, the batches were aerated approximately every 2 weeks. In the subsequent 1-2 month to final curing, the compost batches were not turned. Temperature data were collected every 2 days for the initial month to ascertain the maximum temperature. Final compost were analyzed for bulk density, pH, electrical conductivity, dry matter, organic carbon, total nitrogen, phosphorus, potassium, calcium and magnesium.

Economics A partial budget cost analyses of a 25-sow farm unit comparing the MDL system with a typical water-based waste management system was evaluated.

Outreach and Technology Transfer Dissemination efforts targeted industry organizations, conservation districts and state water quality conferences. In addition, contractual quarterly reports were submitted to regulatory and funding agencies.

RESULTS AND DISCUSSION

Pen Slope and Carbon Material Interaction The key to the system is the sloping pen floors. The sloping floors and the forces of gravity cause the litter material to flow out of the pen and into the holding trench. The steeper slopes resulted in a higher flow of materials through the pens. The movement of material out of the pens will also depend on the type of dry litter. A coarse (> 5 cm) carbon material would require a steeper slope compared to a moderate-sized particle (3-5 cm) litter material. The most effective slope formacadamia nut husks (MNH) was 20 to 1; the steeper slope, 8:1, worked better for commercial tree trimmings (TT). Materials with a low bulk density, like a long hay-type of material will tend to mat down in the pens and clog the system, thus is not recommended as a carbon material in this system. Table 2 shows the effects of the interaction between pen slope and carbon materials and carbon movement out of the pen and change in moisture content.

In Trials 1 and 2, MNH was evaluated under four different pen slopes. Very little material flowed out of the 40:1 pen slope, thus is not recommended. The adequate flow was achieved with the other slopes. In the study, MNH proved to be the best dry litter as it exhibited excellent absorbing qualities and degraded acceptably in the composting phase. It resulted in the compost with the best texture, appearance and ‘earthy’ smell, and was highly preferred by a neighboring organic farmer. The total amount of MNH used per pen during the 49-day test period was approximately12 yd³. Each pen was initially loaded with 4 yd³ and approximately one yd³ was added per week. Bulk density losses of the discharged material were 21% for the 40:1 sloped-pen and an average of 10% in the 20, 13 and 10:1 sloped pens. Bulk density loss in the compost process was estimated at 50%.

In Trial 3, TT did not move out of the pens due to a dam-effect of the large particles being caught in the wire panel and compaction of the fine particles. Modification to the wire panel in the subsequent trial allowed the carbon material to flow out of the pen. TT proved to be adaptable to the MDL system; however requires a steeper slope (8:1). Carbon flow tended to bog down at the 20:1 slope. Breakdown of the coarse trimmings was much slower than MNH. The final product exhibited greater variability in particle sizes; from fine particles to larger material measuring about 4 cm. Bulk density losses were not measured for the TT, but would be less than the MNH material. The TT materials are highly variable; particle sizes may range in length from 2 to 20 cm with a diameter of up to 4 cm. Leaf to stem ratios are variable depending upon the type of woody species, stage of growth and seasonal or other environmental conditions. No animal injury was observed with all carbon resource used in the trials.

In Trial 4, the long hay proved to be the poorest carbon material tested. The material simply clogged the system due to the collapse of the hay fibers. Clean hay had to be continuously top-dressed to keep the animals dry. With the buildup of carbon material and nutrients in the pen, the composting process was initiated as evident by a rise in temperature (101°F) within the pre-compost mixture. However, no deleterious effects on the animals were observed.

The carbon-nutrient mix needs to flow out of the pens in order to achieve the second benefit of the system. The separate composting trench is the second critical design element that prevents the hogs from exposure to the pre-compost material where excessive heat in generated or may be a reservoir of diseases and/or parasites.

It is recommended that carbon sources be secured by long term contracts with the greenwaste generator. Examples of greenwaste generators include utility service tree-trimming contractors, hotels and resorts, agricultural processing facilities, and municipal greenwaste collection sites. The decision on pen slope would be dependent on the type and availability of the carbon resource in your area.

Animal Performance Table 3 shows the results of the trials. A total of 204 animals were used in five trials to evaluate the effects of pen slope and carbon interaction on the performance of nursery and feeder pigs. A total of five different slopes and three carbon materials were evaluated. Due to the herd size of the cooperator farm, only three groups of animals could be evaluated per trial. Different pen slopes were selected and tested. Pen slopes that did not allow the carbon material to flow were dropped from further testing. A different slope was incorporated in the subsequent trial. Trial 1 and 2 evaluated animal performance on MNH on four slopes. There were no differences between the treatments. Trial 3 evaluated the effects of commercial tree trimmings on two slopes. There was no difference in animal performance between the two slopes. However the overall performance, compared to the MNH, may have been lower due to the slightly wetter conditions. In trial 3, the hog panel prevented the TT from flowing through the pens. An adjustment was made by raising hog panel, creating a larger space between the pen floor and bottom of the hog panel. In Trial 3a, selected animals were placed on the steepest sloped pen and continued an additional 28 days. The purpose of this trial was to evaluate the pen modifications made after Trial 3. Increasing the space allowed the larger particles to flow and resulted in a improved gain by the animals. Trial 4 was conducted on a new farm, which adopted the MDL system. Guinea grass hay (GGH) is produced in the district and was incorporated into the testing protocol. All pens at the new facility were built at 20:1 to facilitate the use of MNH as its major carbon resource. As mentioned earlier, the GGH matted down and accumulated in the pens. The pigs remained fairly dry, as compared to Trial 3, thus performance was not affected. The gain advantage observed in Trial 4 is probably due to genetic effects and stage of growth of the pigs. In general, the effects of slope and carbon material on the growth performance of the animals are minimal. Based on the Pork Industry Handbook performance guidelines, weight gains and feed conversion matched the profile for good to excellent production.

Odor Table 4 compares the hydrogen sulfide levels produced by the dry litter system and a well-maintained traditional (water-based) hog waste management system. Measurements for the production, transport and storage components were taken. Odor levels, as estimated by the hydrogen sulfide parameter, was lower in the MDL system. After the initial reading, carbon material was added and a second reading was taken 24 hours later. The addition of carbon material to the pens further reduced the hydrogen sulfide levels in two of the lesser-sloped pens. The carbon interaction and its mechanisms of lowering odors are not well understood, but it is related to optimizing the balance of carbon and nitrogen for efficient microbial assimilation of nutrients. The MDL system offers an alternative management technique to controlling critical odor generating points by adjusting the carbon allocation to the pens. Worker safety and risks, in regards to hydrogen sulfide exposure, is very low. Workers do not enter the production area during normal daily operations. The facility has a high roofline and no solid vertical walls, which allow for more than adequate ventilation. The total daily average man-hours for activities such as feeding and carbon loading is less than 2 hours/day (EPA guidelines; 15 minutes is the limit for worker exposure to hydrogen sulfide at 15 ppm. For longer term work 10 ppm per eight hours is the limit).

Compost The total composting time ranged from 4 to 5 months. Maximum temperatures for the various compost batches were; 151 oF, 142 oF and 130 oF for pen slopes of 40:1, 20:1 and 10:1, respectively. The maximum temperature of the pre-composted discharge from the holding trench was 165 oF. The initial high temperature composting process begins in this storage area. Most of the destruction of pathogens occurs at this stage. Evidence of high temperatures was observed as steam when the discharged material was removed to final composting. For our trials, the retention period was seven weeks. The retention period of the entire facility is estimated at 16.7 weeks (based on six-pens, assuming an equal load and discharge rate of one yd3/week and a 100 yd3 holding trench capacity). Thus the high temperature composting will occur during this time period.

No measurement for heavy metals were not taken since no basis for heavy metal contamination of the compost was evident. A screening for parasite eggs were conducted by Dr. Brad LeaMaster of the Department of Animal Sciences, College of Tropical Agriculture and Human Resource, University of Hawaii at Manoa. No eggs were detected in the compost samples. A set of samples was tested for coliform bacteria. Results confirmed a presence of the bacteria; but no total count was determined in the test procedure. These bacteria are omnipresent in the environment and are very difficult to eliminate. Although high temperatures were achieved in the composting process, contamination via air borne vectors may have occurred.

In general, the compost generated from the MDL system is a high value organic amendment. The final carbon to nitrogen ratio is lower (i.e. more nutrient dense) than most commercially available compost products (Hue, N. V. and H. Ikawa, 1994) due to the low carbon in the source material used and the higher than normal co-composting nutrient loading or incorporation. Co-composting operations minimize the addition of nutrients due to cost factors. The C:N ratio in the compost is very acceptable to cropping systems and will not cause problems due to nitrogen deficit resulting from high carbon mulches. The salinity values were very high in three samples from the same trial set; but with normal use in cropping systems, the compost would not cause burning problems. However, it would not be recommended for pot culture or seedling production.

Sales of the material were brisk when the farmer was aggressive to offer it to the organic farmers who paid as much as $ 70/yd³. Test sales of ten-pound bags proven that there was a high demand for the material. Based on recent sales of a similar product using chicken manure, compost value at the retail would be about $ 135/yd³. The composition of the raw materials and compost are shown in Table 5.

Economics The modified dry litter waste management system is not a separate facility, but its design is incorporated the production unit. Achieving efficiency in cost of construction and in land use, particularly in land-limited island ecosystems. It is a simple design. There are no mechanical parts or specialized equipment required in comparison to methane digesters, solid separators, pumps, flushing systems, or other waste management systems. The design concept is modular to allow for incremental expansion. By-products of the system creates a revenue stream from its high value co-compost, rather than being stored (effluent lagoon) in a dead-end system. The growth of the composting industry across the US has increased the consumer’s awareness and utilization of organic soil amendments. The "organic" revolution is crossing new barriers into mainstream commercial crop production, which, at one time encircled only the niche organic farms.

A cost analyses example comparing the waste management systems of a typical water-base system and the MDL system for a 25 sow hog operation is shown in Table 6. These figures are based on average construction cost in Hawaii and other Pacific Basin islands, which can be two to three times higher than U.S. Mainland figures. The high cost factor can be attributed to the transportation of construction materials and supplies and higher labor costs.

According to the USDA Farm Service Agency, a typical lagoon waste system for a 25-sow unit, with a 60-day storage volume of 400,000 gallons, would cost $32,680. It has a useful life of 20 years. This includes a liner and required perimeter fencing. Utilizing USDA Natural Resources Conservation Service standard values for annual fixed cost (13.3 %), annual repair and maintenance cost (1.5 %) and assuming 2 hours of labor ($8.00/hr) for pen wash down and water costs; the total annual system cost is $11,045.

The cost of the central collection trench in the MDL system was $7,500. It has a useful life of 30 years. Assuming annual fixed (13.3%) and repair and maintenance costs (1,5%), estimated labor to load the carbon material (91.25 hours/year) and labor costs to remove the pre-composted carbon from the trench (6 times at 8 hours/ event = 48 hours/year); the total annual system cost is $2,224. The MDL system is clearly cost-effective. Its initial cost of construction is approximately one-fourth the cost of a typical system. It has lower operational, maintenance, labor and water costs. In addition, the economic costs of responding to complaints of bad odor and potential water pollution fines and legal costs are real cost not accounted for in this analysis. These costs are definitely part of what the typical hog farmer using a lagoon system is faced with in today’s operating environment.

Compost Revenue. Traditional water-based systems in Hawaii was designed basically as a storage facility. Very few farmers actually utilize the nutrients built up in the retention ponds. The MDL system again shifts the standard mode of storage into a focus of product utilization. The market price for compost product range from a high of approximately $150/yd3 for retail, bagged compost product to a low of about $50/yd3 for wholesale, bulk compost product. Based on the trials conducted, approximately 1 yd3 of carbon material was used per pen per week during the 7-week period. Assuming 50 weeks of production, 50% reduction in volume, and a 50% operating cost per pen, more than $600 could be generated from each pen on an annual basis.

Outreach and Technology Transfer The members of the interagency research development team presented and disseminated information about the MDL system at industry meetings, workshops, invited conferences and training programs during the project period from 1995 to 1998. Technology transfer opportunities included state (6), national (4) and international (3) presentations; as well as exposure in state newsletters (2 articles) and national periodicals (2 articles).

The MDL projects have made many impacts in the way people and farmers view the concept of waste management. We have tried to shift paradigms in our outreach efforts; changing keywords such as waste to nutrient, liability to assets, expense to revenue and storage to utilization. Industry adoption examples include: Na'alehu site, a large hog operation in the Kahaluu watershed (Oahu), youth programs and small garden operators in South Kona, and Pacific-Basin outreach.

Na’alehu site. A second prototype unit was constructed in the town of Na’alehu in the district of Ka’u, with funds from the Rural Economic Transition Assistance program. These funds targeted and assisted displaced sugar workers in developing economic generator projects in their communities. This facility is an example of technology transfer and includes modifications and improvements to the original prototype.

Modifications to the layout of the engineered structure were incorporated to allow for a modular approach to future expansion. Economically, it was more cost-effective to build two sets of pens sloping into a common holding trench (Figure 2.). Instead of one long contiguous structure, smaller units were repositioned and connected to form the structure. The new alignment also provides greater eave coverage and proven to be better in preventing rain from entering the holding trench and the pens. A double layer of clear plastic sheeting backed by an 80% shade cloth was used. The clear plastic (double layer of 6-mil polyethylene, $0.20/ft²) was inexpensive, however showed greater evidence of wear, compared to the Kealia site, due to more intense solar radiation and wind. We estimate a replacement cycle of 18-24 months. The manufacturer estimated a 10-year life span on the polyethylene cover. Another type of greenhouse roof cover was available. Although likely to be more durable, the rigid 8mm-polycarbonate material was priced much higher ($1.65/ft²) and not used as the roof cover. The life span of this material under tropical/sub-topical conditions is also not known. Another benefit of the polyethylene/shade cover, besides initial cost, is the ability of the farmer to quickly reduce air resistance in case of high wind conditions (typhoon or hurricane). Reducing air resistance will prevent damage to the tubular structure. With a rigid roof cover, quick removal of the material is not possible. Figure 3 shows a three- dimensional view of the entire structure.

Changes were made to the prototype design to improve the overall system. These design changes were a result of a collaboration of observations and ideas during the course of the applied research work. Summary of these changes include:

1. Increased roof eaves to reduce rainwater intrusion into the pens; also to protect the animals from high solar radiation loads typical of tropical environments.
   
   
2. Expanded pre-compost storage area (12 ft. wide) to allow greater access for farm equipment and to accommodate greater holding capacity for the double-winged design.
   
   
3. Relocated water source to sides of pen to reduce excessive moisture conditions and subsequent clogging in the rear of the pen. Drains for drinking water overflows was not needed with adjustment in waterer location.
   
   
4. Removal of effluent storage tank which was intended for drinking water and rainwater diversion. This component was not needed due to adjustments made in waterer location and moisture losses during the composting process.
   
   
5. Redesigned pens to maximize efficiency of standard building materials (hog panels). The Kealia facility pen were too large and required tailoring the standard 16-foot hog wire panels. The pens at the new facility (8 ft. x16 ft.)were designed to make more efficient use of the standard hog panel and were fully gated to facilitate movement of the animals from pen to pen within the housing structure.
   
   
6. Added liquid feeder to facilitate the typical garbage (wet food wastes) fed operations. Troughs were welded on the alley gates to facilitate the high-moisture feed.
   
   
7. Recommended slopes range between 15:1 and 20:1. In general, the larger particle size material would require a steeper slope. All pen slopes were constructed at 20:1.
   
   
8. Use smaller structures to allow expansion of farm buildings in modular units rather than one large structure.
   
   

Table 2. Carbon Loading Dynamics and Dry Matter Changes.

a MNH=macadamia nut husk, TT=tree trimmings, GGH=guinea grass hay.

Table 4. Comparative Hydrogen Sulfide Levels of the Dry Litter System versus Conventional System.

Table 5. Composition of carbon material and co-compost.

a MNH=macadamia nut husk, TT=tree trimmings, GGH=guinea grass hay.

a Includes connection charges and levy after 12,000 gallons per month.