Harry A. J. Hoitink and Matthew S. Krause
Dispersal of plant pathogens such as Pythium and Phytophthora spp. with irrigation water has caused problems in nurseries for decades. Many procedures have been tested to control this problem. They include chlorination, ozone treatment, heating, ultraviolet (UV) irradiation and others. None of these procedures has been adopted successfully in nurseries. The volume of water used in nurseries simply is too large for some methods. Other methods have proved ineffective or have even aggravated the problem.
In the late 80s, a very old but effective procedure known as slow sand filtration was introduced into nurseries and greenhouses in Geisenheim, Germany, through the research of Dr. W. Wohanka. He determined that this system effectively filtered bacteria and fungi from irrigation water. Several plant pathologists from Europe and Australia have verified the effectiveness and the practical value of this filtration procedure for nursery irrigation systems.
Slow-sand-filtration systems that filter 200,000 gals/day are being used in nurseries today. In Germany, a survey of growers revealed that the quality of nursery crops was improved after these filters had been installed. German and Dutch nurseries have used this technology extensively during the past five years. These nurseries, however, predominantly use container media prepared with Sphagnum peat. Peat media are conducive to Phytophthora root rots, and this allows for the development of high pathogen populations. Most American nurseries use media prepared with tree barks or other types of composted products, and these mixes, when formulated properly, suppress Phytophthora root rots. Unfortunately, these suppressive effects can be overcome by high salinity and this allows Phytophthora spp. to increase in populations.
During prolonged periods of dry weather, the volume of irrigation water available in nurseries often becomes limiting. In addition, the salinity goes up, and it is under these conditions that Phytophthora root rots become most serious. In conclusion, during periods of drought when high-in-salinity irrigation water is recycled onto crops susceptible to Phytophthora root rots, the potential for transmission of these pathogens with infested irrigation water to noninfected plants becomes a problem. Slow sand filtration is designed to control this problem.
Crops in the Ericaceae family unfortunately are affected not only by root rots but also by Phytophthora dieback diseases. They affect the foliage, the stem, and less so, the root system. The Phytophthora spp. that cause these diseases – such as P. cactorum, P. citricola, P. citrophthora, P. parasitica, etc. – sporulate heavily on lesions during high-humidity and high-temperature (75°F or higher) weather conditions. They are spread from plant to plant as inoculum in splashing irrigation water droplets. The plant surface must be wet for at least two hours for infections to occur. Thus, under overhead irrigation, these pathogens can spread within crops even if filtered water were to be used. These dieback diseases do not occur in dry summers in nurseries when leaves dry quickly after overhead sprinkler irrigation.
One way to reduce the severity of Phytophthora die-back diseases is to produce such plants under drip irrigation. Ohio nurserymen have begun to produce rhododendron and lilac, which are highly susceptible to Phytophthora dieback, under drip irrigation in pot-in-pot systems (above ground as well as buried pot systems) using container media naturally suppressive to root rots. The crops also are sprayed with fungicides to further suppress these problems (see OSU Extension Fact Sheet HYG-3073-99 entitled Control of Phytophthora and Other Major Diseases of Ericaceous Plants). This integrated approach has reduced the severity of both types of diseases. The question now becomes whether slow sand filtration can reduce symptoms further in plants that do not show symptoms (latent infections). An answer to this question is not available presently. However, it seems that slow sand filtration should be applied to those crops most susceptible to Phytophthora root rot and dieback diseases. Furthermore, large plants of such crops should be produced under drip irrigation in ground beds or in pot-in-pot systems. Finally, fungicides described in the previously mentioned fact sheet should be applied correctly to further reduce these diseases. Other factors, such as avoidance of puddles on the base where containers are placed, pre-filtration, and settling of irrigation water, should also be included as part of the overall management program.
The principle of a slow sand filter is very simple. Water must be treated before entering a filter to remove as much silt and organic matter as possible. The pre-treated water then is filtered very slowly through a deep bed of sand (Figure 1). The flow rate depends on the size of the microorganism that needs to be removed. For Phytophthora spores this means a flow rate of two to three gallons per square foot filter surface area per hour. An irrigation pond, within which a 1,000 sq-ft filter is installed in the deepest end, could filter 50,000 to 75,000 gals of water per day.
|
| Figure 1. Diagram of a slow sand filter (adapted from W. Wohanka, Geisenheim, Germany). |
| Table1. Quality Requirements for Filter Sand (Adapted from W. Wohanka, Geisenheim, Germany). | |
|---|---|
| effective grain size | 0.15-0.30mm (100 - 50 mesh) |
| uniformity coefficinet (UC) | < 3, maximum 5 |
| silt content | < 1% |
| acid solubility | < 5 % after 30 min |
| effective grain size (d10): sieve opening through which 10% (by weight) of the grains will pass. | |
| uniformity coeffecient (UC): ratio between the sieve opening through which 60% (by weight) of the grains will pass and the effective grain size; UC = d60/d10 | |
Soon after the filtering process begins, a layer of brown material develops in the surface inch layer of the bed. Pseudomonas, Trichoderma, and other microorganisms already known as biocontrol agents in compost-amended container media, seem to slowly destroy the pathogens immobilized through filtration in this layer. Filters are not very effective until after this layer develops, and this requires several weeks. The layer must not be disturbed by water added to the top of the filter. Therefore, a three-feet or deeper layer of water is always maintained on top of the filter. Water must be added to the reservoir through a piping and sprinkler system that minimizes turbulence.
The depth of the sand bed should be three to four feet, and it must be at least two-feet deep. A depth of four-feet allows periodic cleaning of the filter by removal of the surface layer, if flow rates go down due to clogging. The type of sand that must be used is explained in Table 1. Underneath the sand are several layers of gravel and a drainage system. Rockwool (Grodan type 01251 or 012519) has performed better in nurseries than sand, because it becomes clogged less easily than sand filters. It is more costly, but gravel does not have to be used under Rockwool filters.
Below either type of filter, a system of conventional four-inch drainage tubes drains filtered water into a sealed concrete reservoir in the lowest point of the pond filtration system. Water is then pumped up out of this reservoir at a controlled flow rate to a storage facility or the irrigation system. A flow meter with feedback control is essential for effective operation of large pond systems.
The pump must always recirculate some water through the system when the irrigation system is not in use. Without circulation, anaerobic conditions develop in the filter, and this also destroys effectiveness.
The cost of slow sand filtration is very low as compared to other technologies. One California nursery has used a 200,000 gallon-per-day system for three years without problems and is installing a second to further reduce losses caused by plant pathogens for crops not yet treated with filtered water.
For information on Phytophthora control and slow-sand-filtration systems, see the references listed here.