Rhizoremediation experiments conducted by planting and monitoring microbial populations and/or contaminant levels over time are powerful approaches since they can be constructed and operated with replication, randomisation and controlled conditions. Perhaps the major limitation of planted studies is time, since for reasons of practicality planted studies are typically only run for 1-3 years. Because roots are estimated to occupy less than 1% of soil , it is reasonable to anticipate that many years of root exploration would be required to thoroughly treat a contaminated area. Rates of contaminant disappearance also indicate that rhizoremediation should be regarded as a long-term process . Nonetheless, losses in soil pollutant concentration and/or increased populations of degraders have been observed within several years or less [17-21]. In the event that pollutant disappearance occurs too slowly to detect, a comprehensive planted study still affords an excellent opportunity to investigate mechanisms of remediation, such as microbial population shifts at the root-soil interface.
When designing planted studies, as with any controlled experiment, careful and statistically conscious design is critical to ensure conclusive results. Replicates of each treatment including unplanted soil as a control should be included. Numbers of replicates should be as high as reasonably achievable without compromising the quality of the data since higher numbers of replicates produce more accurate results and a lower likelihood of producing false positive/negative results. Randomised arrangements of treatments, whether in pots or field plots, should be employed to ensure that differences observed are due to treatments rather than variations in light, soil composition, moisture, etc. associated with different locations in a greenhouse or outdoor area. Additional guidance may be found in books focusing on basic statistics  and experimental design . The following sections discuss additional design considerations specific to pot or field plot studies.
Performing studies in containers is an immediately attractive option because it affords greater homogeneity than can be achieved in an intact field site, keeps plant roots contained and therefore maintains discrete species-specific influences on soil. However, pot studies are subject to criticism regarding their ability to realistically replicate natural conditions that may impact rhizoremediation processes, especially when conducted in a greenhouse.
Pot experiments performed in a greenhouse are convenient in that they provide independence from seasonal limitations on outdoor research. Unfortunately, plants are shielded from environmental cues that influence plant physiological responses important to some rhizoremediation mechanisms. One issue is that greenhouse glass filters out a portion of ultraviolet light from solar radiation. Ultraviolet light is reported to promote the production of plant secondary compounds , which are thought to play a major role in biostimulation and/or induction of aromatic pollutant degrading bacteria  (detailed in the following chapter by Andrew C. Singer). If studies are conducted in the greenhouse, supplementation with UV light is recommended since it has been demonstrated to increase the concentrations of phenolic compounds produced by plants to levels similar to those obtained under outdoor growth conditions .
Greenhouse studies also shield perennial plants from seasonal signals that dictate cycles of senescence and dormancy, including tree root turnover events . Root turnover may be an important mechanism in the delivery of secondary plant metabolites that support the growth of PCB-degrading bacteria . If a concern, this can be remedied by running the pot study outdoors where plants can be exposed to natural environmental conditions. For additional realism, pots may be buried in the soil to help maintain ambient soil temperature. Whether indoors or outdoors, root growth in pots is often very different from bulk soil. Roots tend to grow along the pot edges, coil along the bottom, and may be significantly less branched than field-grown roots, which could deleteriously effect rhizoremediation results (John Fletcher, 2005, personal communication).
Pot studies are advantageous over field studies since they afford the opportunity to minimize heterogeneity of contaminant concentrations. To take advantage of this, care should be taken to homogenize soil thoroughly in a large soil mixer or similar equipment before filling pots, since soil either collected from a contaminated site or manually spiked with pollutants tends to be heterogeneous. At or prior to the initiation of the study, chemical analyses of replicate samples should be performed to assess the variability of pollutant concentrations within and among pots and help determine adequate sample replication. When natural environmental soils are used without additives to provide loose texture (i.e. vermiculite), they can become very compacted after continued watering. If considering the addition of organic material to the potting mix, note that some materials (i.e. bark mulch) contain high concentrations of plant aromatic compounds that can interfere with gas chromatographic analyses of polyaromatic hydrocarbons in soil as well as impact microbial communities.
Another consideration with pot studies is that contaminants may leach out of drainage holes posing not only a chemical hygiene problem but also creating a false positive response when chemical disappearance data are analyzed. Care should be taken to avoid excessive watering, and leachates should be collected for chemical analyses especially if working with water-soluble contaminants. Unplanted controls introduce unique complications with watering regime. If watered uniformly with planted treatments, soil in unplanted pots may remain saturated, generating hypoxic conditions distinctly different from soil conditions in planted pots. This compromises important control data for background levels of microbial activity as well as for volatilization of pollutants through soil pores. A solution to this problem is to maintain soil moisture at similar levels rather than to use equal water volumes, and to monitor outflow for escaping contaminants.
Planted field plots afford the opportunity to evaluate the rhizoremediation potential of a variety of plant species under more realistic environmental conditions than pot studies. While heterogeneity of soil contamination is often greater, it can generally be overcome with a randomised layout such as Latin square or randomised block design and careful statistical analyses [22, 23]. Many precautions mentioned in the previous section also apply, including attention to the watering regime and chemical analysis of preliminary samples to determine contaminant variability.
A unified experimental design strategy for field investigations of rhizoremediation has been developed by the U.S. Environmental Protection Agency's Remediation Technologies Development Forum (RTDF) Phytoremediation Action Team. The strategy was designed to evaluate the efficacy of various agricultural and non-crop herbaceous plants for the rhizoremediation of weathered petroleum hydrocarbons. Use of the unified protocol permits direct comparisons of data by researchers in a wide range of geographic locations and climates. Having been carefully developed and debated by numerous researchers, the RTDF experimental design provides a useful general framework for field plot studies, and may be easily adapted to study other plants or contaminants. The experimental method involves three different planted treatments in a randomised block design with each block being a minimum of 6.1 m square in size. The three treatments are 1) a mixture of species optimised for local conditions which may include grasses and trees 2) an unplanted control and 3) a standard seed mixture, including some flexibility in plant species to accommodate local conditions, of 10-15% rye, 20-25% legume and 60-70% fescue. Additional blocks may be added with plants of particular interest to the investigator. Plots are planted and monitored over the course of 3 growing seasons for chemical disappearance and microbial populations. Detailed sampling methods, chemical, microbial and plant analyses are provided in the protocol, which is available at http://www.engg.ksu.edu/HSRC/appa.html.
Strategies for sampling forensic or planted rhizoremediation studies are an integral component of experimental design, and ultimately will impact how accurately the data reflect true conditions. The first step in developing a sampling strategy is to define the compartments of interest within the complex root-soil interface (rhizosphere, rhizoplane, bulk soil, etc.) and select a sampling method that is both appropriate and feasible. Multiple samples must be collected from each replicate treatment in a representative manner, with consideration of temporal and spatial issues. Lastly, samples must be transported and stored for microbial or chemical analyses in ways that will minimize the introduction of artefacts.
Was this article helpful?
You Might Just End Up Spending More Time In Planning Your Greenhouse Than Your Home Don’t Blame Us If Your Wife Gets Mad. Don't Be A Conventional Greenhouse Dreamer! Come Out Of The Mould, Build Your Own And Let Your Greenhouse Give A Better Yield Than Any Other In Town! Discover How You Can Start Your Own Greenhouse With Healthier Plants… Anytime Of The Year!