The lab simulation of contaminated soils is very difficult than the wastewater environment simulation. Because, the simulation in the soil is strongly related to the complex system of interactions in contaminated soil:

For instance, INTERACTIONS:

=> b/w particluar components of soil 

=> bw the soil components and the contaminant components

=> b/w microorganisms and soil

=> b/w microorganisms and contaminants

=> b/w the tellurian microorganisms itself

Hence, soil remediation is comparatively difficult and should be approached with a different way.

From another perspective

in water you have interfaces water/air and respective interfaces with microorganisms, contaminant (solid or liquid) interface with water and with microorganism, in addition dissolution and uptake of dissolved matter (contaminant/air) kinetics. This maybe simplified but is already complicated.

Adding additional phase leads to exponential growth of interdependencies. Not to forget complex composition of soil. Degree of bioavailability is different depending on location of contaminant.

In addition to the second answer,

Soil is a heterogeneous system while water is a homogeneous. These for, remedation in water is more rapid.   

Soil is a very big medium involving vast number of interacting factors that makes it difficult.

Soil source is a complex heterogenous material which involves many dependent and associated components like microorganism, pH, temperature, moisture, nutrients etc. It requires a longer incubation time for all factors to synchronize and remediate a potent pollutant in real time. One can simulate in a laboratory conditions as a prototypical work, but it involves various interacting factors.. 

Many pesticides are strongly bound to clays and humic complexes by chemisorbtion and are not as available for chemical reactions. Microbial enzymatic attack is inhibited by many surface interactions, and interference in mass transfer. These would be source of much difficulty.

Also many pesticides are light sensitive and degrade more easily by same reasonings interefering with light is much greater in soils. A chemical look at the recalcitrant analytes, could clue you in on the modes available for degradation.

Precision and repeatibilty of experiments would be result of heterogenous nature of soil microenvironments and microbial makeup. This is what makes scaleup from laboratory results difficult to predict. 

in the first instance, the amount of oxygen present in the soil is very limited thereby making bioremediation an anaerobic process the by-products of which still need to oxidized to get a stable product. Also, many of the micro-organisms responsible for the remediation can survive these by-products except the soil is agitated for aeration. Couple with this, bioremediation is a temperature dependent process and the effective temperature range may not be easily achievable within the soil matrix where temperature may be higher than usual.

Soil is composed a large number of chemicals. Therefore measurement of  a special contaminant in soil is not as easy as in water. However, it is only one reason for it. 

in case soil is saturated with water, yes, then usually environment is anaerobic, however, when the soil is only moist usually environment is aerobic.

Soil is not only a heterogeneous system but it is also a medium where a lot of chemicals lie in water insoluble forms and petrified form. This makes the system very difficult for testing in ideal laboratory conditions. Still there are a lot of process employed for testing various parameters of soil like the process mentioned by Malo, 1977.

Hope this will help you to understand the reason.

Thanks.

In bio/phyto remediation of soil the most critical area of concern is the rhizosphere where most of the interactions and exchanges takes place. This area needs to be researched.

I agree with M.Jamil Khan

Oil is readily dispersed in water than in soil. This dispersion increases the surface area to volume ratio of the oil for biodegradation.

soil is heterogeneous system while water ishomogenous

soil compare to water involved some challenge for bioremediation: low level of oxygen, lower temperature, nutrient less available. Contaminants can be strongly to soil particle or black carbon (i.e. soot, ash) and thus not accessible to bacteria. In addition, bacteria can not move so easily in soil than in water, therefore if the bacteria and the contaminant are not in direct contact, the biodegradation cannot start. This problem does not occur in water where turbulence will homogenize the medium and bring bacteria and contaminants in close contact. Moreover, lower level of oxygen and nutrient availability will induce a low bacterial density. Finally, numerous bacteria live in the soil and will compete with degrader bacteria.

Because unlike soil, irrigation water can be fully contained and treated prior to usage. Whereas containment options for soil is limited to protected planting when pots are used for example. However, once potted plants are placed in the filed, the effect of containment and treatment will start to fade out as the potted soil and plant roots come in touch with the field soil. There are of course other factors due to the fact that soil is solid and much more complex than water and achieving an even distribution and uniformity of treatments is much harder, and is more often than not it is impractical.

Dear Kunal,

Soil is totally heterogeneous system,. If you talk about bio-remediation of heavy metals or other contaminants from soil, then you have to go for Sequential exatration of heavy metals or contaminants. In soil heavy metal is bound in mainly four form: (1) Soluble/ extractable form (2) Bound to carbonates (3) Bound to iron and manganese oxides  (4) Bound to organic matter (5) residual fraction while water ishomogenous. So so can easily see how complex is the soil as a system of soil-metal complex.

    Moreover, in soil availability of heavy metals for plants or microbes is very very less as compared t water.

Because soil is more dense and will have an uneven distribution of both good and bad materials. Water is much more conducive to bio-remediation because 1) it can be managed more readily due to flow characteristics that soil doesn't have unless infused with large amounts of water,  2) bio-active materials such as bacteria or algae can be diffused much more thoroughly and rapidly through a mass of water with a relatively even distribution, and 3) optimal conditions for bio-remediation can be more effectively induced uniformly throughout the water body than through a disparate medium like soil (temperature, pH, etc.)