From my experience with soil humidity measurements your trial is very interesting. But you will find some problems.
In my opinion your circuit isn't Wheatstone bridge, maybe even it isn't bridge. As I found you even don't try to balance it. The main advantage of your circuit is that it works
and it is the most important.
From the beginning. From the picture I found you use tantalum capacitor. Check your cap datasheet, but most of tantalum cap are capacitor up to 1 - 10 MHz. Above there is inductance dominant. Fortunately you have used two similar caps, so for balance it doesn't matter. But I would thought because there is high temperature dependency in such cap parameters.
You don't care about balance of bridge or even you are out of balance, so you don't have to care about imaginary parts of impedance. For now you have to think about final acquisition method, because your meter is the part of circuit and we don’t know what happened with signal. In my opinion you measure something proportional to modulus of impedance with integration constants of a few Hz (DC meter range ???)
I think you should read more about soil humidity measurement methods – especially FDR and TDR. The most important problem is that on such frequencies there are scale problems – sample should be big enough to get proper results, so with small pile of soil you get quite different results than on the field. So you should perform measurements in at least flowerpot (from dielectric material !!!).
From the beginning you should make board for your circuits, because parameters of your circuits depends on “geometry” – with “spider” circuit you can get unrepeatable results.
If you would like to make humidity indicator, I think you have one. Now you should perform measurements in different humidity check characteristics – especially linearity and repeatability with different soils.
If you plan to make humidity meter you should carefully rework your circuit and chose elements to work near bridge balance (real and imaginary parts) to avoid nonlinearity and even saturation.
From my point of view, for now match of circuit impedance it is your smallest problem