Dealing with wet silt
At the planning stage we discussed the following alternatives with contractors:
(1) Using a pump which would take silt directly to a disposal site, which would have to be a nearby field. This could be up to 250 metres away. It would need the permission of the farmer. It would also destroy any remaining wildlife (though in fact there was very little in the pond by this stage.) It would have the advantage of keeping the existing size of the pond. Along two sides of the pond the mud contained a lot of clay and we were dubious about whether this area could be pumped. In fact, once our pond had dried and the mud consolidated, it was clear that pumping any of it would have been impossible.
(2) Excavating the whole pond with machinery and dumping the silt round the edges to shrink it, and building a timber revetement to hold the spoil back (which was also what we did ourselves). Oak was suggested for the posts, but this proved too expensive when we costed the job ourselves.
(3) Excavating with machinery and shrinking the pond in the same way as (2), while placing the spoil behind an artificial bank of geotextile, held under tension by wooden stakes inserted through pockets in the textile. The brand name of the textile proposed was Nicospan. This method was due to be trialled on the Norfolk Broads for reinforcing river banks, and the Broads authority’s engineering department, which we contacted, gave us some positive views on it.
(4) In combination with any of the above, excavating a trench in the garden and using the earth dug out to build a bund round the trench, then filling the resulting it with some of the silt, covering with earth and leaving for a year to settle and drain before spreading it round the garden. This would be useful for very liquid silt.
(5) Using geotextiles to build a bund instead of a trench to contain liquid silt. The water would drain out through the textile. The silt would later be spread round the garden.
(6) Shrinking the pond by dumping the excavated silt around the shallow edges of the pond, then using logs from felled trees from the garden to contain the silt. This seemed ecologically very sound. However, the trees we needed to take down were ash, which rots very quickly in water and would disintegrate within a very few years. We weren’t convinced that this would really do a good enough job of containing the silt, and the contractor who proposed it could not show experience of work on ponds of this size.
(7) Shrinking the pond and building a wall entirely of sandbags filled with concrete to contain the silt. When we costed sandbags, the saving in material costs was modest if there was a saving at all, and the labour involved much greater than building a timber containment. But we did use this method for two short sections where the bottom had a pronounced curve. It is also useful if the containment has to be high – with timber above 75 cm we would have needed to use much bigger stakes to hold it back. In general, we preferred the look of timber.
There was a patch of ground at one end of the pond that flooded in winter, so we decided to use that as a trial area when the pond dried out in September 2009.
- We dug out about a fifth of the pond’s area by hand and wheelbarrow, and dumped it on the wet land.
- Since the digging was not at the deep end of the pond, we removed no more than a tenth of the volume of mud it contained.
- We used scrap timber to shore up the edge of the mud so that it did not slide back into the pond when it refilled, and waited to see how the spoil settled over the winter.
- The operation raised the level of the patch of land by about a metre, but after two years of compaction and drying, the composted silt was down to perhaps a third of a metre deep and supported a thriving growth of grass and weeds, so it was a successful reclamation.
The test gave us a sharp lesson in what not to do:
- When the pond dried out again two years later – a very wet summer intervened – we found our entire excavation of several hundred wheelbarrow loads of silt had filled in again.
- The mud had flowed back from the rest of the pond, and this new deposit was much more liquid, and so harder to handle.
- The lesson was that you cannot half clear a pond. You either have to finish the job in one season or build a barrier to protect the cleared area from the rest of the pond until the next season
- Furthermore, partially clearing the pond seems to have released toxins that wiped out much of the remaining life in the pond. The smell of the gas rising from the silt was very unpleasant.
- With perhaps 10 times as much mud again to take out, the experiment convinced us that we could not dump the rest of the spoil in the garden because of the sheer volume.
- With local farmers ruled out, that left only one place – round the edge of the pond. But we were reluctant at first to do that, because it meant a significant shrinkage of its area. We changed our minds when we realised that the old pond had been smaller, and that it had gradually expanded and shallowed as the banks crumbled under the assault of cows, ducks and frost over the years. Evidence for this was that the silt near the edges had a lot of yellow garden clay mixed into it.
- The main drawback of stacking silt round the margins of the pond is that if the mud is full of nutrients then they will gradually leach back into the water. Too much nutrient in a pond can cause algal blooms and retard other plant life. Experts such as the Suffolk Wildlife Trust advise against this. However, the expense of moving the mud elsewhere and the impossibility of finding space for all of it in the rest of the garden left the margins of the pond as the only practical disposal site.
We needed to be sure that there was nothing too unpleasant in the mud:
- The black surface went brown in an hour after it had been removed from the pond, as the oxygen got to it for the first time since it had accumulated at the bottom.
- By the following spring, it looked and smelled like good garden soil and some of the surplus was distributed on the vegetable patch, which thrived. That was a good enough test for us.
We quantified the job, to see whether we could manage it ourselves:
- The length of the edges was measured with a cord calibrated with knots at 2 metre intervals. This produced an approximate plan, and an area of about 350 square metres.
- Surveying the depth of mud was easy round the edges, using a 2 metre stick with notches at 10 cm intervals. In the middle, because it was impossible to get close, we used a very crude measure that gave a usable estimate of depth. So:
- We cut a 6 metre straight ash pole from a hedge, we calibrated the pole with notches, and then we probed for the bottom from the side while keeping the pole at as near a constant angle to the surface as we could.
- We reproduced the angle of the pole to the surface on flat ground and measured the vertical distance to the point on the pole that the mud had reached.
- We then cleaned the mud from the pole and did it again in another spot.
- It sounds horribly unreliable. But after the pond was completely cleared of mud we could see that this crude method had been accurate to about 10 cms, which was good enough. The result: maximum mud depth 1.2 metres (it actually proved to be 1.3 metres), average depth 50- 60 cms and total volume of mud around 175 cubic metres.
The decision to reduce the size of the pond allowed 100 sq metres of mud to be left in place, with new spoil dumped on top. Since the mud was much shallower round the edges, this reduced the total volume to be shifted by only 20 to 30 cubic metres, to around 150 cubic metres (these numbers are all rough estimates).
Trials showed that it was feasible to shift 1 to cubic 2 metres an hour by shovel (depending whether a wheelbarrow was needed).