P. nuntia populations in tub-based trials were quantitatively shown to
reduce the organic matter in sand bed surface layers (from 4–5% to 1–2%),
since worm biomass levels (0–112 g m−2
) were negatively correlated
with the residual organic content of the sand surface (R2=0.811,
Pb0.01). By contrast, organic matter levels in subsurface (core) sand
samples (1–2%) were not affected (PN0.05) by P. nuntia densities (Fig. 2).
This organic matter clearance from the leading (upper) surface of
sand filters facilitated the maintenance of prescribed filtration
(percolation) rates. Water depths above operating sand filtration
beds tended to remain lower when high densities of polychaetes were
present, indicating the bed's lower resistance to percolating water.
This was clearly shown in standardised (equal mid-tidal flows) tubbased
trials with 14-week-old P. helleri (see Fig. 3). Tubs without
worms operated with much higher (Pb0.05) water depths than those
with worms (≥1000 m−2
), and this effect was most pronounced with
the highest stocking density (4000 m−2
). Although there was no
evidence of P. helleri densities (2000 and 6000 m−2
) differentially
affecting water depths or percolation rates in the larger-scale farm
trial in 2007, the volumes of pond water filtered by the six sand beds
demonstrated a consistent ability to handle the nominated commercially
relevant water treatment rates (1500 L m−2 d−1
).
3.3. Treated water qualities
Different water percolation rates (approx 1–4Lm−2 min−1
)
tested in the smaller scale tub trials did not differentially affect