Measuring the effects of bivalve mariculture on water quality in northern New Zealand using 15 years of MODIS-Aqua satellite observations

ABSTRACT: This study demonstrates that satellite information can be used to quantify the effect of a large coastal mussel farm on water quality. Our study focussed on a 1400 ha area of maricultured green-lipped mussels Perna canaliculus in the Firth of Thames in northern New Zealand. Observations by NASA’s MODIS-Aqua sensor were used to estimate the concentration of chlorophyll a (chl a, as a proxy for phytoplankton), turbidity and sea surface temperature at 500 m resolution. We used approximately 890 single-pass satellite images, each with <50% cloud cover, spanning 2002 to 2017. For each image, the area over the mussel farm was removed and then filled in using kriging with parameters derived from semivariogram analysis of each individual image using linear and exponential models. Three control areas were used to test whether the differences between the kriged data and the observed data in the farm area were significant. We found that, over the 15 yr of the observation, the farm had a significant effect on chl a of -1.6% (reduction) averaged over an area of 1.5 times the area of the farm itself. No significant effect of the farm on turbidity was detected. The farm caused a seasonal pattern of winter warming (+0.03°C) and summer cooling (-0.11°C). The method described here could potentially be applied to other satellite data and other large (multi-kilometre) mariculture facilities to observe their effects on water quality.