| one publication added to basket [312673] | An empirical remote sensing algorithm for retrieving total suspended matter in a large estuarine region
Camiolo, M.D.; Cozzolino, E.; Dogliotti, A.I.; Simionato, C.G.; Lasta, C.A. (2019). An empirical remote sensing algorithm for retrieving total suspended matter in a large estuarine region. Sci. Mar. (Barc.) 83(1): 53-60. https://dx.doi.org/10.3989/scimar.04847.22a In: Scientia Marina (Barcelona). Consejo Superior de Investigaciones Científicas. Institut de Ciènces del Mar: Barcelona. ISSN 0214-8358; e-ISSN 1886-8134, more |   |
| Keyword | | | Author keywords | empirical algorithm; suspended particulate matter; Río de la Plata; in situ measurements; MODIS-Aqua satellite image |
| Authors | | Top | - Camiolo, M.D.
- Cozzolino, E.
- Dogliotti, A.I., more
| - Simionato, C.G.
- Lasta, C.A.
| |
| Abstract | The Río de la Plata is a large, shallow estuary located at approximately 35°S and flowing into the southwestern Atlantic Ocean. It carries a high amount of nutrients and suspended particulate matter, both organic and inorganic, to the adjacent shelf waters and is considered among the most turbid estuarine systems in the world. Knowledge of the concentration and spatial and temporal variability of these materials is critical for any biological study in the Río de la Plata. In this work, the relationship between suspended particulate matter and turbidity is empirically established in order to derive suspended particulate matter maps from satellite data (MODIS-Aqua) for the Río de la Plata region. A strong correlation between suspended particulate matter and turbidity was found (Pearson correlation coefficient =0.91) and the linear regression (slope =0.76 and intercepts =12.78, R2=0.83) explained 83% of the variance. The validation of the empirical algorithm, using co-located and coincident satellite and in situ measurements, showed good results with a low mean absolute error (14.60%) and a small and positive bias (3.04%), indicating that the estimated suspended particulate matter values tend to slightly overestimate the field values. |
|
