Image processing techniques applied to the study of Lake Trasimeno, Italy

This study presents the contribution of remote sensing to improve the environmental knowledge of Lake Trasimeno (Central Italy) and surrounding basin, in order support the local authorities (ARPA Umbria) in implementing the basin management plan. Remote sensing data were acquired to analyze water quality and coastal vegetation status, together with their interactions and connection with land cover and use dynamics of the surrounding areas through the last 30 years. Different satellite sensors have been considered in this study: (1) high revisiting time sensors as MERIS and MODIS for coarse scale regular monitoring of the lake water quality (2005-2008); and (2) high/medium spatial resolution satellite sensors (e.g., Landsat) for intermediate/fine scale studies on aquatic vegetation and surrounding lands (1979-2008).
Time-series MERIS data from 2005 to 2008 were processed to retrieve water quality parameters according to physically based approaches (calibrated using in situ data from 2008). The image-derived water quality trends were in agreement with in situ measurements (r>0.8). Satellite ALOS and ASTER data integrated with spectroradiometric in situ data were elaborated in order to investigate the common-reed area. Satellite-derived vegetation indexes (e.g., NDVI) were adopted to assess the vegetation status. The Leaf Area Indexes for about 12 sites collected in 4 campaigns of summer 2008 allowed to understand the differences of common reed conditions in relation to water levels and sediment quality observed in the different visited sites. A multisource mid to high resolution dataset, ranging from Landsat MSS to Landsat TM/ETM+ to Terra ASTER and ALOS AVNIR-2 scenes, has been exploited for multitemporal study of land cover change covering the period from 1979 to 2008. Satellite scenes were subjected to radiometric normalization and then analyzed to map land cover changes through time. The analysis was focused over agricultural areas in order to investigate the consequences of changing on the water quality status of the lake.
This study showed the capabilities of multi-sensor satellite data to derive practical data about the complex and heterogeneous environment of Lake Trasimeno basin. We expect that the presented study may be useful to the project, developed by ARPA Umbria, aiming to refine the environmental knowledge of the study area and hence to define a conceptual model for the basin management plan.