**1. Introduction**

The Water Framework Directive (WFD) [1] is a framework by European Commission that requires Member States to monitor each relevant biological quality element (BQE) in order to assess the ecological status of each water body (WB), which represents the classification and managemen<sup>t</sup> unit of the WFD [2]. Macroalgae, phanerogams, macroinvertebrates, and fish faunal are the BQEs to be evaluated for the assessment of the ecological status of European transitional WBs under the WFD. In addition to BQEs, the physicochemical and hydromorphological supporting elements contribute to the ecological classification, confirming or not the classification provided by the BQEs. Biological monitoring results have to be expressed as ecological quality ratios (EQRs), by comparing sampled data with those equivalents from undisturbed or minimally disturbed reference sites. Depending on the scale of deviation from reference conditions, the EQR range is divided into five status classes: High,

good, moderate, poor, and bad. The most important distinction is between good and moderate status: the first is defined as a "slight deviation from reference condition", whilst the latter is a "moderate deviation from reference conditions". When the quality status is less than good, Member States/River Basin District Authorities must adopt a plan of measures to improve a WB until good status is achieved [1,2].

The WFD also requires all Member States to provide information about the estimation of the confidence interval and precision attained by the monitoring system used in their classification [1] (Annex V). As the critical good/moderate boundary can require expensive remedial measures to be included in the River Basin Management Plan, it is important that water body monitoring and managemen<sup>t</sup> organizations estimate the confidence to which an individual WB can be assigned to an ecological status class [3]. The biological, physical, and chemical data from current monitoring programs underlie large sources of uncertainty, which can be partitioned into spatial variations, temporal variations, sampling, and analytical errors, in addition to the uncertainty introduced by the classification system, especially by the reference conditions [4]. Some authors proposed assessment systems based on statistical principles to provide information on the level of confidence and precision of monitoring program results, focusing on field sampling variability and laboratory procedures and protocols [3–7]. Mascarò et al. [8] investigated the sources of variability associated with sampling design of a selection of macrophyte-based (both macroalgae and seagrasses) monitoring programs and observed that uncertainty is mainly a ffected by spatial variation. The authors consequently suggested that spatial uncertainty analyses be performed to optimize sampling strategy and to improve the reliability of classification of ecological status [8].

The MaQI (Macrophyte Quality Index) is the macrophyte index adopted by the Italian Law (Italian Ministry decree 260/2010) in agreemen<sup>t</sup> with the WFD requirements for ecological classification in transitional waters, and it was successfully intercalibrated in the framework of the WFD intercalibration exercise [9]. The MaQI was applied to assess the ecological status by macrophyte assemblages in the Venice Lagoon, Italy's largest wetland and one of the most important coastal ecosystems in the whole Mediterranean basin [10–12]. During the first cycle of the WFD operational monitoring, 114 stations were monitored in the Venice Lagoon. The ecological status classification of the Lagoon, resulting from the above monitoring, has been adopted by Veneto Region as published as Decision n. 140/2014 in the first revision of the Management Plan of the hydrographic district of the Eastern Alps [10]. However, for subsequent assessments, due to limitations of budget, optimization of the monitoring strategy was needed. Reducing monitoring e fforts is a common consequence of cuts in governmen<sup>t</sup> spending, adopted by many countries to save money in order to face the current global crisis [13]. Therefore, the balance between sampling e ffort and classification confidence becomes a critical issue in implementing the WFD [14,15].

In this context, the aim of this study was first to assess the reliability of ecological status classification in transitional waters, using results from monitoring of macrophyte assemblages in the Venice Lagoon, as a case study. Then, a multi-approach method to optimize the monitoring strategies minimizing both sampling e ffort and risk of misclassification was proposed, based on inferential statistical principles, geostatistical analyses, and expert judgment.

### **2. Materials and Methods**
