Giuseppe Passarella

A methodology for treating missing data applied to daily rainfall data in the Candelaro River Basin (Italy)

Environmental time series are often affected by the “presence” of missing data, but when dealing statistically with data, the need to fill in the gaps estimating the missing values must be considered. At present, a large number of statistical techniques are available to achieve this objective; they range from very simple methods, such as using the sample mean, to very sophisticated ones, such as multiple imputation. A brand new methodology for missing data estimation is proposed, which tries to merge the obvious advantages of the simplest techniques (e.g. their vocation to be easily implemented) with the strength of the newest techniques. The proposed method consists in the application of two consecutive stages: once it has been ascertained that a specific monitoring station is affected by missing data, the “most similar” monitoring stations are identified among neighbouring stations on the basis of a suitable similarity coefficient; in the second stage, a regressive method is applied in order to estimate the missing data. In this paper, four different regressive methods are applied and compared, in order to determine which is the most reliable for filling in the gaps, using rainfall data series measured in the Candelaro River Basin located in South Italy.

MSANOS: Data-Driven, Multi-Approach Software for Optimal Redesign of Environmental Monitoring Networks

Within the recent EU Water Framework Directive and the modification introduced into national water-related legislation, monitoring assumes great importance in the frame of territorial managerial activities. Recently, a number of public environmental agencies have invested resources in planning improvements to existing monitoring networks. In effect, many reasons justify having a monitoring network that is optimally arranged in the territory of interest. In fact, modest or sparse coverage of the monitored area or redundancies and clustering of monitoring locations often make it impossible to provide the manager with sufficient knowledge for decision-making processes. The above mentioned are typical cases requiring optimal redesign of the whole network; fortunately, using appropriate stochastic or deterministic methods, it is possible to rearrange the existing network by eliminating, adding, or moving monitoring locations and producing the optimal arrangement with regard to specific managerial objectives. This paper describes a new software application, MSANOS, containing some spatial optimization methods selected as the most effective among those reported in literature. In the following, it is shown that MSANOS is actually able to carry out a complete redesign of an existing monitoring network in either the addition or the reduction sense. Both model-based and design-based objective functions have been embedded in the software with the option of choosing, case by case, the most suitable with regard to the available information and the managerial optimization objectives. Finally, two applications for testing the goodness of an existing monitoring network and the optimal reduction of an existing groundwater-level monitoring network of the aquifer of Tavoliere located in Apulia (South Italy), constrained to limit the information loss, are presented.

A methodology for rapid assessment of the environmental status of the shallow aquifer of "Tavoliere di Puglia" (Southern Italy)

Anticipating the European Water Framework Directive (2000/60/EC), the Italian Government issued Legislative Decree n.152/99 which sets out rules for classifying the environmental status of national water bodies in order to achieve specific qualitative objectives by 2016. The most recent European Groundwater Directive (2006/118/EC), which was only recognized by Italy in early 2009 (Legislative Decree 30/09), requires such resources to be characterized from a qualitative standpoint and the risk of their being polluted by individual pollutants or groups of pollutants to be evaluated. This paper reports a simple methodology, based on easy-to-apply rules, for the rapid classification of groundwater, and the results of its application to the shallow aquifer of the plain of Tavoliere delle Puglie located in south Italy. Data collected during well-water monitoring campaigns carried out from 2002 to 2003 made it possible to assess the environmental status of the Tavoliere which, unfortunately, was found to be characterized by “significant anthropic pressures on quality and/or quantity of groundwater and necessitating specific improvement actions”.

Integration of electromagnetic induction sensor data in soil sampling scheme optimization using simulated annealing

Soil survey is generally time-consuming, labor-intensive, and costly. Optimization of sampling scheme allows one to reduce the number of sampling points without decreasing or even increasing the accuracy of investigated attribute. Maps of bulk soil electrical conductivity (ECa) recorded with electromagnetic induction (EMI) sensors could be effectively used to direct soil sampling design for assessing spatial variability of soil moisture. A protocol, using a field-scale bulk ECa survey, has been applied in an agricultural field in Apulia region (southeastern Italy). Spatial simulated annealing was used as a method to optimize spatial soil sampling scheme taking into account sampling constraints, field boundaries, and preliminary observations. Three optimization criteria were used. the first criterion (minimization of mean of the shortest distances, MMSD) optimizes the spreading of the point observations over the entire field by minimizing the expectation of the distance between an arbitrarily chosen point and its nearest observation; the second criterion (minimization of weighted mean of the shortest distances, MWMSD) is a weighted version of the MMSD, which uses the digital gradient of the grid ECa data as weighting function; and the third criterion (mean of average ordinary kriging variance, MAOKV) minimizes mean kriging estimation variance of the target variable. The last criterion utilizes the variogram model of soil water content estimated in a previous trial. The procedures, or a combination of them, were tested and compared in a real case. Simulated annealing was implemented by the software MSANOS able to define or redesign any sampling scheme by increasing or decreasing the original sampling locations. The output consists of the computed sampling scheme, the convergence time, and the cooling law, which can be an invaluable support to the process of sampling design. The proposed approach has found the optimal solution in a reasonable computation time. The use of bulk ECa gradient as an exhaustive variable, known at any node of an interpolation grid, has allowed the optimization of the sampling scheme, distinguishing among areas with different priority levels.

An automated decision support system for aided assessment of variogram models

In the present paper, an extensive cross-validation procedure, based on the analysis of numerical indices and graphical tools, is described and discussed. The procedure has been implemented in a software application designed to support practitioners in the variogram model assessment. It provides an extensive report, which summarizes a large post-processing stage and suggests how to interpret the performed analysis to rate the model to be validated. Besides classical accuracy indices, two new integrated tools based on the variogram of residuals are introduced, which take the spatial nature of the dataset into account. Finally, inspecting the summary report, the user can decide whether the considered model is satisfactory for his/her goals or it needs to be improved. Finally, a case study is presented related to the variogram assessment of groundwater level measured in a porous shallow aquifer of the Apulia Region (South-Italy). A new extensive cross-validation procedure is described and implemented.An automated decision support system for supporting practitioners with the variogram calibration and validation is presented.Two novel tools for the spatial correlation assessment are defined and implemented.

Cross-Calibration of Two Independent Groundwater Balance Models and Evaluation of Unknown Terms: The Case of the Shallow Aquifer of “Tavoliere di Puglia” (South Italy)

Groundwater represents an essential water resource for human purposes, mainly in those areas characterised by a scarcity of surface water and dry climate. Consequently, tools for assessing the groundwater balance are fundamental for its suitable management. The conventional groundwater balance equation, which considers all the natural and human-induced terms of the balance, such as rainfall, withdrawals, irrigation, etc., sometimes lacks of some important terms. One of the terms of the balance that is most difficult assess is the volume of water exchanged with other neighbouring water bodies (subsurface inflow/outflow). In this case, the estimation must be considered as a poor approximation. In this paper, a novel methodology is proposed that is capable of significantly increasing the accuracy of the groundwater balance when subsurface inflows and outflows are unknown. The improvement is accomplished by comparing two corresponding time series of annual groundwater balances assessed by means of different balance models. The first time series is evaluated by means of the conventional balance equation and the second one by directly estimating the groundwater volumes by means of geostatistical methods. Both these models are supposed to lack specific, even though different, information. Their comparison through simple statistical tools allows them to be calibrated and to recover missing average information. A study case is presented considering the inflow/outflow term and the specific yield as missing information for the conventional and the geostatistical approaches, respectively. The study area is the shallow porous aquifer of the Tavoliere di Puglia (South Italy).

GTest: a software tool for graphical assessment of empirical distributions’ Gaussianity

In the present paper, the novel software GTest is introduced, designed for testing the normality of a user-specified empirical distribution. It has been implemented with two unusual characteristics; the first is the user option of selecting four different versions of the normality test, each of them suited to be applied to a specific dataset or goal, and the second is the inferential paradigm that informs the output of such tests: it is basically graphical and intrinsically self-explanatory. The concept of inference-by-eye is an emerging inferential approach which will find a successful application in the near future due to the growing need of widening the audience of users of statistical methods to people with informal statistical skills. For instance, the latest European regulation concerning environmental issues introduced strict protocols for data handling (data quality assurance, outliers detection, etc.) and information exchange (areal statistics, trend detection, etc.) between regional and central environmental agencies. Therefore, more and more frequently, laboratory and field technicians will be requested to utilize complex software applications for subjecting data coming from monitoring, surveying or laboratory activities to specific statistical analyses. Unfortunately, inferential statistics, which actually influence the decisional processes for the correct managing of environmental resources, are often implemented in a way which expresses its outcomes in a numerical form with brief comments in a strict statistical jargon (degrees of freedom, level of significance, accepted/rejected H0, etc.). Therefore, often, the interpretation of such outcomes is really difficult for people with poor statistical knowledge. In such framework, the paradigm of the visual inference can contribute to fill in such gap, providing outcomes in self-explanatory graphical forms with a brief comment in the common language. Actually, the difficulties experienced by colleagues and their request for an effective tool for addressing such difficulties motivated us in adopting the inference-by-eye paradigm and implementing an easy-to-use, quick and reliable statistical tool. GTest visualizes its outcomes as a modified version of the Q-Q plot. The application has been developed in Visual Basic for Applications (VBA) within MS Excel 2010, which demonstrated to have all the characteristics of robustness and reliability needed. GTest provides true graphical normality tests which are as reliable as any statistical quantitative approach but much easier to understand. The Q-Q plots have been integrated with the outlining of an acceptance region around the representation of the theoretical distribution, defined in accordance with the alpha level of significance and the data sample size. The test decision rule is the following: if the empirical scatterplot falls completely within the acceptance region, then it can be concluded that the empirical distribution fits the theoretical one at the given alpha level. A comprehensive case study has been carried out with simulated and real-world data in order to check the robustness and reliability of the software.

Similarity indices of meteo-climatic gauging stations: definition and comparison

Space-time dependencies among monitoring network stations have been investigated to detect and quantify similarity relationships among gauging stations. In this work, besides the well-known rank correlation index, two new similarity indices have been defined and applied to compute the similarity matrix related to the Apulian meteo-climatic monitoring network. The similarity matrices can be applied to address reliably the issue of missing data in space-time series. In order to establish the effectiveness of the similarity indices, a simulation test was then designed and performed with the aim of estimating missing monthly rainfall rates in a suitably selected gauging station. The results of the simulation allowed us to evaluate the effectiveness of the proposed similarity indices. Finally, the multiple imputation by chained equations method was used as a benchmark to have an absolute yardstick for comparing the outcomes of the test. In conclusion, the new proposed multiplicative similarity index resulted at least as reliable as the selected benchmark.

Groundwater Exploitation as Thermal Fluid in Very-Low Enthalpy Geothermal Plants in Coastal Aquifers

Recently, in Italy, the interest for very low enthalpy geothermal resources (T < 20 °C) is growing. This is mainly because, these resources are widely available throughout the country and also unlike the other green energy sources (eg. solar and wind energy), and they do not need to be stored. Among the direct-use of geothermal resources, the open-loop groundwater heat pump (GWHP) system needs particular attention in terms of potential environmental impact. In coastal areas, that are generally densely populated, the installation of GWHP system is particularly appealing because the presence of shallow aquifers. This means significant savings of economic resources in terms of pumping energy and drilling costs. Nevertheless, vast areas of the Italian coastlines, as well as those of other Mediterranean countries, are often affected by seawater intrusion and hence are ruled by restrictive laws aimed to protect the groundwater quality and quantity.

Assessment of Groundwater Balance Terms Based on the Cross-Calibration of Two Different Independent Approaches

A reliable groundwater balance assessment is a fundamental tool for any effective resource exploitation plan. Nevertheless, some terms of the balance equation are, generally, very difficult to be estimated, even on average, especially when large and heterogeneous groundwater bodies are considered.