Random Forest model and TRIX used in combination to assess and diagnose the trophic status of Bizerte Lagoon, southern Mediterranean

Béjaoui, Béchir; Armi, Zina; Ottaviani, Ennio; Barelli, Enrico; Gargouri-Ellouz, Emna; Chérif, Rim; Turki, Souâd; Solidoro, Cosimo; Aleya, Lotfi

doi:10.1016/j.ecolind.2016.07.010

Cited by 53 publications

(23 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A direct comparison (scatter plot) of the observed and predicted Chl-a concentrations is shown in Figure 3. The fitted RF model was much better than the one reported by Béjaoui et al (2016) for Bizerte lagoon (R 2 =0.51),…”

Section: Figcontrasting

confidence: 57%

See 1 more Smart Citation

Machine Learning modeling techniques for forecasting the trophic level in a restored South Mediterranean lagoon using Chlorophyll-α

Hadid

Goyet

Chaar

et al. 2021

Preprint

View full text Add to dashboard Cite

An Artificial Neural Network (ANN), a Machine Learning (ML) modeling approach is proposed to predict the ecological state of the North Lagoon of Tunis, a shallow restored Mediterranean coastal ecosystem. A Nonlinear Auto Regressive with exogenous input (NARX) neural network model was fitted to predict Chlorophyll- a (Chl- a ) concentrations in the North Lagoon of Tunis as an eutrophication indicator. The modeling is based on approximately three decades of monitoring water quality data (from January 1989 to April 2018) to train, validate and test the NARX model. The most relevant predictor variables used in this model were those having a high permutation importance ranking with Random Forest (RF) model, which simplified the structure of the network, resulting in a more accurate and efficient procedure. Those predictor variables are secchi depth, and dissolved oxygen. Various model scenarios with different NARX architectures were tested for Chl- a prediction. To verify the model performances, the trained models were applied to field monitoring data. Results indicated that the developed NARX model can predict Chl- a concentrations in the North Lagoon of Tunis with high accuracy (R= 0.79; MSE= 0.31). In addition, results showed that the NARX model generally performed better than multivariate linear regression (MVLR). This approach could provide a quick assessment of Chl- a variations for lagoon management and eco-restoration.

show abstract

Section: Figcontrasting

confidence: 57%

“…Algae produce oxygen during the day and absorb it during the night. Oxygen absorption also occurs during the process of algae death and decay (Béjaoui et al 2016). In agreement with that, our findings have shown that dissolved oxygen is also associated with Chl-a concentrations.…”

Section: Random Forestmentioning

confidence: 99%

Machine Learning modeling techniques for forecasting the trophic level in a restored South Mediterranean lagoon using Chlorophyll-α

Hadid

Goyet

Chaar

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…e TRIX has been applied to many ecosystems of coastal marine waters, such as the Tyrrhenian [33] and the Caspian Seas [34], Bizerte Lagoon in Tunisia [35], and the Greek coastal lagoon [36] are among others.…”

Section: Discussionmentioning

confidence: 99%

Assessment of the Trophic Status of the South Lagoon of Tunis (Tunisia, Mediterranean Sea): Geochemical and Statistical Approaches

Abidi

Amor

Gueddari

2018

Journal of Chemistry

View full text Add to dashboard Cite

The trophic status assessment of the South Lagoon of Tunis, a shallow Mediterranean coastal area after its restoration, is addressed herein with respect to its various environmental settings which are taken as indicators of water quality. The lagoon had, in the past, witnessed severe environmental quality issues. To resolve these problems, a large restoration project of the lagoon was undertaken which consisted of dredging the bottom sediments removing areas of water stagnation and improving water circulation. After this restoration work, the lagoon morphology has radically changed. In this paper, we attempt to evaluate the lagoon water’s trophic state to analyze the eutrophication risk after almost 16 years. In order to achieve these purposes, two water quality monitoring campaigns were conducted (July 2013 and February 2014). Natural and anthropogenic factors controlling the nutrient content of the lagoon water have been assessed through both geochemical methods and multivariate statistical tools. The results show that the nutrients are from external sources due to the discharge of municipal and industrial wastewater from the surrounding city of the catchment in the lagoon’s south side. According to the TRIX index, the lagoon remains eutrophic presenting a “poor” water quality, notwithstanding the engineering project due to the high level of nutrients.

show abstract

“…In the north of Tunisia, the main mussel breeding site is the lagoon of Bizerte (37° 11′ 48″ N, 9° 51′ 23″ E), where the local industry produces on average 115 tons/year ( DPGA, 2017 ). Seawater temperature varies widely from 9.8 to 28°C, although it can reach up to 30.8°C in summer ( Béjaoui et al , 2016 ). The maximum summer temperature in the northwestern Mediterranean Sea increased by ~1°C between 2002 and 2010, relative to the 1980–2000 average ( Marbá and Duarte, 2010 ), and rapid warming of 2.8 ± 1.1°C is expected by the end of the century ( Jordá et al , 2012 ).…”

Section: Introductionmentioning

confidence: 99%

The Mediterranean mussel Mytilus galloprovincialis: responses to climate change scenarios as a function of the original habitat

Lassoued

Pad́ın

Comeau

et al. 2021

Conservation Physiology

View full text Add to dashboard Cite

The impact of simulated seawater acidification and warming conditions on specimens of the mussel Mytilus galloprovincialis locally adapted to very distinct, widely separated sites in the Mediterranean Sea (Tunisia) and Atlantic Sea (Galicia, NW Spain) was evaluated in relation to key behavioural and eco-physiological parameters. Over the 2-month exposure to the experimental conditions, mussels were fed optimally to ensure that there are no synergistic interactions between climate change drivers and energetic status of the individuals. In general, regardless of origin (Atlantic or Mediterranean), the mussels were rather resilient to acidification for most of the parameters considered and they were able to grow in strongly acidified seawater through an increased feeding activity. However, shell strength decreased (40%) consistently in both mussel populations held in moderately and highly acidified seawater. The observed reduction in shell strength was not explained by slight alterations in organic matter, shell thickness or aragonite:calcite ratio. The combined effects of high acidification and warming on the key response of byssus strength caused a strong decline in mussel performance, although only in Galician mussels, in which the valve opening time decreased sharply as well as condition index (soft tissue state) and shell growth. By contrast, the observed negative effect of highly acidified scenario on the strength of Tunisian mussel shells was (partly but not totally) counterbalanced by the higher seawater temperature. Eco-physiological and behavioural interactions in mussels in relation to climate change are complex, and future scenarios for the ecology of the species and also the feasibility of cultivating them in Atlantic and Mediterranean zones are discussed.

show abstract

Random Forest model and TRIX used in combination to assess and diagnose the trophic status of Bizerte Lagoon, southern Mediterranean

Cited by 53 publications

References 48 publications

Machine Learning modeling techniques for forecasting the trophic level in a restored South Mediterranean lagoon using Chlorophyll-α

Machine Learning modeling techniques for forecasting the trophic level in a restored South Mediterranean lagoon using Chlorophyll-α

Assessment of the Trophic Status of the South Lagoon of Tunis (Tunisia, Mediterranean Sea): Geochemical and Statistical Approaches

The Mediterranean mussel Mytilus galloprovincialis: responses to climate change scenarios as a function of the original habitat

Contact Info

Product

Resources

About