Hydraulic Efﬁciency of Green-Blue Flood Control Scenarios for Vegetated Rivers: 1D and 2D Unsteady Simulations

: Flood hazard mitigation in urban areas crossed by vegetated ﬂows can be achieved through two distinct approaches, based on structural and eco-friendly solutions, referred to as grey and green–blue engineering scenarios, respectively; this one is often based on best management practices (BMP) and low-impact developments (LID). In this study, the hydraulic efﬁciency of two green–blue scenarios in reducing ﬂood hazards of an urban area crossed by a vegetated river located in Central Tuscany (Italy), named Morra Creek, were evaluated for a return period of 200 years, by analyzing the ﬂooding outcomes of 1D and 2D unsteady hydraulic simulations. In the ﬁrst scenario, the impact of a diffuse effect of ﬂood peak reduction along Morra Creek was assessed by considering an overall real-scale growth of common reed beds. In the second scenario, riverine vegetation along Morra Creek was preserved, while ﬂood hazard was mitigated using a single vegetated ﬂood control area. This study demonstrates well the beneﬁts of employing green–blue solutions for reducing ﬂood hazards in vegetated rivers intersecting agro-forestry and urban areas while preserving their riverine ecosystems. It emerged that the ﬁrst scenario is a valuable alternative to the more impacting second scenario, given the presence of ﬂood control areas.


Introduction
The geo-hydrological and ecohydraulic challenges linked to current and future climate processes [1][2][3][4][5][6] highlight the growing need to protect water resources quantitatively and qualitatively in an ever more decisive way, especially in sensitive areas within both natural and urban territories [7,8]. In this context, the hydraulic conveyance of vegetated open channels intersecting anthropogenic settlements is dramatically affected by the temporal evolution of riverine vegetation properties [9][10][11], mainly associated with riverine plants' growth, foliage, and density overall [12,13]. In fact, in the case of vegetated flows, the morphometric and bio-physical changes over time in riverine vegetation canopy features represent a source of hydraulic roughness, in addition to that due to the only riverbanks and bed, to be meticulously considered in the field-scale analysis of global flow resistance [14,15].
Managing riverine vegetation biomass growth plays a key role in mitigating the flooding risk associated with urban and agro-forestry areas crossed by vegetated water bodies [16][17][18][19]. In the past, flood peak control has been essentially achieved by adopting hydraulic engineering solutions based on so-called "grey engineering scenarios", aiming at reducing peak hydrological discharge and water levels [20,21] through traditional engineering infrastructures that prevent any development of terrestrial or aquatic ecosystems through the years [22,23]. Thus, grey scenarios do not deliver multiple environmental benefits, also known as "Ecosystem Services", apart from flood control or peak discharge and water level reduction effects. On the other hand, based on the proposal of low-impact developments (LID) and best management practices (BMP), which aim at balancing the need for improved hydraulic efficiency and the need to mitigate the environmental impact of the hydraulic infrastructures, green-blue scenarios constitute a very promising scientific and practical advance in terms of flood control engineering solutions at low ecological impacts [24][25][26][27][28][29].
To properly model and simulate the actual biomechanical and botanical traits of riverine vegetation stands at field scale, the analysis of the green volumes involved in their phenological and morphometric evolution over time is essential [30][31][32]. As suggested by previous ecohydraulic studies and reviews [33][34][35], riverine vegetation's canopy morphometric trends can be easily described by the well-known leaf area index (LAI). This parameter must be properly considered in predictive and numerical modeling of vegetated rivers in urban areas for taking rigorously into account the real-scale impacts of riverine vegetation evolution on flow resistance in vegetated streams, to be then robustly validated by vegetational and water flow measurements acquired during experimental ecohydraulic field campaigns [36,37].
There still exists a need for the proposal of an accurate analytical methodology for evaluating the role of riverine vegetation canopy growth on the effectiveness of green-blue flood risk mitigation systems. To respond to this research question, dedicated hydraulic simulations were carried out in this article to demonstrate the peak flood lamination associated with both riverine vegetation (1D) and natural vegetated flood control areas (2D), as practical examples of green-blue flood control solutions. In detail, the aim of this study is the evaluation of the hydraulic performance (expressed in percentage, %) of two green-blue flood control scenarios associated with peak water level and discharge values for a return period (hereinafter referred to as T) of 200 years in an Italian vegetated river named Morra Creek, colonized by riverine Phragmites australis (Cav.) Trin. Ex. Steudel., mostly known as common reed beds. In the first scenario, hereinafter indicated as greenblue Scenario I, the effect of riverine canopy growth on the hydraulic conveyance of the examined vegetated river was simulated by varying the values of Manning's hydraulic roughness coefficients (hereinafter indicated as n) from 0.05 m −1/3 s (very young plants) to 0.40 m −1/3 s (mature plants) continuously. In the second case, hereinafter referred to as green-blue scenario II, the impacts of the vegetated flood control areas were modeled and simulated, while assuming the cover of the examined riverine vegetation stands along Morra Creek unchanged.
The novel approach proposed in this study is embodied by the analysis of the hydraulic efficiency associated with green-blue flood control proposals, based on the hydrodynamic interaction between the specific riverine vegetation and water flow within a real vegetated watercourse, representing an advance in flood risk mitigation planning applied mainly to vegetated flows crossing agro-forestry and urban areas.

Study Site
Morra Basin mainly develops in the two Municipalities of Collesalvetti and Torretta Vecchia (Province of Livorno-Central Tuscany, Northern Italy), from 457 to 18 m a.m.s.l. The land use is mainly constituted by agricultural land, croplands in detail, with about 50% of its surface used for this purpose, while forest and semi-natural environments occupy 43%. The manmade territories cover only 7% but have more than tripled in the last 70 years, affecting mostly floodplains located in the lower part of the Morra Basin watershed. Thus, it is a matter of fact that all this uncontrolled urbanization process contributed importantly to modify the natural hydraulic and ecological conditions of the vegetated stream examined in the present study, which has been progressively forced into increasingly smaller or even channeled both bed and riverbanks through the years. Figure 1 shows in detail the Morra Creek's length and location.

Study Site
Morra Basin mainly develops in the two Municipalities of Collesalvetti and Torretta Vecchia (Province of Livorno-Central Tuscany, Northern Italy), from 457 to 18 m a.m.s.l. The land use is mainly constituted by agricultural land, croplands in detail, with about 50% of its surface used for this purpose, while forest and semi-natural environments occupy 43%. The manmade territories cover only 7% but have more than tripled in the last 70 years, affecting mostly floodplains located in the lower part of the Morra Basin watershed. Thus, it is a matter of fact that all this uncontrolled urbanization process contributed importantly to modify the natural hydraulic and ecological conditions of the vegetated stream examined in the present study, which has been progressively forced into increasingly smaller or even channeled both bed and riverbanks through the years. Figure 1 shows in detail the Morra Creek's length and location. Morra Creek (43°33′41″ N-10°28′47″ E, at the vegetated stream's mouth) is a vegetated river belonging to the hydrographic network of Morra Basin. This vegetated watercourse extends for approximately 7 km and the dominant riverine vegetation stands species identified along its length is Phragmites australis (Cav.) Trin. Ex. Steudel., commonly known as common reed beds.
In Figure 2, two hydraulic cross-sections of Morra Creek vegetated river are depicted, characterized by the existence of two bridges hereinafter referred to as "Bridge A" (reinforced concrete road bridge) and "Bridge B" (masonry road and pedestrian bridge), respectively. Morra Creek (43 • 33 41" N-10 • 28 47" E, at the vegetated stream's mouth) is a vegetated river belonging to the hydrographic network of Morra Basin. This vegetated watercourse extends for approximately 7 km and the dominant riverine vegetation stands species identified along its length is Phragmites australis (Cav.) Trin. Ex. Steudel., commonly known as common reed beds.
In Figure 2, two hydraulic cross-sections of Morra Creek vegetated river are depicted, characterized by the existence of two bridges hereinafter referred to as "Bridge A" (reinforced concrete road bridge) and "Bridge B" (masonry road and pedestrian bridge), respectively. In the last 30 years of hydraulic and hydrological observations, the largest flood was recorded in September 2017 when 250 mm of rain fell in three hours within the Morra Basin watershed. The hydraulic risk is due to the restriction at "Bridge B" cross-section of Morra Creek vegetated river.  In the last 30 years of hydraulic and hydrological observations, the largest flood was recorded in September 2017 when 250 mm of rain fell in three hours within the Morra Basin watershed. The hydraulic risk is due to the restriction at "Bridge B" cross-section of Morra Creek vegetated river. Figure 3a shows the Morra Creek hydraulic conditions during September 2017 flood, whilst Figure 3b,c report a view of "Bridge A" (43 • 33 23" N-10 • 28 41" E) and "Bridge B" (43 • 33 26" N-10 • 28 44" E) structures and cross-sections, respectively.

1D and 2D Hydraulic Simulations
In the present study, the hydraulic efficiency in reducing both peak discharge and water level at "Bridge B" cross-section of Morra Creek was analyzed and discussed, based on the proposal of two different green-blue flood control scenarios, here respectively referred to as green-blue scenario I and green-blue scenario II. In the first scenario (greenblue scenario I), a diffuse lamination effect was modeled and simulated within Morra Creek, as an innovative proposal of environment-friendly management practice of the riverine common reed beds. In the second case (green-blue scenario II), riverine vegetation was modeled to be at its current phenological development stage and flood mitigation was simulated by considering two vegetated flood control areas first and then just a deeper single vegetated flood control area.
All the hydraulic simulations were conducted with HEC-RAS v5.0 freeware software under unsteady flow conditions, aiming at evaluating the flooding lamination effects along the entire vegetated river over time. In detail, for green-blue scenario I, only 1D geometric and numerical schemes were adopted. In contrast, in the case of green-blue scenario II, a 1D scheme was employed to simulate flow along Morra Creek, and 2D geometric and numerical schemes were considered for simulating the vegetated flood control areas (e.g., customized meshes). The effects of the two Green-Blue flood mitigation scenarios proposed in the present study were quantitatively evaluated for an input flow based on ahydrological hydrograph characterized by a return period T of 200 years.

1D and 2D Hydraulic Simulations
In the present study, the hydraulic efficiency in reducing both peak discharge and water level at "Bridge B" cross-section of Morra Creek was analyzed and discussed, based on the proposal of two different green-blue flood control scenarios, here respectively referred to as green-blue scenario I and green-blue scenario II. In the first scenario (greenblue scenario I), a diffuse lamination effect was modeled and simulated within Morra Creek, as an innovative proposal of environment-friendly management practice of the riverine common reed beds. In the second case (green-blue scenario II), riverine vegetation was modeled to be at its current phenological development stage and flood mitigation was simulated by considering two vegetated flood control areas first and then just a deeper single vegetated flood control area.
All the hydraulic simulations were conducted with HEC-RAS v5.0 freeware software under unsteady flow conditions, aiming at evaluating the flooding lamination effects along the entire vegetated river over time. In detail, for green-blue scenario I, only 1D geometric and numerical schemes were adopted. In contrast, in the case of green-blue scenario II, a 1D scheme was employed to simulate flow along Morra Creek, and 2D geometric and numerical schemes were considered for simulating the vegetated flood control areas (e.g., customized meshes). The effects of the two Green-Blue flood mitigation scenarios proposed in the present study were quantitatively evaluated for an input flow based on ahydrological hydrograph characterized by a return period T of 200 years.
The dimensional features of "Bridge A" and "Bridge B" structures and cross-sections were reproduced in the HEC-RAS v5.0 geometric model. In particular, the spans of the two bridges were modeled to evaluate the effective flood peak reduction efficiencies for the two green-blue flood control scenarios herein proposed, in terms of peak water level and discharge values.
The "Bridge A" and "Bridge B" actual geometries are summarized in Tables 1 and 2. The dimensional features of "Bridge A" and "Bridge B" structures and cross-sections were reproduced in the HEC-RAS v5.0 geometric model. In particular, the spans of the two bridges were modeled to evaluate the effective flood peak reduction efficiencies for the two green-blue flood control scenarios herein proposed, in terms of peak water level and discharge values.
The "Bridge A" and "Bridge B" actual geometries are summarized in Tables 1 and 2.  Figure 4a displays the HEC-RAS v5.0 geometry model of Morra Creek vegetated river, adopted here for running the initial 1D hydraulic simulation. In the present numerical study, "Bridge A" and "Bridge B" cross-sections are indicated as "3019.5" (white unfilled circle) and "3015.3" (blue unfilled circle), respectively. As shown in Figure 4b,c, "Bridge A" and "Bridge B" geometries and cross-sections were modeled and then simulated here to assess the hydraulic efficiency of the two examined green-blue scenarios in reducing water level and discharge flood peaks under unsteady flow conditions. Table 2. "Bridge B" geometry and dimensions: height from Morra Creek bed (m), span width (m), and length (m).

Span Features
Dimensions (m) Height 4.50 Width 8.07 Length 5.23 Figure 4a displays the HEC-RAS v5.0 geometry model of Morra Creek vegetated river, adopted here for running the initial 1D hydraulic simulation. In the present numerical study, "Bridge A" and "Bridge B" cross-sections are indicated as "3019.5" (white unfilled circle) and "3015.3" (blue unfilled circle), respectively. As shown in Figure 4b,c, "Bridge A" and "Bridge B" geometries and cross-sections were modeled and then simulated here to assess the hydraulic efficiency of the two examined green-blue scenarios in reducing water level and discharge flood peaks under unsteady flow conditions.    To properly model and analyze the actual ecohydraulic conditions observed within Morra Creek, upstream of "Bridge A" cross-section, values of Manning's n hydraulic roughness coefficients equal to 0.05 m −1/3 s (rocks and very low common reed plants) and 0.06 m −1/3 s (very low grassy shrubs) were employed here to simulate hydraulic roughness at both riverbed and floodplain stages, respectively. Instead, downstream of "Bridge A" cross-section, a value of Manning's n equal to 0.033 m −1/3 s was adopted for Morra Creek riverbed, where the whole bottom was cleaner and straighter and to 0.04 m −1/3 s at the floodplains, where it was possible mostly observing short vegetation related to agricultural cultivation.
As reported in the next section, the main morphometric and phenological features of the riverine vegetation species identified along Morra Creek were rigorously analyzed to define the most suitable values of Manning's n hydraulic roughness coefficients to be employed in the hydraulic simulations performed in the case of the green-blue Scenario I for properly representing the evolutive trends of the riverine vegetation canopy analyzed in the present study case.  To properly model and analyze the actual ecohydraulic conditions observed within Morra Creek, upstream of "Bridge A" cross-section, values of Manning's n hydraulic roughness coefficients equal to 0.05 m −1/3 s (rocks and very low common reed plants) and 0.06 m −1/3 s (very low grassy shrubs) were employed here to simulate hydraulic roughness at both riverbed and floodplain stages, respectively. Instead, downstream of "Bridge A" cross-section, a value of Manning's n equal to 0.033 m −1/3 s was adopted for Morra Creek riverbed, where the whole bottom was cleaner and straighter and to 0.04 m −1/3 s at the floodplains, where it was possible mostly observing short vegetation related to agricultural cultivation.
As reported in the next section, the main morphometric and phenological features of the riverine vegetation species identified along Morra Creek were rigorously analyzed to define the most suitable values of Manning's n hydraulic roughness coefficients to be employed in the hydraulic simulations performed in the case of the green-blue Scenario I for properly representing the evolutive trends of the riverine vegetation canopy analyzed in the present study case.

Riverine Vegetation Growth
The only riverine vegetation species recognized across the vegetated stream examined in the present study is a weed species named Phragmites australis (Cav.) Trin. Ex. Steudel., most known as common reed beds.
As depicted in Figure 6, increasing values of Manning's n coefficients were considered for the green-blue scenario I to simulate the growth in biomass area of the examined common reed beds (indicated by the green-yellow rectangles in Figure 6) along Morra Creek, corresponding to an augmentation in leaf area index (LAI), precisely defined as the ratio between the total leaf area distributed on the riverine plants' height (in m 2 ) and the projected ground area (in m 2 ) in the field [12,31,38,39].

Green-blue Flood Control Scenario I Riverine Vegetation Growth
The only riverine vegetation species recognized across the vegetated stream examined in the present study is a weed species named Phragmites australis (Cav.) Trin. Ex. Steudel., most known as common reed beds.
As depicted in Figure 6, increasing values of Manning's n coefficients were considered for the green-blue scenario I to simulate the growth in biomass area of the examined common reed beds (indicated by the green-yellow rectangles in Figure 6) along Morra Creek, corresponding to an augmentation in leaf area index (LAI), precisely defined as the ratio between the total leaf area distributed on the riverine plants' height (in m 2 ) and the projected ground area (in m 2 ) in the field [12,31,38,39].
As suggested by many previous analytical and modeling studies on the real-scale ecohydrodynamic response of common reed stands [11,40,41], the vertical distribution of plants' green leaf volumes is very similar on the whole reed height, with negligible LAI values in the first 0.10-0.15 m from the ground [38,41]. Thus, LAI is rigorously the same as plant area index (PAI) in the case of common reed beds, with PAI quantitatively obtained by dividing the total stems and leaves areas (in m 2 ) within the examined riverine plants' height (in m 2 ) and the projected ground area (in m 2 ), both measured in dedicated ecohydraulic field campaigns [42,43].
In the present study, to properly consider the actual phenological and morphometric processes associated with riverine vegetation canopy traits in the hydraulic simulations, overall Morra Creek's hydraulic roughness was modeled in the case of green-blue scenario I by considering increasing Manning's n values ranging from 0.05 m −1/3 s (no vegetation cover) to 0.40 m −1/3 s (massive common reed cover), to assess the flood peak control efficiencies associated with a diffuse augmentation in hydraulic roughness due to the only common reed beds' growth. Thus, for green-blue scenario I, the hydraulic simulations aimed at identifying the minimum value of Manning's n hydraulic roughness coefficient corresponding to a surface water elevation (SWE, m a.m.s.l.) value compatible with span height at "Bridge B" cross-section, equal to 4.50 m (See Table 2). Water 2021, 13, x FOR PEER REVIEW 8 of 20 As suggested by many previous analytical and modeling studies on the real-scale ecohydrodynamic response of common reed stands [11,40,41], the vertical distribution of plants' green leaf volumes is very similar on the whole reed height, with negligible LAI values in the first 0.10-0.15 m from the ground [38,41]. Thus, LAI is rigorously the same as plant area index (PAI) in the case of common reed beds, with PAI quantitatively obtained by dividing the total stems and leaves areas (in m 2 ) within the examined riverine plants' height (in m 2 ) and the projected ground area (in m 2 ), both measured in dedicated ecohydraulic field campaigns [42,43].
In the present study, to properly consider the actual phenological and morphometric processes associated with riverine vegetation canopy traits in the hydraulic simulations, overall Morra Creek's hydraulic roughness was modeled in the case of green-blue scenario I by considering increasing Manning's n values ranging from 0.05 m −1/3 s (no vegetation cover) to 0.40 m −1/3 s (massive common reed cover), to assess the flood peak control efficiencies associated with a diffuse augmentation in hydraulic roughness due to the only common reed beds' growth. Thus, for green-blue scenario I, the hydraulic simulations aimed at identifying the minimum value of Manning's n hydraulic roughness coefficient corresponding to a surface water elevation (SWE, m a.m.s.l.) value compatible with span height at "Bridge B" cross-section, equal to 4.50 m (See Table 2).
A total of twenty 1D unsteady hydraulic simulations were carried out for green-blue scenario I, by applying an increase of 0.02 m −1/3 s to Manning's n hydraulic roughness coefficient at each HEC-RAS v5.0 run. A total of twenty 1D unsteady hydraulic simulations were carried out for green-blue scenario I, by applying an increase of 0.02 m −1/3 s to Manning's n hydraulic roughness coefficient at each HEC-RAS v5.0 run.

Green-Blue Flood Control Scenario II
The HEC-RAS v5.0 geometry model adopted in the 2D hydraulic simulation is reported in Figure 7a, with FC A1 and FC A2 representing here two vegetated flood control areas located at Morra Creek left and right orographic banks, respectively (indicated in Figure 7a by the green polygons). In detail, FC A1 and FC A2 are two agricultural areas, to be employed here as vegetated flood control areas, representing a proposal of engineering solution based on a perspective of low environmental and ecological impacts. The 2D unsteady hydraulic simulation was first carried out by connecting the 2D geometry networks of FC A1 and FC A2 vegetated flood control areas to the 1D geometry model through side green structures to allow the bidirectional exchange of water volumes between the 1D and the 2D geometry models. 0.05 m −1/3 s was assigned to the whole vegetated water body to simulate the total riverine vegetation removal along the vegetated river examined in the present study. As indicated in Figure 7b, a further 2D hydraulic simulation was carried out by replacing the two vegetated flood control areas FCA1 and FCA2 areas with a single area, indicated here as FC*A1, narrower and 2 m deeper than the original FCA1, to evaluate its hydraulic efficiency in achieving the desired peak reduction effect for both discharge and water level values at "Bridge B" cross-section of Morra Creek.  To exclude the riverine vegetation growth from the 2D hydraulic simulations associated with Morra Creek in the green-blue scenario II, a value of Manning's n coefficient of 0.05 m −1/3 s was assigned to the whole vegetated water body to simulate the total riverine vegetation removal along the vegetated river examined in the present study. As indicated in Figure 7b, a further 2D hydraulic simulation was carried out by replacing the two vegetated flood control areas FC A1 and FC A2 areas with a single area, indicated here as FC* A1 , narrower and 2 m deeper than the original FC A1 , to evaluate its hydraulic efficiency in achieving the desired peak reduction effect for both discharge and water level values at "Bridge B" cross-section of Morra Creek.
As shown in Figure 7b, the proposal to realize a natural wet area and the planting of trees and shrubs buffer strips or riparian hygrophilous plants would result in eco-friendly and sustainable protection of water resources quality and, therefore, in the contemporary improvement of the ecological services associated with the presence of the single vegetated flood control area FC* A1 .
The following Table 3 summarizes the main dimensional features of FC* A1 . Width and depth are indicated here as average values. In Table 4 are reported the most relevant numerical features of the 1D and 2D unsteady hydraulic simulations carried out in the present study for the two green-blue flood control scenarios.

Flood Control Efficiency: Peak Discharge and Water Level Reduction
As remarked by Del Giudice et al. [44] among others, the hydraulic efficiencies (expressed here in percentage, %) of the flood control effects corresponding to both green-blue scenario I and green-blue scenario II in terms of peak discharge Q (m 3 s −1 ) and water level h (m) reduction associated with a return period T of 200 years, were quantitively computed as follows: η (%) = (peak in − peak out )/peak in , where peak in and peak out indicate, respectively, the peak discharge Q (m 3 s −1 ) and water level h (m) peak values in input and output at "Bridge B" cross-section of Morra Creek for a return period T of 200 years. The corresponding peak reduction efficiencies are here indicated as η Q (%) and η h (%), respectively.

Morra Creek Current Ecohydraulic Conditions
The flooding outcomes resulting from the 1D unsteady hydraulic simulation of the actual Morra Creek ecohydraulic conditions are displayed in Figure 8a,b, representing, respectively, surface water elevation (SWE) and flow average velocities U (m s −1 ) at each Morra Creek cross-section identified in the HEC-RAS v5.0 geometry model, for a hydrological hydrograph based on a return period T of 200 years, indicated in the present study as Q 200 (m 3 s −1 ).
As it emerges from the analysis of Figure 8a, the water level value at "Bridge B" crosssection of Morra Creek resulting from the 1D unsteady hydraulic simulation under the actual riverine vegetation conditions is equal to approximately 6.50 m, which is higher than "Bridge B" span height, equal to 4.50 m, as shown in Table 2. Thus, both green-blue scenario I and green-blue scenario II were modeled and then simulated in the following sections aiming at reducing the hydrological peaks for T of 200 years at "Bridge B" cross-section of Morra Creek. r 2021, 13, x FOR PEER REVIEW 11 of 20 As it emerges from the analysis of Figure 8a, the water level value at "Bridge B" crosssection of Morra Creek resulting from the 1D unsteady hydraulic simulation under the actual riverine vegetation conditions is equal to approximately 6.50 m, which is higher than "Bridge B" span height, equal to 4.50 m, as shown in Table 2. Thus, both green-blue scenario I and green-blue scenario II were modeled and then simulated in the following sections aiming at reducing the hydrological peaks for T of 200 years at "Bridge B" crosssection of Morra Creek.

green-blue Flood Control Scenario I
In Figure 9a,b are, respectively, displayed the values in percentage of peak discharge Q (m 3 s −1 ) and water level h (m) reduction efficiencies associated with T = 200 years for green-blue Scenario I at "Bridge B" cross-section, as a function of the increasing Manning's n coefficients values reproducing the real-scale augmentation in LAI values characterizing the riverine vegetation stands examined in the present study. In detail, both Figure 9a,b were obtained through twenty 1D unsteady hydraulic simulations (see Table  4).

Green-Blue Flood Control Scenario I
In Figure 9a,b are, respectively, displayed the values in percentage of peak discharge Q (m 3 s −1 ) and water level h (m) reduction efficiencies associated with T = 200 years for green-blue Scenario I at "Bridge B" cross-section, as a function of the increasing Manning's n coefficients values reproducing the real-scale augmentation in LAI values characterizing the riverine vegetation stands examined in the present study. In detail, both Figure 9a,b were obtained through twenty 1D unsteady hydraulic simulations (see Table 4).
As expected, it is possible to observe, in Figure 9a,b, that for green-blue scenario I, the peak reduction efficiencies in terms of both peak discharge and water level values at "Bridge B" cross-sections increase with common reed biomass growth over Morra Creek vegetated river, according to two very similar trends. In detail, η Q I values vary from 28.35% to 29.70%, while η hI ones vary from 25.09% to 31.46%. These values indicate that η hI is slightly more sensitive than η Q I to the real-scale changes in Manning's n hydraulic roughness coefficients, corresponding to the phenological evolutive trend of riverine common reed beds under their natural conditions across the whole length of the vegetated watercourse. As expected, it is possible to observe, in Figure 9a,b, that for green-blue scenario I, the peak reduction efficiencies in terms of both peak discharge and water level values at "Bridge B" cross-sections increase with common reed biomass growth over Morra Creek vegetated river, according to two very similar trends. In detail, η values vary from 28.35% to 29.70%, while η ones vary from 25.09% to 31.46%. These values indicate that η is slightly more sensitive than η to the real-scale changes in Manning's n hydraulic  Figure 10a,b illustrate, respectively, the outcomes of the 2D unsteady simulation in terms of SWE (m a.m.s.l.) for the whole Morra Creek length (Figure 10a), originally considered in green-blue scenario II and a detailed view of the two vegetated flood control areas FC A1 and FC A2 (Figure 10b). watercourse. Figure 10a,b illustrate, respectively, the outcomes of the 2D unsteady simulation in terms of SWE (m a.m.s.l.) for the whole Morra Creek length (Figure 10a), originally considered in green-blue scenario II and a detailed view of the two vegetated flood control areas FCA1 and FCA2 (Figure 10b). The results of the 2D hydraulic simulation reported in Figure 10a,b show that the combined action of the two vegetated flood control areas FCA1 and FCA2 was not sufficient for reducing peak SWE (m a.m.s.l.) at the examined "Bridge B" cross-section, equal to 6.0 m, indicating the need for replacing FCA1 and FCA2 areas with the only vegetated flood control area FC*A1 in the further 2D hydraulic model and simulation.

Green-Blue Flood Control Scenario II
In Figure 11a,b, the HEC-RAS v5.0 geometry model and the flooding outcomes are shown, resulting from the 2D hydraulic simulation implemented here for green-blue scenario II in terms of SWE (m a.m.s.l.), by considering the only single vegetated flood control area FC*A1. The results of the 2D hydraulic simulation reported in Figure 10a,b show that the combined action of the two vegetated flood control areas FC A1 and FC A2 was not sufficient for reducing peak SWE (m a.m.s.l.) at the examined "Bridge B" cross-section, equal to 6.0 m, indicating the need for replacing FC A1 and FC A2 areas with the only vegetated flood control area FC* A1 in the further 2D hydraulic model and simulation.
In Figure 11a,b, the HEC-RAS v5.0 geometry model and the flooding outcomes are shown, resulting from the 2D hydraulic simulation implemented here for green-blue scenario II in terms of SWE (m a.m.s.l.), by considering the only single vegetated flood control area FC* A1 .
It emerges from the analyses of Figure 11a,b that for green-blue scenario II the surface water elevations SWE (m a.m.s.l.) is equal to 4.04 m, which is quantitatively compatible with "Bridge B" span height, equal to 4.50 m. These trends can also be easily observed by analyzing the effects on the overall flood risk assessment of the flooding outcomes computed along the Morra Creek cross-sections through the 2D unsteady hydraulic simulation, displayed in Figure 12a,b. In detail, in Figure 12a,b the values of SWE (m a.m.s.l.) and flow average velocities U (m s −1 ) at each modeled and then simulated Morra Creek cross-sections, respectively.
In Table 5 are indicated the discharge Q II (m 3 s −1 ) and water level h II (m) values of the output hydrograph of the 2D unsteady hydraulic simulation at "Bridge B" cross-section of Morra Creek for green-blue scenario II because of the existence of the only vegetated flood control area FC* A1 . It is possible to observe from Table 5 that, given the presence of the single vegetated flood control area FC* A1 , the value of peak discharge Q II (m 3 s −1 ) at "Bridge B" crosssection moved from 155.15 to 109.53 m 3 s −1 , while the value of peak water level h II (m) effectively decreased from 6.50 to 4.04 m. It emerges from the analyses of Figure 11a,b that for green-blue scenario II the surface water elevations SWE (m a.m.s.l.) is equal to 4.04 m, which is quantitatively compatible with "Bridge B" span height, equal to 4.50 m. These trends can also be easily observed by analyzing the effects on the overall flood risk assessment of the flooding outcomes computed along the Morra Creek cross-sections through the 2D unsteady hydraulic sim- In Table 5 are indicated the discharge QII (m 3 s −1 ) and water level hII (m) values of the output hydrograph of the 2D unsteady hydraulic simulation at "Bridge B" cross-section of Morra Creek for green-blue scenario II because of the existence of the only vegetated flood control area FC*A1.

Discussion
Based on the two green-blue flood control scenarios herein proposed, the results of the present study in terms of peak discharge and water level control effects for Morra Creek well demonstrated that both of them are capable of obtaining a satisfactory flood risk mitigation, with low ecological and environmental impacts compared to grey-or structural-hydraulic engineering solutions.
Values of Manning's n coefficients, ranging between 0.10 m −1/3 s and 0.40 m −1/3 s, are highly comparable with those obtained by the two previous ecohydraulic studies performed by Liu and Shan [45] and Lama et al. [46], which analyzed the average and turbulent fluctuations of flow velocity fields experimentally measured in vegetated streams to evaluate the main field-scale hydraulic and hydrodynamic effects of the riverine vegetation stands' phenology on the hydraulic conveyance of real vegetated water bodies. Both studies were carried out at increasing discharge and water levels. In the study of Liu and Shan [45], the values of Manning's n coefficients ranged from 0.12 m −1/3 s to 0.36 m −1/3 s, while they varied between 0.27 m −1/3 s and 0.49 m −1/3 s in the study carried out by Lama et al. [46].
In addition, it is important to emphasize here that Manning's n coefficient values resulting from the 1D hydraulic simulations (0.10-0.40 m −1/3 s) well represent the actual vegetative flow resistance induced by patchy riverine vegetation, as reported by the previous ecohydraulic studies carried out by West et al. [47] and Zhu et al. [48]. Their works, respectively, modeled the impacts of real-scale riverine vegetation distribution on water level argumentation and hydrodynamic dispersion within vegetated open channels under different growth conditions of riverine plants in both flume laboratory and field-scale study cases.
It is crucial to highlight that the outcomes of the present study refer to a single riparian vegetation species (common reed beds), while, in a more realistic perspective, the combined effect of two or more species represent a source of uncertainty which inevitably propagates in the vegetative Manning's n hydraulic roughness coefficient associated with each simulated cross-section, quantitatively varying in a wider range of values. Thus, this uncertainty could be reduced by implementing more complex and rigorous numerical analyses, obtained by coupling high-performance computational fluid dynamics (CFD) models [49][50][51][52][53] and simulations [54][55][56][57][58][59][60][61] with the most advanced remote sensing techniques of fluvial hydraulics, agro-forestry and land use governance interest [62][63][64][65][66][67][68][69][70][71].

Conclusions
From the analysis and the discussion of the outcomes of this study was possible to observe that green-blue scenario I, obtained by considering the diffuse peak lamination effect associated with common reed beds' growth along the whole Morra Creek vegetated watercourses, constitutes a valuable alternative to the fairly more impacting green-blue scenario II proposal.
In addition, it emerges that the proposals of green-blue scenarios in both urban and agricultural areas effectively constitute decisive steps in the definition of a new engineering and scientific methodology for flooding risk control associated with vegetated rivers with low ecological and environmental impacts [72][73][74][75][76], preserving their riverine and riparian ecosystems.
It is possible to assess that the main findings of the hydraulic simulations performed in the present study represent a useful tool for ecohydrological and ecohydraulic forecasters and modelers for flood risk control associated with both urban and agro-forestry territories. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.