Impact of hydrological connectivity on rainfed annual crops in Mediterranean hilly landscapes: a numerical exploration

doi:10.21203/rs.3.rs-3745046/v1

Download PDF

Research Article

Impact of hydrological connectivity on rainfed annual crops in Mediterranean hilly landscapes: a numerical exploration

https://doi.org/10.21203/rs.3.rs-3745046/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Within hilly agricultural landscapes, topography induces lateral transfers of runoff water, so-called interplot hydrological connectivity. Runoff water generated in upstream plots can infiltrate downstream plots, thus influencing the water content in the root zone that drives crop functioning. As the impact of runoff on crop functioning in the context of Mediterranean rainfed annual crops has been understudied thus far, the current study aims to comprehensively investigate this effect. To quantify how the downstream infiltration of upstream runoff affects crop functioning, we conduct a numerical experiment using the AquaCrop crop model and consider two hydrologically connected plots. The experiment explores a range of upstream and downstream agro-pedo-climatic conditions: crop type, soil texture and depth, climate forcing, and the area of the upstream plot. The experiment relies on data collected over the last 25 years within the Kamech catchment, an environment research observatory in northeastern Tunisia. Overall, the results show that the downstream infiltration of upstream runoff has a positive impact on crop functioning in a moderate number of situations, ranging from 16% (wheat) to 33% (faba bean) as the average across aboveground biomass and yield. This positive impact is mostly found for soils with higher available water capacity and under semiarid and dry subhumid climate conditions, with a significant impact of rainfall intra-annual distribution in relation to crop phenology. These results need to be deepened by considering both a wider range of crops and future climate conditions, for which larger occurrences of semiarid climate conditions are expected.

Hydrological connectivity

Runoff-runon process

Water infiltration

Rainfed agriculture

Annual crops

Crop production

Mediterranean

Within the Mediterranean basin, water resources are limited and unevenly distributed in time and space, partly due to contrasting rainfall patterns (Blinda and Thivet 2009; Daccache et al. 2016; Fader et al. 2020). This water context is set to worsen because of (1) water resource overexploitation to meet the growing food demand (Karabulut et al. 2018; Souissi et al. 2019) and (2) climate change consequences such as rainfall decreases, up to 30% in some regions (Lange et al. 2020), increases in evaporative demand (Fader et al. 2020), and concentrations of intra-annual rainfall distributions (Ramos and Martínez-Casasnovas 2006). As the largest water user, the agricultural sector has long been under threat, with subsequent challenges for food security (Yang and Zehnder 2002). Rainfed annual crops are significantly affected by water issues because (1) they fully depend upon rainfall as a water resource, and (2) their shallow root systems make them vulnerable to water shortages (Hossain et al. 2020).

Mediterranean policies for water resource management have mainly focused on mobilising these resources to support irrigated agriculture (Besbes et al. 2014; Nouri et al. 2020). As a result, less attention has been given to rainfed agriculture, which uses less water per unit area and has significant room for improvement (Anderson et al. 2016). Indeed, it is possible to improve rainfed agricultural productivity by implementing strategies that (1) reduce crop water needs, with species suited to drought conditions, or (2) increase water availability in the root zone by favouring runoff/rainwater infiltration or minimising evaporative losses. The amount of water that infiltrates the root zone can be increased using water harvesting techniques (e.g., planting pits, terraces) at different spatial scales from the plot to the landscape (Yadari et al. 2019; Tadros et al. 2021; Molénat et al. 2023). These techniques are suitable for landscapes with hilly topography that allows for the spatial redistribution of surface runoff (Ammar et al. 2016; Mekki et al. 2018). The benefits of these techniques in reducing runoff, promoting infiltration, increasing the soil water content, and enhancing crop yield have been demonstrated, particularly in Sub-Saharan Africa (see Wolka et al. (2018) for a review). Most of the studies conducted in the Mediterranean basin have demonstrated the benefits of the techniques in terms of reducing runoff or increasing the soil water content (e.g., Schiettecatte et al. (2005)). However, only a few studies have specifically focused on the impact on crop yield.

Hydrological connectivity can be defined as water transfer from one part of the landscape to another (Bracken and Croke 2007). In most Mediterranean hilly landscapes, the interplot hydrological connectivity predominantly arises from surface runoff, which induces the spatial redistribution of rainfall between plots. In this context, the runoff-runon process occurs when runoff generated within an upstream area has the potential to infiltrate downslope cultivated plots with higher infiltration capacity (Jones et al. 2013; Van Loo and Verstraeten 2021) (Fig. 1). This process contributes to augmented water availability in the root zone (Howes and Abrahams 2003). While the influence of interplot hydrological connectivity on hydrological processes, such as stream flow generation, is widely acknowledged (Nanda et al. 2019; Zuecco et al. 2019; Saco et al. 2020), there are limited studies assessing the impact of this connectivity due to the runoff-runon process on crop functioning. Typically, crop functioning and agricultural production are often studied in a spatially distributed manner using multilocal methods, assuming that plots are hydrologically independent of each other and thus neglecting the impact of hydrological connectivity (Van Gaelen et al. 2017). However, understanding this impact on crop functioning is crucial for optimising strategies in agricultural landscape management, where optimisation aims to modulate the crop consumption of rain water to enhance agricultural production (Saco et al. 2020). This consideration is particularly pertinent in arid to semiarid Mediterranean regions where water is considered the primary limiting factor for crop functioning and growth (Daccache et al. 2016; Araya et al. 2017) and where lateral water transfer is primarily attributed to surface runoff (Peredo et al. 2022).

Studying the impact of interplot hydrological connectivity on rainfed annual crops in hilly Mediterranean regions requires careful consideration of key environmental factors affecting both downstream crop functioning and upstream runoff generation (Fig. 1). Considering these factors enables the exploration of their potential influence on hydrological connectivity. Various elements can influence surface runoff, including rainfall patterns (Chen et al. 2016), hydrodynamic properties, soil moisture (Schoener and Stone 2019), agricultural practices (Prosdocimi et al. 2016), vegetation cover type (Nunes et al. 2011; Liu and Lobb 2021), and the impluvium area, representing the upstream contributing area for runoff input (Gnouma 2006). Furthermore, the functioning of downstream crops can be affected by the infiltration of upstream runoff and other environmental factors, such as climate (comprising rainfall and evapotranspiration demand), soil properties (such as the organic matter content and available water capacity), and agricultural practices (including fertilisation and soil management) (Mbava et al. 2020). Investigating the potential influences of these environmental factors on hydrological connectivity can be pursued through field campaigns or numerical experiments using modelling. The latter approach is more suitable, as it allows for (1) the consideration of a wide range of environmental factors and (2) the disentanglement of the combined effects of environmental factors with hydrological connectivity.

[Fig. 1 about here.]

The objective of this article is to study the impact of water infiltration due to the runoff-runon process on crop functioning. We focus on rainfed annual crops in a Mediterranean hilly landscape, emphasising two main agronomic variables: aboveground biomass and yield. In Section 2, we introduce the numerical experiment by describing the chosen modelling approach, the various influential factors to consider, and the strategy for analysing the results. Section 3 presents and discusses the analysis of modelling simulations: we first examine the occurrence of situations with a significant impact on hydrological connectivity and then assess the importance of environmental conditions (climate, upstream runoff, soil texture and depth) on this impact. Finally, we discuss the prospects of this study, considering that it represents a preliminary step toward an integrated catchment-scale approach.

2.1. General framework

The numerical experiment is based on simulating crop functioning in a downstream plot receiving surface runoff from an upstream plot that is hydrologically connected (Fig. 2). In addition to simulating crop functioning within the downstream plot, our approach relies on simulating the upstream runoff. Thus, the downstream plot is supplied with water from both rainfall and runoff simulated in the upstream plot. Both plots are typified by Mediterranean conditions in terms of crops, soils, and climate. The experiment relies on a few key assumptions. First, each of the upstream and downstream plots is assumed to be homogeneous in terms of parameters, state variables, and fluxes. Second, hydrological connectivity induces a complete transfer of upstream runoff toward the downstream plot. Here, we provide an overview of the numerical experiment setting, and a detailed presentation is provided in Supplementary materials - Section 1.

The numerical experiment is conducted using the AquaCrop model, which meets a set of requirements. This crop model (1) is a water-driven model in accordance with arid to semiarid Mediterranean regions where water is the principal limiting factor for crop functioning, (2) has been extensively validated across various Mediterranean conditions and for a range of state variables related to water budgets and crop growth, and (3) has the capability to simulate runoff, thus minimising costs in numerical simulations by preventing the use of a second simulation model.

[Fig. 2 about here.]

The simulations of downstream crop functioning, which account for additional water supply from upstream runoff, are generated to encompass a wide range of situations typical of Mediterranean conditions. This is achieved by considering a diversity of environmental drivers that influence crop functioning, such as (1) crop type, whose hydrological functioning and phenology are expected to vary significantly from crop to crop, (2) soil available water capacity, which is expected to vary significantly in relation to soil depth and texture via hydrodynamic properties (Cousin et al. 2022), and (3) climate forcing, including rainfall, temperature and reference evapotranspiration, which is expected to encompass variability over several decades.

In each situation of the downstream plot, we consider a range of upstream runoff magnitudes to account for the variability in Mediterranean hydrological conditions. This is achieved by simulating the upstream runoff based on a set of environmental drivers of runoff generation. Some of these drivers are the same as those identified for downstream crop functioning, including crop type, soil available water capacity, and climate forcing. In addition to these drivers, the drained area of the upstream plot is also considered for upstream runoff. This driver, hereafter referred to as the impluvium, is characterised using the ratio of upstream to downstream plot areas. A small value of the ratio would indicate a downstream plot near a hillslope summit in a landscape. Conversely, a large value would signify a downstream plot along the slope or at the bottom of a hillslope.

To generate simulations that are typical of Mediterranean conditions, we rely on the database provided by the environment research observatory OMERE (Observatoire Méditerranéen de l’Environnement Rural et de l’Eau, www.obs-omere.org, Molénat et al. 2018), as this database meets a set of requirements. First, it allows for the consideration of a range of climate forcings, as it includes data collected over the last three decades. Second, OMERE observations are collected within the Kamech catchment (Cap Bon Peninsula, northeastern Tunisia), which is representative of semiarid Mediterranean regions in terms of crops, soil, and climate. Third, the AquaCrop model was shown to acceptably simulate runoff and crop functioning under the agro-pedo-climatic conditions of the Kamech catchment (Dhouib et al. 2022).

2.2. Overview of the AquaCrop crop model

Detailed presentations of AquaCrop (https://www.fao.org/aquacrop/en/) are provided by Raes et al. (2009) and Steduto et al. (2009). Here, we outline the specificities related to our methodological choices.

AquaCrop is a crop model designed to simulate crop functioning and the principal components of the water balance. Specifically, tailored for arid and semiarid environments, it is categorised as a water-driven model (Todorovic et al. 2009). Indeed, it adjusts crop growth based on vegetation transpiration, itself driven by root zone soil moisture. This characteristic makes it well suited for Mediterranean regions, where water acts as the principal limiting factor for agricultural production.

AquaCrop simulates, on a daily time step, the components of soil water balance across the soil‒plant-atmosphere continuum (infiltration and runoff, deep percolation and capillary rise, soil evaporation and vegetation transpiration), as well as plant growth and production (canopy crop cover, root growth, aboveground biomass, yield). Crop transpiration (Tr) is derived from canopy crop cover (CC) and reference evapotranspiration (ET₀). Aboveground biomass (AGB) is then derived from Tr and normalised water productivity (WP*), which accounts for atmospheric CO₂ concentration. Yield (Yld) is calculated as the product of AGB and the harvest index (HI). Runoff is determined using the empirical curve number method that accounts for crop type, agricultural practice, and hydrological soil group in relation to the soil infiltration rate and antecedent soil moisture. The soil water balance is calculated by discretising the soil into five horizons based on pedological characteristics.

The AquaCrop forcing variables encompass climate data (e.g., air temperature, reference evapotranspiration ET₀, rainfall, and atmospheric CO₂ concentration) on a daily timescale. The model parameters consist of soil properties (texture and depth, soil moisture at field capacity, permanent wilting point and saturation, saturated hydraulic conductivity), cultural parameters (e.g., maximum canopy cover, crop coefficient), and agricultural practice data (e.g. fertilisation, sowing date).

2.3. Setting the agro-pedo-climatic conditions

We assumed the same variability for the agro-pedological conditions within the upstream and downstream plots. All possible scenarios for each of the two plots are next combined to ensure the representativeness of the resulting AquaCrop simulations. When dealing with climate conditions, including rainfall, air temperature and ET_0, for instance, we assumed uniformity across the two plots.

2.3.1. Crop types and subsequent crop parameters

We chose wheat as the cereal crop and faba bean as the legume crop. There are two reasons for this choice. First, these two crops are among the main rainfed crops within the Kamech catchment (Mekki et al. 2006) and in the Mediterranean region (FAO 2022). Second, wheat and faba bean are two very different crops, especially in terms of phenological cycle duration and agricultural practices (sowing and harvest dates), as well as in terms of hydrological functioning (different soil cover rates implying different infiltration-runoff ratios). Indeed, faba bean is a row crop with a short phenological cycle, whereas wheat is a cover crop with a longer phenological cycle. Supplementary materials - Sections 2 and 3 detail the setting, according to the study area, of the crop parameters, the choice of sowing dates, and the fertilisation rates.

2.3.2. Soil characteristics and hydrodynamic properties

In the current study, we consider soil to be homogeneous with depth. Soil hydrodynamic properties include (1) soil moisture at field capacity (FC), at permanent wilting point (PWP), and at saturation (Sat), as well as (2) the saturated hydraulic conductivity (Ksat). These properties are determined based on soil textures observed within the Kamech catchment. For this, we use data analysis from 10 soil pits collected at various locations within Kamech (Coulouma et al. 2017). This leads to the identification of three dominant textures in accordance with the USDA classification triangle, namely, clay (C), clay-loam (CL) and sandy-clay-loam (SCL). Then, soil textures are converted into hydrodynamic properties using the nominal values proposed by the AquaCrop user guide (Table ST3 in Supplementary materials - Section 2).

When dealing with soil depth, we set three different depths, namely, 0.5 m, 1 m, and 1.5 m. These values are chosen on the basis of the variability of soil depth observed in situ within the study site, where soil depths range from a few centimetres to two metres (Molénat et al. 2018).

By combining three soil textures and three soil depths, we simulate nine situations for soil available water capacity, which is defined as the maximum amount of plant available water that a soil can provide (Cousin et al. 2022).

2.3.3. Climate forcing

To account for the inter- and intra-annual variability in climate conditions, especially rainfall and reference evapotranspiration (ET₀), we consider a climate period consisting of 25 years from September 1, 1995, to August 31, 2019. This period is the maximum window for which the OMERE data are available. The climate forcing data used (e.g., air temperature, rainfall, reference evapotranspiration ET₀) are collected by the meteorological station located at the outlet of the Kamech catchment.

For this climate series, the annual averages for rainfall, air temperature during the vegetation growing season (October to May) and ET₀ are 629 mm, 14.8°C and 1310 mm, respectively (Supplementary Materials - Section 4, Fig. SF2). The years 1996 and 2019 are the wettest, with cumulative rainfall of 1036 mm and 862 mm, respectively. The years 1997, 2002 and 2016 are the driest, with cumulative rainfall of 406 mm, 394 mm and 416 mm, respectively. Regarding air temperature, 1999 and 2009 are the coldest years, with an average air temperature of 14.2°C over the crop growth period [October - May]. The years 2001, 2002 and 2007 are the warmest, with an average air temperature of 15.4°C over the crop growth period [October - May] (Supplementary Materials - Section 4, Fig. SF2).

To deepen the analysis of AquaCrop simulations, we classify the years of the climate series using the FAO aridity index (Spinoni et al. 2014). This index expresses aridity as the ratio of atmospheric water supply (rainfall) to atmospheric water demand (ET₀). We opt for this index because (1) it considers several climate variables when using ET₀ to quantify aridity, and (2) it is suitable for analysing AquaCrop simulations since AquaCrop involves ET₀ when calculating the aboveground biomass. According to the FAO aridity index, the climate series comprises two subhumid years (SH), 10 dry subhumid years (DSH), and 13 semiarid years (SA), accounting for occurrences of 8%, 40%, and 52%, respectively (Fig. 3). Additional details about the calculation of the FAO aridity index are provided in the Supplementary Materials - Section 4.

[Fig. 3 about here.]

2.4. Simulating the upstream runoff

The upstream runoff is quantified using AquaCrop simulations based on the agro-pedo-climatic conditions discussed in Section 2.3. The agro-pedo-climatic conditions include two crop types, nine situations for soil available water capacity (three soil textures and three soil depths), and 25 years of climate. To account for the impluvium area, simulated upstream runoff is weighted by the α ratio, that is, the ratio of the upstream to the downstream plot area (Fig. 2), which is set to three nominal values: 0.5, 1 and 2. By combining two crop types, nine conditions for soil available water capacity, and three ratios of the upstream to the downstream plot area, we obtain 54 situations of upstream runoff and thus 54 simulated time series of runoff, each spanning 25 years. Subsequently, each simulated time series of upstream runoff is added to the corresponding time series of rainfall in the downstream plot.

The set of simulated time series of upstream runoff, after weighting by the α ratio, depicts a range of annual cumulative values from 9 mm to 691 mm, representing 2–97% of the annual rainfall, depending on the year. To further analyse the impact of upstream runoff on downstream crop functioning, we classify these annual cumulative values into four classes relative to three quartiles (Table 1). We refer hereafter to classes of upstream runoff.

[Table 1 about here.]

2.5. Simulating downstream crop functioning

Downstream crop functioning is simulated with AquaCrop, considering the agro-pedo-climatic conditions (Section 2.3) and the upstream runoff (Section 2.4). For each of the two downstream crops (wheat and faba bean) and each of the nine downstream situations in terms of soil available water capacity, 54 AquaCrop simulations are run, varying in the simulated input of upstream runoff. This results in 486 simulations for each of the two downstream crops to be linked for comparison purposes to the corresponding nine reference simulations (3 soil depths, 3 soil textures) of crop functioning without upstream runoff from connectivity. On a yearly basis, these 486 simulations amount to 12,150 simulations for each downstream crop, totaling 24,300 for both.

2.6. Simulation analysis

To study the impact of water infiltration due to the runoff-runon process on downstream crop functioning, we focus on two agronomic variables driven by crop functioning, namely, aboveground biomass (AGB) and yield (Yld). We conduct a quantitative analysis, which involves calculating the relative differences in AGB and Yld between simulations with and without connectivity (Eq. 1, X_WC and X_OC stand for the value of the simulated variable with and without connectivity, respectively). This allows us to (1) globally quantify, for all considered situations, the impact of upstream runoff by hydrological connectivity on the functioning of the downstream crops (wheat and faba bean) and (2) understand the influence, on this impact, of environmental conditions within the downstream plot (upstream runoff, climate forcing, soil texture and depth, crop).

For each of the two downstream crops, the relative difference Δ is calculated at the annual timescale along the 25-year time series for any of the 486 combinations (3 soil depths, 3 soil textures, 54 upstream runoff). A year Y is considered a hydrological year spanning from the beginning of September of the calendar year [Y-1] to the end of August of the calendar year [Y]. This results in a total of 12,150 relative differences calculated for AGB and Yld for each crop in the downstream plot and for each of the 25 years. Δ > 0 (< 0) indicates that additional water input through hydrological connectivity has a positive (negative) impact, since it leads to an increase (a decrease) in AGB and Yld compared to the case without connectivity.

Before analysing all relative differences Δ, it is necessary to conduct some filtering in relation to simulation realism and AquaCrop modelling accuracy.

A first filtering step prevents the analysis of unrealistic simulations. For this, we filter AquaCrop simulations based on an agro-economic constraint, namely, yield. Following field-based expert recommendations, we select simulations with Yld (wheat) > 0.5 ton/ha and Yld (faba bean) > 0.25 ton/ha, knowing that yields below these values are considered null. To avoid eliminating significant impact changes between with and without connectivity, this filter is applied to simulations with connectivity if Δ > 0 and without connectivity if Δ < 0.
A second filtering step defines a threshold value for Δ to account for uncertainties in the AquaCrop simulations. For this, we refer to Dhouib et al. (2022), who reported that the model satisfactorily simulates AGB, with a relative error between observations and simulations of approximately 11%. Therefore, we choose a threshold of 0.11 for the absolute value for Δ, above which the impact of water input through connectivity is considered significant as it exceeds the modelling uncertainty. If negative (positive) Δ values are greater (lower) than or equal to -0.11 (0.11), we consider that the impact of water input through hydrological connectivity on crop functioning is insignificant. Since the model has not been evaluated for yield in the study area, we use the same threshold on Δ for AGB and Yld.

3.1. Overall results for aboveground biomass (AGB) and yield (Yld)

The analysis of relative differences between simulations with and without connectivity indicates that in most situations (combinations of soil available water capacity derived from soil texture and depth, upstream runoff, and climate year), the contribution of hydrological connectivity through runoff infiltration has a nonsignificant impact on AGB/Yld. This holds true for both downstream crops of wheat and faba bean, with more than 85%/77% and 67%/62% of the calculated differences falling between − 0.11 and 0.11, respectively (Fig. 4).

[Fig. 4 about here.]

Beyond the overall results, there are situations in which the contribution of hydrological connectivity through runoff infiltration significantly increases AGB and Yld for both wheat and faba bean crops. The increase is more pronounced for faba bean than for wheat, with 33% of the relative differences (average over AGB and Yld values) being greater than 0.11 for faba bean, compared to 16% only for wheat. This suggests that faba bean is more sensitive to water shortages than wheat and that additional water input via the infiltration of upstream runoff contributes to alleviating this shortage. This is confirmed by the analysis of the water stress coefficient Ks. Indeed, in 69% of the situations considered, faba bean is more often stressed than wheat since it has a lower Ks (Fig. SF3 in Supplementary Material - Section 5), while the increase in Ks induced by upstream runoff averages 9% for faba bean and only 2% for wheat (data not shown). This greater sensitivity of faba bean to water shortage can be explained by (1) a shorter phenological cycle, making its functioning more sensitive to intra-annual variations in rainfall, and (2) a shorter root system (Hamblin and Tennant 1987) that does not allow the crop to use water stored in deeper soil layers. These simulation results converge with the observations of Daryanto et al. (2017), who reported a drought-based yield reduction more important for legume crops than for cereal crops.

When a positive impact is observed, the increase induced by water input through hydrological connectivity is higher for grain yield (Yld) than for aboveground biomass (AGB). This is ascribed to the way that AquaCrop calculates Yld as the product of AGB and the harvest index (HI), where HI is sensitive to water stress (Ali and Talukder 2008; AquaCrop user manual). Thus, the latter has a double effect on Yld via both AGB and HI.

We also observe another type of situation, much rarer (only 3% of relative differences), where water input through hydrological connectivity leads to negative relative differences (Fig. 4). This indicates a decrease in both aboveground biomass (AGB) and yield (Yld), which may be attributed to waterlogging (Liu et al. 2020). Since the frequency of such a situation is low, we mainly focus on situations with a positive impact of hydrological connectivity for the remainder of this paper.

3.2. Influence of environmental conditions on the impact of hydrological connectivity

In the previous section, we showed that hydrological connectivity through the runoff-runon process can increase crop production for wheat and faba bean in some situations. In this section, we investigate the influence of environmental conditions of the downstream plot (hydrological conditions, soil texture and depth, and climate) to understand how some of these conditions can lead to positive impacts. For each downstream crop, we categorise the set of significant Δ values (Δ > 0.11) for both aboveground biomass (AGB) and yield (Yld) based on the respective classes of a given environmental factor (i.e., four classes for upstream runoff, three classes for texture, three classes for soil depth and three classes for climate years). Thus, the cumulative distribution across all classes for any environmental factor and any Δ type (AGB, Yld) adds up to 100%. The resulting statistics are presented in Tables 2, ST5, ST6 and ST7 and are utilised in the subsequent three subsections.

[Table 2 about here.]

3.2.1. Influence of hydrological conditions

We emphasise three significant results related to the simulated water dynamics, from runoff to the soil water content, in connection with AGB and Yld variations.

First, in situations with insignificant impact (Δ ∈ [-0.11 ; +0.11]), the occurrences of the ∆ values are evenly distributed across the four classes of upstream runoff (Supplementary materials - Section 6, Table ST5). In situations with positive impact, the occurrences of the ∆ values are lower for the first two classes of upstream runoff that correspond to the lowest cumulative values (R1 and R2 classes in Table 1), as they account for 40–46% of cases depending on the crop and the production variable (AGB or Yld). Conversely, the occurrence of the ∆ values is larger for the last two classes (R3 and R4) that correspond to the highest cumulative values, with 54–60% of cases.
Second, in situations with a positive impact, the increase in the water infiltration amount induced by upstream runoff averages 10% and 7% relative to wheat and faba bean, respectively (Table ST6). Meanwhile, the increase in infiltration for situations with an insignificant impact is only 8% and 6% for the two crops.
Third, situations with a positive impact show an average increase in the simulated root zone water content (wRZ) of 41% and 24% for wheat and faba bean, respectively (Table ST6). In contrast, for situations with an insignificant impact, the simulated root zone water content increases by only 2% and 1% for these crops.

From these three main results, we can infer that the increase in AGB and Yld depends only partially on the amount of upstream runoff. While positive impact situations are predominantly associated with a significant increase in upstream runoff, even small increases in upstream runoff (Class R1 in Tables 1 and ST5) and in the resulting infiltration (Table ST5) can lead to positive impacts. For our case study, the impact of hydrological connectivity through the runoff-runon process on crop production seems to be primarily determined by the increase in the root zone water content during the crop cycle.

3.2.2. Influence of soil texture and depth

For both crops (wheat and faba bean) and both agronomic variables (aboveground biomass AGB and yield Yld), the deeper the soil, the larger the occurrence of significant Δ values (Table 2). Indeed, the occurrence intervals are [12% − 15%], [31% − 42%], and [43% − 57%] for 0.5, 1, and 1.5 m deep soil, respectively. When dealing with soil texture, a large occurrence of significant Δ values is observed for clay-loam (CL) soils, with an average value of 47% when merging the two crops and two agronomic variables (Table 2). Compared to CL soils, lower occurrences are observed for C and SCL, with average values (over the two crops and two agronomic variables) of 30% and 24%, respectively (Table 2).

These results are consistent with the concept of soil available water capacity and align with recent literature on this topic. We observed increases in aboveground biomass (AGB) and yield (Yld) following the infiltration of upstream runoff, mainly for clay-loam (CL) soils and deeper soils. This observation is explained by (1) a larger soil available water capacity for CL and deeper soils, allowing for a greater storage of water in the root zone (Alkassem et al. 2022; Cousin et al. 2022) and (2) a soil conducive to root zone development in terms of depth (van Leeuwen 2022).

To illustrate these statements, we use the example of wheat sown in 2002 in clay-loam soil and compare crop functioning across the three soil depths considered (Fig. 5). For the same temporal pattern and amount of water received from upstream runoff through hydrological connectivity, the aboveground biomass increases from 1.5 ton/ha at a soil depth of 0.5 m to 6.1 ton/ha at a soil depth of 1.5 m. This difference can be explained by changes in the root zone water content in relation to soil depth. Indeed, upstream runoff increases the amount of water that infiltrates into the soil by 8% for the three soils between simulations with and without connectivity (data not shown). This further increases wRZ by 2%, 35% and 88% for the 0.5 m, 1 m and 1.5 m soil depths, respectively, between simulations with and without connectivity (data not shown). These changes in the root zone water content can be related to changes in crop functioning throughout the crop growth cycle, as discussed below.

The comparison of crop functioning variables for different soil depths, including water storage in the root zone, transpiration, canopy cover, and aboveground biomass, suggests that from the beginning of the crop cycle to January 10, the wheat crop exhibits a similar rooting depth for the three soil depths, thus accessing the same amount of water in the soil.
From January 10 onwards, the crop roots continue to expand downward in the 1 m and 1.5 m deep soils, while they reach their maximum growth in the shallow soil (0.5 m depth). Consequently, the root zone water content decreases for the crop in the shallow soil (0.5 m depth), whereas it increases for the crop in the deeper soils (1 m and 1.5 m depth). This leads, for the shallow soil, to reductions in canopy cover (CC), crop transpiration (Tr), and aboveground biomass (AGB) 50 days later (March 1) when the root zone water content reaches a critical level. Conversely, these reductions are not observed for the crop in the 1 m and 1.5 m deep soils that still benefit from sufficient water content within the root zone.
For wheat sown in 1 m and 1.5 m deep soils, the initial deviation between the two temporal courses of the root zone water content occurs on March 2. Specifically, the root zone water content slightly increases for the 1.5 m deep soil due to deepening root growth, whereas it decreases for the 1 m deep soil as the root system has reached its maximum depth. This leads, for the 1 m depth soil, to reductions in CC, Tr, and AGB 50 days later (April 15), when the soil water content reaches a critical level. Conversely, these reductions are not observed until senescence of the crop in the 1.5 m deep soils that still benefit from sufficient water content within the root zone.

[Fig. 5 about here.]

3.2.3. Influence of climate forcing

When the influence of climate conditions is analysed, the increases in aboveground biomass (AGB) and yield (Yld) due to water input through hydrological connectivity are primarily noticeable in dry subhumid (DSH) and semiarid (SA) years, while being almost negligible in subhumid (SH) years (Table 2). Indeed, occurrence intervals for significant Δ values (Δ > 0.11) are [0% − 5%], [40% − 49%], and [46–60%] for the SH, DSH, and SA years, respectively. For wheat, 0% of significant Δ values are observed in SH years, 42% in DSH years, and 58% in SA years. For faba bean, 4% of significant Δ values are observed in SH years, and the remainder are evenly distributed between DSH and SA years (approximately 48%). Notably, these results may be biassed by the uneven distribution of climate years, comprising 8% of SH years, 40% of DSH years, and 52% SA years. This uneven distribution is an inherent limitation of the study that relies on field observations of climate forcing, although the distribution of significant Δ values does not completely align on that of climate years.

3.2.3.1. During semi-arid years

During SA years, the downstream infiltration of upstream runoff, even in small quantities, increases the water content in the root zone, thereby increasing AGB and Yld. On average, during these years, the increase in the soil water content in the root zone due to upstream runoff is 29% for wheat and 18% for faba bean (Supplementary materials - Section 6, Table ST7). The example of the year 2016 is characteristic of the increase in the root zone water content induced by infiltration. This year is classified as an SA year according to the FAO aridity index (Fig. 3), with annual rainfall accumulation largely below the 25-year average (Supplementary materials - Section 4, Fig. SF2). During this year, the downstream plot benefits from additional runoff between a lower value of 9 mm and an upper value of 105 mm (data not shown), depending on the environmental conditions of the upstream plot (soil, crop, and impluvium area). For the two limits mentioned above, infiltration in the downstream plot (cultivated with wheat) increases by 1% and 18%, respectively, wRZ increases by 4% and 29%, AGB increases by 27% and 112%, and Yld increases by 35% and 406% (median values of all downstream soils, data not shown).

3.2.3.2. During dry subhumid years

The influence of climate forcing on the increase in aboveground biomass (AGB) and yield (Yld) may be linked to the intra-annual variability in rainfall, especially in dry subhumid years. Indeed, when analysing the occurrences of positive impact, calculated for faba bean in dry subhumid years considering all downstream soil textures and depths and all runoff amounts (Supplementary Materials - Section 7, Fig. SF4), we note the following trends.

On the one hand, water input through hydrological connectivity has a frequent positive impact on AGB and Yld (occurrence > 10% in Fig. SF4) for a group of years, namely, 1998, 1999, 2004, 2010, and 2013. These years are characterised by annual rainfall accumulations of 750 mm, 700 mm, 708 mm, 650 mm, and 622 mm, respectively (Supplementary materials - Section 4, Fig. SF2).
On the other hand, water input through connectivity has a less frequent positive impact on AGB and Yld (occurrence < 10% in Fig. SF4) for 2003, 2005, and 2009, characterised by very similar annual rainfall accumulations: 728 mm, 651 mm, and 793 mm, respectively Supplementary materials - Section 4, Fig. SF2).

Therefore, the infiltration of runoff water from upstream is more likely to increase the aboveground biomass (AGB) and yield (Yld) of the faba bean crop for the first group of years, but this is less likely for the second group. To gain further insight into this observation, we examine simulations without hydrological connectivity considering a clay-loam soil with a 1 m depth for two dry subhumid years: in 2003 (728 mm of rainfall), the occurrence of positive impacts among all tested situations is rather low (8%), whereas it is much greater (61%) in 2004 (708 mm of rainfall) (Supplementary Materials - Section 7, Fig. SF4). In the situation without hydrological connectivity, AGB and Yld are lower in 2004 than in 2003. Therefore, the potential crop production increase by upstream runoff is higher for 2004 than for 2003. The analysis of monthly rainfall (Fig. 6a) reveals differences in the intra-annual distribution of rainfall between the two years, leading to subsequent variations in crop growth.

During the hydrological year 2004, September, December, and March are characterised by high monthly rainfall, thereby increasing the amount of water that infiltrates into the soil (Fig. 6b). In contrast, the hydrological year 2003 has a drier start compared to 2004 (except for November). From January onwards, the rainfall in 2003 is more substantial than that in 2004. This intra-annual distribution of rainfall results in a higher water content in the root zone between September and February in 2004 than in 2003 (Fig. 6c). However, this difference in the soil water content between the two years does not have a particularly positive impact on the growth of faba bean in 2004, as the crop is in the early stages of its cycle (Fig. 6d).
From February onwards, the water content in the root zone in 2003 is higher than that in 2004 due to greater rainfall in January, February, and April. This explains the better vegetation development in 2003.

[Fig. 6 about here.]

To compare the impact of hydrological connectivity on crop production during these two years (2003 and 2004), we consider the same agro-pedological situation considered above (faba bean cultivated in a clay-loam soil with 1 m depth) for a downstream plot receiving typical upstream runoff (from an upstream plot cultivated with faba bean in a 1 m deep clayey soil). Then, water input from hydrological connectivity increases infiltration and wRZ by 8% and 10%, respectively, in 2004 and by 11% and 6% in 2003, respectively. Furthermore, the resulting increase in yield is far larger in 2004 (73%) than in 2003 (13%), which is ascribed to the intra-annual variability in rainfall discussed above.

We report the first complete numerical experiment aimed at quantifying the impact of hydrological connectivity between plots on Mediterranean rainfed crop production.

The results show that water input through hydrological connectivity has a positive impact on agricultural production depending on environmental conditions related to crops, climate forcing, and soil properties. Climate forcing significantly influences the impact of hydrological connectivity from both yearly rainfall and the intra-annual distribution of rainfall. A positive impact is observed in dry subhumid and semiarid climate years, suggesting that this impact may become more pronounced in light of the forecasted climate change.

This study represents an initial step in understanding the impact of hydrological connectivity on crop functioning and demands further refinements. Future investigations should broaden the range of considered environmental conditions and address limitations arising from the foundational assumptions of the current study. These improvements encompass the following:

a more detailed characterisation of spatial heterogeneity within agricultural plots;
the incorporation of successive cultivated plots along a hydrological transect to account for cumulative effects;
consideration of the influence of shallow aquifers, which may be substantial in hilly areas; and
examination of changes within agricultural plots induced by agricultural practices throughout the crop growth cycle and by crop rotation.

Moreover, the present study relies on a straightforward approach for runoff. The coupling of crop models with distributed hydrological models should facilitate more realistic simulations of runoff propagation within cultivated landscapes.

Acknowledgements

The OMERE observatory (www.obs-omere.org), funded by the French institutes INRAE and IRD and coordinated by INAT Tunis, INRGREF Tunis, UMR Hydrosciences Montpellier and UMR LISAH Montpellier, is acknowledged for providing a portion of the data used in this study. The OMERE observatory is part of OZCAR Research Infrastructure (https://www.ozcar-ri.org/ ) that is supported by the French Ministry of Research, French Research Institutions and Universities.

Funding

This work was financially supported by the Tunisian Ministry of Higher Education and Scientific Research (MESRS), the French National Research Institute for Sustainable Development (IRD). The current study is part of the ALTOS project in the framework of the PRIMA 2018 program, with financial contributions from France and Tunisia. A CC-BY public copyright license has been applied by the authors to the present document and will be applied to all subsequent versions up to the Author Accepted Manuscript arising from this submission, in accordance with the grant’s open access conditions.

Conflicts of interest/Competing interests

The authors have no conflicts of interest to declare that are relevant to the content of this article.

Ethics approval

Not applicable.

Consent to participate

Not applicable

Consent for publication

Not applicable

Availability of data and material

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Code availability

Code will be made available on request

Authors' contributions

Conceptualization, F.J., I.M., J.M., L.P. and R.Z-C.; Methodology, F.J., J.M., L.P. and M.D.; Software, M.D.; Validation, F.J., I.M., J.M., L.P., M.D. and R.Z-C.; Formal analysis, J.M., L.P. and M.D.; Investigation, F.J., I.M., J.M., L.P., M.D. and R.Z-C.; Resources, I.M. and R.Z-C.; Data Curation, M.D.; Writing – original draft preparation, F.J., J.M., L.P. and M.D.; Writing – review and editing, F.J., J.M. and M.D.; Visualization, M.D.; Supervision, F.J. and J.M.; Project administration, F.J.; Funding acquisition, F.J., I.M., J.M., M.D. and R.Z-C.

Ali MH, Talukder MSU (2008) Increasing water productivity in crop production—A synthesis. Agric Water Manage 95:1201–1213. https://doi.org/10.1016/j.agwat.2008.06.008
Alkassem M, Buis S, Coulouma G et al (2022) Estimating soil available water capacity within a Mediterranean vineyard watershed using satellite imagery and crop model inversion. Geoderma 425:116081. https://doi.org/10.1016/j.geoderma.2022.116081
Ammar A, Riksen M, Ouessar M, Ritsema C (2016) Identification of suitable sites for rainwater harvesting structures in arid and semi-arid regions: A review. Int Soil Water Conserv Res 4:108–120. https://doi.org/10.1016/j.iswcr.2016.03.001
Anderson W, Johansen C, Siddique KHM (2016) Addressing the yield gap in rainfed crops: a review. Agron Sustain Dev 36:18. https://doi.org/10.1007/s13593-015-0341-y
Araya A, Kisekka I, Gowda PH, Prasad PVV (2017) Evaluation of water-limited cropping systems in a semi-arid climate using DSSAT-CSM. Agric Syst 150:86–98. https://doi.org/10.1016/j.agsy.2016.10.007
Besbes M, Chahed J, Hamdane A (2014) Sécurité hydrique de la Tunisie: gérer l’eau en conditions de pénuries. l’Harmattan, Paris
Blinda M, Thivet G (2009) Ressources et demandes en eau en Méditerranée: situation et perspectives. Sécheresse 20:009–016. https://doi.org/10.1684/sec.2009.0162
Bracken LJ, Croke J (2007) The concept of hydrological connectivity and its contribution to understanding runoff-dominated geomorphic systems. Hydrol Process 21:1749–1763. https://doi.org/10.1002/hyp.6313
Chen L, Sela S, Svoray T, Assouline S (2016) Scale dependence of Hortonian rainfall-runoff processes in a semiarid environment: SCALE DEPENDENCE OF SEMIARID RAINFALL-RUNOFF PROCESSES. Water Resour Res 52:5149–5166. https://doi.org/10.1002/2015WR018315
Coulouma G, Ciampalini R, Annabi M et al (2017) Synthèse de la prospection pédologique sur le bassin versant de kamech dans le cadre du projet MASCC. UMR LISAH, Montpellier
Cousin I, Buis S, Lagacherie P et al (2022) Available water capacity from a multidisciplinary and multiscale viewpoint. A review. Agron Sustain Dev 42:46. https://doi.org/10.1007/s13593-022-00774-8
Daccache A, Elbana MA, Fouial A et al (2016) Gestion des ressources en eau. In: Mediterra: zéro gaspillage en Méditerranée: ressources naturelles, alimentations et connaissances. pp 71–93
Daryanto S, Wang L, Jacinthe P-A (2017) Global synthesis of drought effects on cereal, legume, tuber and root crops production: A review. Agric Water Manage 179:18–33. https://doi.org/10.1016/j.agwat.2016.04.022
Dhouib M, Zitouna-Chebbi R, Prévot L et al (2022) Multicriteria evaluation of the AquaCrop crop model in a hilly rainfed Mediterranean agrosystem. Agric Water Manage 273:107912. https://doi.org/10.1016/j.agwat.2022.107912
Fader M, Giupponi C, Burak S et al (2020) First Mediterranean Assessment Report – Chap. 3.1: Resources – Water. Zenodo
FAO (2022) L’agriculture pluviale face aux changements climatiques en Afrique du Nord: Impact et perspective avec l’agroécologie. FAO, Tunis, Tunisia
Gnouma R (2006) Aide à la calibration d’un modèle hydrologique distribué au moyen d’une analyse des processus hydrologiques: application au bassin versant de l’Yzeron. L’institut national des sciences appliquées de Lyon
Hamblin AP, Tennant D (1987) Root length density and water uptake in cereals and grain legumes: how well are they correlated. Aust J Agric Res 38:513–527. https://doi.org/10.1071/ar9870513
Hossain A, Sab AE, Barutcular C et al (2020) Sustainable crop production to ensuring food security under climate change: A Mediterranean perspective. Aust J Crop Sci 14:439–446. https://doi.org/10.3316/informit.121077065862236
Howes DA, Abrahams AD (2003) Modeling runoff and runon in a desert shrubland ecosystem, Jornada Basin, New Mexico. Geomorphology 53:45–73. https://doi.org/10.1016/S0169-555X(02)00347-1
Jones OD, Sheridan GJ, Lane PN (2013) Using queuing theory to describe steady-state runoff-runon phenomena and connectivity under spatially variable conditions. Water Resour Res 49:7487–7497. https://doi.org/10.1002/2013WR013803
Karabulut AA, Crenna E, Sala S, Udias A (2018) A proposal for integration of the ecosystem-water-food-land-energy (EWFLE) nexus concept into life cycle assessment: A synthesis matrix system for food security. J Clean Prod 172:3874–3889. https://doi.org/10.1016/j.jclepro.2017.05.092
Lange MA, Llasat MC, Snoussi M et al (2020) First Mediterranean Assessment Report – Chap. 1: Introduction. Zenodo
Liu J, Lobb DA (2021) An Overview of Crop and Crop Residue Management Impacts on Crop Water Use and Runoff in the Canadian Prairies. Water 13:2929. https://doi.org/10.3390/w13202929
Liu K, Harrison MT, Shabala S et al (2020) The State of the Art in Modeling Waterlogging Impacts on Plants: What Do We Know and What Do We Need to Know. Earth’s Future 8. https://doi.org/10.1029/2020EF001801
Mbava N, Mutema M, Zengeni R et al (2020) Factors affecting crop water use efficiency: A worldwide meta-analysis. Agric Water Manage 228:105878. https://doi.org/10.1016/j.agwat.2019.105878
Mekki I, Albergel J, Ben Mechlia N, Voltz M (2006) Assessment of overland flow variation and blue water production in a farmed semi-arid water harvesting catchment. Phys Chem Earth Parts A/B/C 31:1048–1061. https://doi.org/10.1016/j.pce.2006.07.003
Mekki I, Zitouna-Chebbi R, JACOB F et al (2018) Impact of land use on soil water content in a hilly rainfed agrosystem: a case study in the Cap Bon peninsula in Tunisia. https://doi.org/10.7251/AGRENG1801064M. AGROFOR 3:
Molénat J, Barkaoui K, Benyoussef S et al (2023) Diversification from field to landscape to adapt Mediterranean rainfed agriculture to water scarcity in climate change context. Curr Opin Environ Sustain 65:101336. https://doi.org/10.1016/j.cosust.2023.101336
Molénat J, Raclot D, Zitouna R et al (2018) OMERE: A Long-Term Observatory of Soil and Water Resources, in Interaction with Agricultural and Land Management in Mediterranean Hilly Catchments. Vadose Zone J 17:180086. https://doi.org/10.2136/vzj2018.04.0086
Nanda A, Sen S, McNamara JP (2019) How spatiotemporal variation of soil moisture can explain hydrological connectivity of infiltration-excess dominated hillslope: Observations from lesser Himalayan landscape. J Hydrol 579:124146. https://doi.org/10.1016/j.jhydrol.2019.124146
Nouri H, Stokvis B, Chavoshi Borujeni S et al (2020) Reduce blue water scarcity and increase nutritional and economic water productivity through changing the cropping pattern in a catchment. J Hydrol 588:125086. https://doi.org/10.1016/j.jhydrol.2020.125086
Nunes AN, de Almeida AC, Coelho COA (2011) Impacts of land use and cover type on runoff and soil erosion in a marginal area of Portugal. Appl Geogr 31:687–699. https://doi.org/10.1016/j.apgeog.2010.12.006
Peredo D, Ramos M-H, Andréassian V, Oudin L (2022) Investigating hydrological model versatility to simulate extreme flood events. Hydrol Sci J 67:628–645. https://doi.org/10.1080/02626667.2022.2030864
Prosdocimi M, Jordán A, Tarolli P et al (2016) The immediate effectiveness of barley straw mulch in reducing soil erodibility and surface runoff generation in Mediterranean vineyards. Sci Total Environ 547:323–330. https://doi.org/10.1016/j.scitotenv.2015.12.076
Raes D, Steduto P, Hsiao TC, Fereres E (2009) AquaCrop -The FAO Crop Model to Simulate Yield Response to Water: II. Main Algorithms and Software Description. Agron J 101:438–447. https://doi.org/10.2134/agronj2008.0140s
Ramos MC, Martínez-Casasnovas JA (2006) Trends in Precipitation Concentration and Extremes in the Mediterranean Penedès-Anoia Region, Ne Spain. Clim Change 74:457–474. https://doi.org/10.1007/s10584-006-3458-9
Saco PM, Rodríguez JF, Moreno-de las Heras M et al (2020) Using hydrological connectivity to detect transitions and degradation thresholds: Applications to dryland systems. CATENA 186:104354. https://doi.org/10.1016/j.catena.2019.104354
Schiettecatte W, Ouessar M, Gabriels D et al (2005) Impact of water harvesting techniques on soil and water conservation: a case study on a micro catchment in southeastern Tunisia. J Arid Environ 61:297–313. https://doi.org/10.1016/j.jaridenv.2004.09.022
Schoener G, Stone MC (2019) Impact of antecedent soil moisture on runoff from a semiarid catchment. J Hydrol 569:627–636. https://doi.org/10.1016/j.jhydrol.2018.12.025
Souissi A, Mtimet N, Thabet C et al (2019) Impact of food consumption on water footprint and food security in Tunisia. Food Sec 11:989–1008. https://doi.org/10.1007/s12571-019-00966-3
Spinoni J, Vogt J, Naumann G et al (2014) Towards identifying areas at climatological risk of desertification using the Köppen-Geiger classification and FAO aridity index: TOWARDS IDENTIFYING AREAS AT CLIMATOLOGICAL RISK OF DESERTIFICATION. Int J Climatol 35:2210–2222. https://doi.org/10.1002/joc.4124
Steduto P, Hsiao TC, Raes D, Fereres E (2009) AquaCrop-The FAO Crop Model to Simulate Yield Response to Water: I. Concepts and Underlying Principles. Agron J 101:426–437. https://doi.org/10.2134/agronj2008.0139s
Tadros MJ, Al-Mefleh NK, Othman YA, Al-Assaf A (2021) Water harvesting techniques for improving soil water content, and morpho-physiology of pistachio trees under rainfed conditions. Agric Water Manage 243:106464. https://doi.org/10.1016/j.agwat.2020.106464
Todorovic M, Albrizio R, Zivotic L et al (2009) Assessment of AquaCrop, CropSyst, and WOFOST Models in the Simulation of Sunflower Growth under Different Water Regimes. Agron J 101:509–521. https://doi.org/10.2134/agronj2008.0166s
Van Gaelen H, Vanuytrecht E, Willems P et al (2017) Bridging rigorous assessment of water availability from field to catchment scale with a parsimonious agro-hydrological model. Environ Model Softw 94:140–156. https://doi.org/10.1016/j.envsoft.2017.02.014
Van Leeuwen C (2022) 9 - Terroir: The effect of the physical environment on vine growth, grape ripening, and wine sensory attributes. In: Reynolds AG (ed) Managing Wine Quality (Second Edition). Woodhead Publishing, pp 341–393
Van Loo M, Verstraeten G (2021) A Spatially Explicit Crop Yield Model to Simulate Agricultural Productivity for Past Societies under Changing Environmental Conditions. Water 13:2023. https://doi.org/10.3390/w13152023
Wolka K, Mulder J, Biazin B (2018) Effects of soil and water conservation techniques on crop yield, runoff and soil loss in Sub-Saharan Africa: A review. Agric Water Manage 207:67–79. https://doi.org/10.1016/j.agwat.2018.05.016
Yadari HE, Chikhaoui M, Naimi M et al (2019) Techniques de conservation des eaux et des sols au Maroc: Aperçu et perspectives. 1 7
Yang H, Zehnder AJB (2002) Water Scarcity and Food Import: A Case Study for Southern Mediterranean Countries. World Dev 30:1413–1430. https://doi.org/10.1016/S0305-750X(02)00047-5
Zuecco G, Rinderer M, Penna D et al (2019) Quantification of subsurface hydrologic connectivity in four headwater catchments using graph theory. Sci Total Environ 646:1265–1280. https://doi.org/10.1016/j.scitotenv.2018.07.269

Tables 1 and 2 are available in the Supplementary Files section.

Table1.docx.pdf
Table 1: Classification of the cumulative values of upstream runoff. R1, R2, R3, and R4 correspond to Classes 1, 2, 3, and 4, respectively, of cumulative values of upstream runoff. Q1, Q2, and Q3 correspond to the 1st quartile, median, and 3rd quartile, respectively.
Table2.docx.pdf
Table 2: Percentage occurrence of significant and positive relative differences (Δ > 0.11) categorised by environmental factors. The abbreviations SH, DSH, and SA represent subhumid, dry subhumid, and semiarid years, respectively. The labels R1, R2, R3, and R4 correspond to upstream runoff classes 1, 2, 3, and 4, respectively. The labels C, CL, and SCL stand for clay, clay-loam, and sandy-clay-loam textures, respectively.
SupplementaryMaterialsV2.docx.pdf

Download PDF

Reviewers agreed at journal
03 Jan, 2024
Reviewers invited by journal
02 Jan, 2024
Editor invited by journal
20 Dec, 2023
Editor assigned by journal
13 Dec, 2023
First submitted to journal
13 Dec, 2023

You are reading this latest preprint version

Impact of hydrological connectivity on rainfed annual crops in Mediterranean hilly landscapes: a numerical exploration

Status:

Version 1

Abstract

Figures

1. Introduction

2. Materials and methods

2.1. General framework

2.2. Overview of the AquaCrop crop model

2.3. Setting the agro-pedo-climatic conditions

2.3.1. Crop types and subsequent crop parameters

2.3.2. Soil characteristics and hydrodynamic properties

2.3.3. Climate forcing

2.4. Simulating the upstream runoff

2.5. Simulating downstream crop functioning

2.6. Simulation analysis

3. Results and discussion

3.1. Overall results for aboveground biomass (AGB) and yield (Yld)

3.2. Influence of environmental conditions on the impact of hydrological connectivity

3.2.1. Influence of hydrological conditions

3.2.2. Influence of soil texture and depth

3.2.3. Influence of climate forcing

3.2.3.1. During semi-arid years

3.2.3.2. During dry subhumid years

4. Conclusion

Declarations

References

Tables

Supplementary Files

Status:

Version 1