Quantification of Ecohydrological Sensitivities and Their Influencing Factors at the Seasonal Scale

Ecohydrological sensitivity is defined as the response intensity of streamflow to per unit vegetation change. Understanding of 20 ecohydrological sensitivity and its influencing factors is important for managing water supply, reducing water-related hazards and ensuring aquatic functions by vegetation management. However, this topic has rarely been examined. In this study, 14 large watersheds across various environmental gradients in China were selected to quantify ecohydrological sensitivities at the seasonal scale and to examine their influencing factors such as climate, vegetation, topography, soil and landscape. Based on the variables identified by correlation analysis and factor analysis, the prediction models of seasonal ecohydrological 25 sensitivity were constructed to test their utilities for the design of watershed management and protection strategies. Our key findings were: (1) ecohydrological sensitivities were more sensitive in dry conditions than in wet conditions, for example, 1% LAI (leaf area index) change averagely induced 5.05% and 1.96% change in dry and wet season streamflows, respectively; (2) seasonal ecohydrological sensitivities were highly variable across the study watersheds with different climate condition, dominant soil type and hydrological regime; and (3) the dry season ecohydrological sensitivity was mostly determined by 30 topography (slope, slope length, valley depth, downslope distance gradient), soil (topsoil organic carbon, topsoil bulk density) and vegetation (LAI), while the wet season ecohydrological sensitivity was mainly controlled by soil (topsoil available water holding capacity), landscape (edge density) and vegetation (leaf area index). Our study provided a useful and practical framework to assess and predict ecohydrological sensitivities at the seasonal scale. We expect that ecohydrological sensitivity prediction models can be applied to ungauged watersheds or watersheds with limited hydrological data to help decision makers 35 https://doi.org/10.5194/hess-2020-336 Preprint. Discussion started: 21 July 2020 c © Author(s) 2020. CC BY 4.0 License.

and watershed managers to effectively manage hydrological impacts through vegetation restoration programs. We conclude that ecohydrological sensitivities at the seasonal scale were varied by climate, vegetation and watershed property, and their understanding can greatly support management of hydrological risks and protection of aquatic functions.

Introduction
Natural rivers often have a distinctive seasonal pattern of flow, where flow is highly related to precipitation and shows large 40 variations over dry and wet seasons. Seasonal flows determine ecosystem functions (Toledo-Aceves et al., 2011;Bruijnzeel et al., 2011;Salve et al., 2011), and their responses to vegetation change are highly variable and consequently affect watershed ecosystem equilibrium (Maeda et al., 2015). On the one hand, wet season flows and their variability regulate flood magnitudes (Arias et al., 2012), determine the structure of floodplains and channel morphology (Jansen and Nanson, 2010), and provide opportunities of recruitment of large woody debris (Warfe et al., 2011;de Paula et al., 2011). On the other hand, dry season 45 flows are critical for maintaining a stable water supply and protecting aquatic ecosystem, as well as playing important roles in sustaining aquatic biota and refuging juvenile fishes (Bunn et al., 2006;Palmer and Ruhi, 2019). However, seasonal streamflow can be significantly affected by forest or vegetation change (Dai, 2011;Hirabayashi et al., 2013). Researches have showed that vegetation change can influence water retention time (Moore and Wondzell, 2005;Baker and Wiley, 2009;Bisantino et al., 2015), alter snow accumulation and snowmelt processes (Lin and Wei, 2008;Calder, 2005), and route 50 river flow quickly to downstream (Winkler et al., 2010;Chang, 2012) and consequently increase the frequency and size of floods in wet season. Vegetation change can also affect dry season flow, which may influence baseflow level and the risk of droughts, and degrade or enrich in-channel habitat for aquatic species (Simonit and Perrings, 2013;Sun et al., 2016). Thus, understanding of seasonal hydrological variations to vegetation change is critical for maintaining sustainable water supply, preventing large floods and droughts, and developing sound watershed management plans. 55 Obviously, seasonal streamflow response to vegetation change is highly variable among watersheds worldwide. To better understand the general pattern of streamflow response to vegetation change, a uniform indicator named ecohydrological sensitivity (defined as the response intensity of streamflow to per unit forest or vegetation change) has been firstly introduced by Zhang et al. (2017). And ecohydrological sensitivity is believed to be controlled not only by the proportion of forest or vegetation cover change but also by climate condition, hydrological regime and forest or vegetation type (Zhang et al., 2017;60 Li et al., 2017). Assessing ecohydrological sensitivity can provide various benefits. For example, it provides a dimensionless index on vegetation-water relationship so that any watersheds can be effectively compared. It also allows for predicting ecohydrological sensitivities for a landscape or region so that negative hydrological impacts in the areas with high ecohydrological sensitivities can be minimized through suitable arrangements of vegetation or watershed management strategies. However, in spite of its usefulness, ecohydrological sensitivity and its influencing factors have been rarely 65 quantified. To our knowledge, there is no any study on quantification of seasonal ecohydrological sensitivity. https://doi.org/10.5194/hess-2020-336 Preprint. Discussion started: 21 July 2020 c Author(s) 2020. CC BY 4.0 License.
Ecohydrological sensitivity is likely varied with time scales. The hydrological responses to vegetation change at the annual scale are the averaged and cumulative effects from those at shorter time intervals, which are typically associated with total annual magnitudes such as water yield or production, while those at daily or monthly or seasonal scales affect flow patterns and are normally related to floods and droughts. The seasonal scale is a medium level between daily and annual scales, 70 which can affect both magnitude and pattern in terms of hydrological response and sensitivity. For example, the interactions between vegetation and water are quite different between dry and wet seasons (Donohue et al., 2010;Asbjornsen et al., 2011).
Abundant water is available for vegetation growth in wet season, while vegetation in dry season mostly relies on limited soil moisture or groundwater for limited photosynthesis and transpiration. Besides, streamflow generation in wet season is mainly based on precipitation or water input, whereas dry season flow is controlled by soil moisture in the antecedent wet season and 75 groundwater discharge. Thus, the contrasted processes in different seasons suggest that ecohydrological sensitivity must be examined at a seasonal scale.
Various factors including climate, vegetation and watershed property affect hydrological responses or sensitivities (Zhou et al., 2015;Li et al., 2017;Zhang et al., 2017). For examples, hydrological responses to land cover change tend to be more sensitive in non-humid regions (Zhou et al., 2015). Evapotranspiration change related to vegetation change is controlled 80 by energy and water (Zhang et al., 2004;Creed et al., 2014;Yang et al., 2007). Topography controls hydrological processes by affecting the distribution and routing of water (Woods, 2007). Soil and landscape conditions are important for erosion, sediment and flow connectivity (Borselli et al., 2008). Clearly, fully assessing and understanding ecohydrological sensitivity requires a consideration of various influencing variables. Although past studies have focused on the hydrological influences of a single type of variables such as vegetation change (Beck et al., 2013;Feng et al., 2016;van Dijk et al., 2012), climate 85 (Creed et al., 2014;Miara et al., 2017), topography (Lyon et al., 2012;Jencso and McGlynn, 2011;Li et al., 2018a) and landscape (Nippgen et al., 2011;Buma and Livneh, 2017;Teutschbein et al., 2018), the inclusion of various types of variables into an integrated assessment framework of hydrological responses has been a challenging subject.
China has experienced substantial and dynamic vegetation change over the past few decades. Deforestation and biomass loss dominated vegetation change from 1950s to 1980s , while the large-scale revegetation programs 90 have been implemented since 1980s (Li et al., 2018b). Substantial and dynamic vegetation change in China provides such a great opportunity for assessing seasonal ecohydrological sensitivity and its influencing factors. The objectives of this study were: (1) to evaluate seasonal ecohydrological sensitivity in the selected large watersheds across environmental gradients; (2) to examine the role of climate, vegetation, topography, soil and landscape in seasonal ecohydrological sensitivity; and (3) to simulate and predict seasonal ecohydrological sensitivity based on the selected factors. 95 2 Study watersheds and data

Study watersheds
Fourteen large watersheds across climatic zones with the area ranging from 832 to 19189 km 2 in China were selected in this study. They include the Pingjiang and Xiangshui watersheds in Southeast China, the Tangwang River and Xinancha River watersheds in Northeast China, the Upper Zagunao, Zagunao, Upper Heishui River, Heishui River, Gongbujiangda and 100 Gengzhang watersheds in Southwest China and the Dongchuan, Heishuichuan, Jingchuan and Rui River watersheds in Northwest China (Fig. 1). Table 1 provides a brief summarise of climate, vegetation and topography in the study watersheds.
The selected watersheds are mainly located in subtropical monsoon climate, temperate continental monsoon climate, alpine climate and temperate continental climate zones. In addition, substantial vegetation restoration programs caused large-scale vegetation change from 1980s onwards. To evaluate seasonal ecohydrological sensitivity, the study periods start from 1983. 105 Detailed information about climate, topography, soil and vegetation in each watershed can be found in the Supplement (Sect. S1).

Data
Daily or monthly discharges for 14 watersheds were obtained from various government agencies. The details about the study periods and hydrometric stations can be found in the Supplement (Table S3). Discharges (m 3 /s) were converted into the unit 110 of mm according to drainage area. A hydrological year was equally divided into dry season and wet season, and then seasonal flows were calculated accordingly.
The historical climate data used in this study include three sources: daily climate records from National Meteorological Information Centre of China Meteorological Administration (CMA: http://data.cma.cn/), spatial-interpolated gridded climate data by use of the ANUSPLIN model and meteorological data collected at the associated hydrological stations 115 or rain gauges (Table S3). In this study, daily or monthly climate data including mean temperature (Tmean), minimum temperature (Tmin), maximum temperature (Tmax) and precipitation (P) were derived and calculated accordingly. Monthly potential evapotranspiration (PET) was calculated based on estimated Tmax and Tmin by using Hargreaves' equation (Hargreaves and Samani, 1985).  (Table S2). 125 Leaf area index (LAI) derived from the Global Land Surface Satellite LAI Product (GLASS LAI) was used as a vegetation index to express vegetation change in this study (GLASS: http://glass-product.bnu.edu.cn/). The GLASS LAI https://doi.org/10.5194/hess-2020-336 Preprint. Discussion started: 21 July 2020 c Author(s) 2020. CC BY 4.0 License. product dataset provides continuous global LAI at a high temporal resolution of eight days (Liang et al., 2013;Xiao et al., 2014). There are two types of GLASS LAI products with different spatial resolutions and available periods. The first GLASS LAI product is based on Advanced Very High Resolution Radiometer (AVHRR) reflectance data with spatial resolution of 130 0.05°, and this dataset is available from 1982 to 2016. The other one, with a higher spatial resolution of 1 km is retrieved from Moderate Resolution Imaging Spectroradiometer (MODIS) reflectance data, but it only covers a period of 17 years from 2000 to 2016. As the study watersheds are large (>500 km 2 ) and the study periods are ended before 2006, the former GLASS LAI product was chosen for this study.
Harmonized World Soil Database (HWSD) published by Food and Agriculture Organization (FAO) and International 135 Institute for Applied Systems Analysis (IIASA) with the spatial resolution of 1km was used to collect soil indices (Wieder, 2014). HWSD classifies soil into topsoil from surface to 30 cm below ground, and subsoil between 30 cm and 100 cm below ground.
DEMs with the spatial resolution of 30m derived from GDEMDEM were provided by Geospatial Data Cloud site, Computer Network Information Centre, Chinese Academy of Sciences (http://www.gscloud.cn). Topographical information 140 of the study watersheds was derived from DEMs. Zhang et al. (2017) introduced the ecohydrological sensitivity as the response intensity of annual streamflow to forest cover change. In this study, we defined seasonal ecohydrological sensitivity (Sf) as the response intensity of seasonal streamflow 145 variations to per unit vegetation change (using the leaf area index (LAI) as a proxy), which can be computed with equations (1)-(2). Seasonal streamflow variations to vegetation change (∆Qv) were determined by an improved single watershed approach (see Sect. 2 in the Supplement for more details) (Hou et al., 2018a;Hou et al., 2018b). The value of seasonal ecohydrological sensitivity refers to the percentage of seasonal streamflow changes induced by 1% of LAI change. Ecohydrological sensitivity is highly dependent on factors such as climate, hydrological regime, vegetation type, soil condition, and etc. 150

Definition and calculation of ecohydrological sensitivity
where, , refers to the long-term mean seasonal streamflow during the study period; ∆Qv is seasonal streamflow response to vegetation change in mm, which is calculated by the improved single watershed approach; ∆Qv% is seasonal streamflow response to vegetation change in percentage (%); and ∆LAI is LAI variation compared to average LAI in the reference period 155 in %. https://doi.org/10.5194/hess-2020-336 Preprint. Discussion started: 21 July 2020 c Author(s) 2020. CC BY 4.0 License.

Quantification of influencing drivers to seasonal ecohydrological sensitivities
Five types of indices including climate, vegetation, topography, soil and landscape were adopted in this study. Detailed information on the interpretations and calculations of 40 indices were presented in Table 2. Climate indices, including dryness index and effective precipitation can demonstrate water input and climate condition in a given watershed (van Dijk et al., 2012;160 Jones et al., 2012;Zhang et al., 2004). Dryness index is calculated at the annual scale to demonstrate dryness condition, while effective precipitation (an integrated index of climatic variability) in dry season and wet season denotes seasonal water inputs.
Vegetation growth is highly dependent on temperature, water, soil and geographical location (Chang, 2012). Vegetation coverage or forest coverage indicates a proportion of vegetation or forest in a watershed, but it cannot express vegetation growth, mortality, and seasonality. Thus, LAI is recognized as a better indicator mainly because it is an important biophysical 165 variable relating to photosynthesis, transpiration and energy balance (Launiainen et al., 2016;Verrelst et al., 2016;González-Sanpedro et al., 2008). Topographic indices can be classified into two groups: primary and secondary (also known as compounded topographic indices) Moore et al., 1991). Primary topographic indices can be directly derived from DEM, whilst compounded topographic indices are based on one or more primary indices . In this study, 17 topographic indices including 5 primary indices and 12 compounded indices were selected to describe watershed 170 characteristics like visibility, generation process and morphology (Yokoyama et al., 2002;Park et al., 2001;Jenness, 2004).
Calculations of the topographic indices were made in ArcGIS 10.2 (ERSI) and SAGA GIS 2.1. Soil types were based on the FAO-85 system classification, while soil organic carbon and sanity were directly derived from HWSD in ArcGIS 10.2 (ERSI), and soil available water holding capacity, saturated hydraulic conductivity and bulk density were calculated using Soil-Plant-Air-Water (SPAW) hydrology model. We used weighted average value to represent watershed-scale soil indices. Seven 175 landscape indices including patch number (PN), patch density (PD), largest patch index (LPI), edge density (ED), contagion index (CONTAG), Shannon's diversity index (SHDI) and Simpson's diversity index (SIDI) at the landscape level were derived from FRAGSTATS 4.2 software.
According to dryness index (DI), watersheds were grouped into energy-limited (EL), equitant (EQ) and water-limited (WL) conditions (McVicar et al., 2012). The most widely distributed soil type in a watershed was treated as the dominant soil 180 type. Following our analysis, four dominant soil types (LIXISOLS, LUVISOLS, LEPTOSOLS and CAMBISOLS) were shown in this study. Additionally, the selected watersheds were categorized into rain-dominated (RD) and rain-snow hybrid (Hybrid) watersheds according to their hydrological regimes. Table 3 showed the detailed classifications for each watershed in terms of climate condition, dominant soil type and hydrological regime.
Non-parametric Mann Whitney U test was performed to detect the statistically significant differences between the 185 watershed groups. Mann Whitney U test was applied to test whether there are significant differences in the median values of seasonal ecohydrological sensitivities between two groups (Birnbaum, 1956). Kendall correlation analysis and linear regression were used to identify statistically significant correlations between seasonal ecohydrological sensitivities and 40 indices at a significant level of p=0.10. The insignificant indices were excluded for prediction described below. https://doi.org/10.5194/hess-2020-336 Preprint. Discussion started: 21 July 2020 c Author(s) 2020. CC BY 4.0 License.

Prediction of seasonal ecohydrological sensitivity 190
Seasonal ecohydrological sensitivity can be predicted and simulated based on those identified significant variables. To accomplish this, factor analysis (FA) was introduced to further reduce the redundancy of indices. Similar to principal component analysis (PCA), indices after filtering by factor analysis could retain important information, which means that less indices can be used to represent most information (Lyon et al., 2012). Three criteria were used to pick highly related indices: the coefficient of Kaiser-Meyer-Olkin (KMO) test, p value of Bartlett's test, and the diagonal coefficients of anti-image 195 correlation matrix . Indices filtered by factor analysis with coefficient of KMO being greater than 0.50, the p value of Bartlett's test being less than 0.05 and the diagonal coefficients of anti-image correlation matrix being greater than 0.50 were selected for further analysis.
Multiple linear regression model modified by stepwise regression was employed to predict seasonal ecohydrological sensitivity. Influencing factors filtered by correlation analysis and factor analysis were regarded as independent variables and 200 ecohydrological sensitivity was considered as a dependent variable in a linear regression model. Independent variables were inputted into a model one by one, and ANOVA test was conducted accordingly. Once the p value of ANOVA test was greater than 0.1, the input independent variable at this stage would be dropped. The optimal linear regression model was reached when no independent variables were inputted and no variables were dropped. The Akaike Information Criterion (AIC) and R 2 were used to find optimal multiple linear regression model for prediction. Except for quantitative indices, climate condition, 205 dominant soil type and hydrological regime might also make contributions to prediction of ecohydrological sensitivity. As a result, we introduced dummy variables to quantify the influence of climate condition, dominant soil type and hydrological regime on model accuracy (Hardy, 1993). In this study, ecohydrological sensitivity based on the improved single watershed approach was called the observed Sf, while ecohydrological sensitivity from the multiple linear regression model was named as the predicted Sf. 210 Table 4 showed the comparison of ecohydrological sensitivities between the dry and wet seasons. The ecohydrological sensitivities in the dry season were significantly greater than those in the wet season ( Fig. 2 and Fig. S8-S10). As shown in  (Table S4). 220

Seasonal ecohydrological sensitivity and its variations
Comparisons of seasonal ecohydrological sensitivities were made among the study watersheds grouped by their climate conditions, dominant soil types and hydrological regimes (Fig. 3, Fig. 4 and Fig. 5). As suggested by Fig. 3 and Table   5, significant differences in both dry season and wet season ecohydrological sensitivities between energy-limited (EL) and equitant (EQ) watersheds and between energy-limited and water-limited (WL) watersheds were found. Significant differences in the medians of wet season ecohydrological sensitivity in the pair of EQ-WL were also detected. 1% vegetation change 225 caused 2.09%, 5.86% and 5.23% change of dry season streamflow in the energy-limited, equitant and water-limited watersheds, respectively (Fig. 3a), while it only led to 0.59%, 2.82% and 1.64% change of wet season streamflow in the EL, EQ and WL watersheds, respectively (Fig. 3b). These results clearly demonstrated that ecohydrological sensitivity was greater in the EQ and WL conditions, particularly in the dry season.
Fig. 5 demonstrated the differences of seasonal ecohydrological sensitivity in watersheds grouped by hydrological 240 regime. Mann-Whitney U test showed that there were significant differences between rain-dominated and hybrid watersheds in medians of dry season ecohydrological sensitivity (Table 5). 1% vegetation change can result in 6.51% and 3.29% change of dry season streamflow in rain-dominated and hybrid watersheds, respectively (Fig. 5a), while it only led to 1.75% and 2.20% change of wet season streamflow in rain-dominated and hybrid watersheds, respectively (Fig. 5b).

Prediction models for seasonal ecohydrological sensitivity 245
According to correlations between seasonal ecohydrological sensitivity and 40 indices detected by Kendall correlation and linear regression, 17 indices being significantly related to dry season ecohydrological sensitivity were identified (Table 6) (Table 7). Meanwhile, factor analysis identified 6 indices 260 (Pe, CON, Tw, Thy, Shy and ED) associated with wet season ecohydrological sensitivity based on correlation analysis. For wet season subset, the coefficient of KMO with the value of 0.634 was lower than that in dry season subset, but diagonal coefficients of anti-image correlation matrix were higher than those in wet season subset (≥0.57). The p value of Bartlett's test was 0.00. Given it is an important ecohydrological indicator for vegetation status in a watershed, LAI was also manually added as a predictor in the predicted model. Fig. 6 showed the structure, parameters and statistics of the established prediction models 265 for ecohydrological sensitivity. The dry season model had a better performance with a higher R 2 of 0.966 (Fig. 6a), while the R 2 was only 0.501 for the wet season model (Fig. 6b).

Seasonal ecohydrological sensitivity and climate conditions
Climate conditions in terms of energy (temperature) and water (precipitation) are the most important drivers for vegetation 270 growth. Ecohydrological processes of vegetative watersheds vary greatly with climate conditions (Donohue et al., 2010). As suggested by our study, both dry season and wet season ecohydrological sensitivities of the water-limited watersheds were higher than those of the energy-limited watersheds (Fig. 3), and the dry season ecohydrological sensitivities were much higher than the wet season ecohydrological sensitivities (Fig. 2). In addition, the dry season ecohydrological sensitivity significantly increased with rising dryness index while the wet season ecohydrological sensitivity significantly decreased with increasing 275 effective precipitation (Table 6). In other words, under dry conditions (during dry periods or in dry regions), streamflow is more sensitive to vegetation change than under wet conditions (during wet periods or in wet regions). These findings are in accordance with results from previous studies, which indicate streamflow response to vegetation in drier regions might be more pronounced than in wetter regions Vose et al., 2011;Li et al., 2017;Zhang et al., 2017). For example, Farley et al. (2005) demonstrated that afforestation produced 27% water yield reduction in wetter sites, whilst 62% 280 water yield reduction was identified in drier sites based on the analysis of 26 catchments globally. Sun et al. (2006) modelled streamflow responses to large-scale reforestation in China, and found increased vegetation cover produced a nearly 30% reduction in streamflow in humid regions, but the streamflow reduction rose to approximately 50% in semi-arid and arid areas. Creed et al. (2014) indicated water use efficiencies in forests were higher in drier years than in wetter years by assessing water yield variations in North America. The different ecohydrological sensitivities between dry and wet seasons might be explained by their various mechanisms of water use by vegetation. Vegetation growth in wet conditions with abundant available water, sufficient soil moisture and saturated aquifers is more sensitive to energy factors including temperature, radiation and heat input (Newman et al., 2006;Hou et al., 2018a;Zhang et al., 2011;Brooks et al., 2012). Changes in energy input in wet conditions can alter stomatal conductance and transpiration, and consequently affect the photosynthesis, transpiration and biomass of vegetation (de Sarrau et al., 2012;Van Dover and Lutz, 2004). In contrast, in dry conditions with limited 290 precipitation input, water is more critical for vegetation growth where vegetation mainly relies on its access to soil water through root systems to support photosynthesis and transpiration (Zhou et al., 2015).

Seasonal ecohydrological sensitivity and soil
Soils as the interface between streamflow and groundwater play vital roles in water cycle (Bockheim and Gennadiyev, 2010;Schoonover and Crim, 2015). Our study showed that watersheds with different dominant soil types could have contrasting 295 seasonal ecohydrological sensitivity. As shown in Fig. 4, the ecohydrological sensitivities in both dry and wet seasons in the LIXISOLS-dominated watersheds were the lowest as compared with those of CAMBISOLS-, LEPTOSOLS-and LUVISOLSdominated watersheds. This result clearly illustrates the importance of soil types in hydrological responses and sensitivities (Rieu and Sposito, 1991;Srivastava et al., 2010;Chadli, 2016). Soil properties including organic carbon, salinity, available water holding capacity, saturated hydraulic conductivity and bulk density can affect soil water infiltration and lateral movement 300 (Hillel, 1974;Leu et al., 2010). For example, soil with higher available water holding capacity has the ability to store more water for vegetation growth (Mukundan et al., 2010). Saturated hydraulic conductivity is positively correlated to available water holding capacity, suggesting that soils in a watershed with higher value of saturated hydraulic conductivity might promote interactions between streamflow and groundwater (Sulis et al., 2010). Large differences between topsoil and subsoil bulk densities suggest a frequent moisture movement, leading to more active interactions and feedbacks above and below soil 305 (Zhao et al., 2010). LIXISOLS is characterized by lowest saturated hydraulic conductivity and smallest difference between topsoil and subsoil bulk densities as compared to other three types of soils (Table S1), indicating its lowest water storage capacity and less frequent water movement between topsoil and subsoil. Therefore, hydrological responses in the LIXISOLSdominated watersheds were less sensitive to vegetation change, and consequently led to lowest seasonal ecohydrological sensitivity. 310

Seasonal ecohydrological sensitivity and hydrological regimes
Hydrological regime is another determinant for ecohydrological sensitivity (Zhang et al., 2017). Our study found that the dry season ecohydrological sensitivity in the rain-dominated watersheds was significantly higher than that in the hybrid watersheds ( Fig. 5), while insignificant difference in wet season ecohydrological sensitivity between them (Table 5). The differences in dry season ecohydrological sensitivity between the rain-dominated and hybrid watersheds can be associated with their 315 differences in the mechanisms of streamflow generation. In the rain-dominated watersheds, dry season streamflow is mainly maintained by groundwater discharge while both groundwater and snow water might be the sources of dry season streamflow https://doi.org/10.5194/hess-2020-336 Preprint. Discussion started: 21 July 2020 c Author(s) 2020. CC BY 4.0 License.
in the hybrid watersheds. Thus, the generation of the dry season streamflow in the hybrid watersheds tend to be more complex and stable, and can be more resilient to vegetation change in comparison with that in rain-dominated watersheds. This is supported by several reviews which found that forest cover change in rain-dominated watersheds can produce greater 320 hydrological impacts than in snow-dominated watersheds (Zhang et al., 2017;Moore and Wondzell, 2005). In hybrid watersheds, forestation or vegetation removal can lead to changes in snowmelt processes by altering snow accumulation, melting timing, energy input and wind speed in dry season (Frank et al., 2015), resulting in hydrological de-synchronization effects. These de-synchronization effects may offset negative impacts of vegetation change on dry season streamflow, and eventually lower dry season ecohydrological sensitivity in the hybrid watersheds. 325 The lack of a significant difference in the wet season ecohydrological sensitivity between the rain-dominated and hybrid watersheds might be due to the fact that only precipitation form during wet season is rainfall. It is expected that there are similar interactions and feedback mechanisms between vegetation and water in wet season in all watersheds, leading to insignificant differences in wet season ecohydrological sensitivity between the rain-dominated and hybrid watersheds.

Seasonal ecohydrological sensitivity and topography 330
Topography as an dominating factor for hydrological processes Jenness, 2004;Scown et al., 2015;Yokoyama et al., 2002;Park et al., 2001;Li et al., 2018a) plays an important role in determining streamflow response to vegetation change (Price, 2011;Smakhtin, 2001). According to the established prediction model of dry season ecohydrological sensitivity (Fig. 6a), topographic factors including slope and downslope distance gradient had positive effects on dry season ecohydrological sensitivity, while slope length factor and valley depth yielded negative effects. The vegetated watersheds with 335 steeper slopes often have faster water movement from slopes to river channel and severe soil erosion in wet season if vegetation cover is destroyed, which can greatly reduce wet season soil water storage for supply to dry season streamflow, and therefore have greater dry season ecohydrological sensitivity (Desmet and Govers, 1996;. Similarly, vegetated watersheds with smaller slope length factor and valley depth can also have greater dry season ecohydrological sensitivity. This is possibly because these watersheds generally have a generally flatter topography and longer water residence time, and 340 consequently allow for more interactions between vegetation and water, which likely lead to greater ecohydrological sensitivity in dry season. Unlike the dry season ecohydrological sensitivity, no topographic indices were associated with wet season ecohydrological sensitivity (Fig. 6b). As we know, climate and vegetation are two major drivers to hydrological variations in forested watersheds (Wei et al., 2018;Li et al., 2017). This indicates that in wet season, climate plays a more dominate role in 345 hydrological responses or variations, which means a decreasing role of vegetation on streamflow and consequently reduction of ecohydrological sensitivity. The decreasing role of vegetation on streamfow in wet season may explain the insignificant impact of topographic indices on wet season ecohydrological sensitivity.

Seasonal ecohydrological sensitivity and landscape
Landscape pattern can directly affect hydrological connectivity within a watershed, and can also indirectly influence 350 hydrological processes by controlling soil activities such as soil erosion and sediment (Buma and Livneh, 2017;Teutschbein et al., 2018;Karlsen et al., 2016). Based on the prediction models of seasonal ecohydrological sensitivity, landscape pattern played a more important role in wet season ecohydrological sensitivity than in dry season ecohydrological sensitivity. Only edge density was identified as an effective, negative landscape predictor for wet season ecohydrological sensitivity.
Watersheds with a higher value of edge density are often featured by landscape fragmentation and segmentation, e.g., scatter 355 distributed vegetation, higher road densities, leading to poor hydrological connectivity and high risk of soil erosion. The increasing role of watershed property (edge density) means that the relative role of vegetaion in hydrological reponse would be lower, which consequently lead to decreasing of wet season ecohydrological sensitivity.

Implications
Ecohydrological studies at the seasonal scale are limited due to the lack of the understanding of complex and variable 360 streamflow responses to climate, vegetation, topography, soil and landscape (McDonnell et al., 2018;Singh et al., 2014;Wei et al., 2018;Li et al., 2018a;Oppel and Schumann, 2020;Guswa et al., 2020). Our findings clearly showed that seasonal ecohydrological sensitivity was not only highly associated with climate and vegetation change, but also significantly related to watershed properties like topography, soil and landscape. As indicated by the constructed prediction models, the dry season ecohydrological sensitivity could be better described by vegetation, topography and soil (Fig. 6a) while the wet season 365 hydrological response was mainly controlled by vegetation (leaf area index), soil (topsoil available water holding capacity) and landscape (edge density) (Fig. 6b). Given complex and variable hydrological responses to vegetation change among the study watersheds due to their differences in watershed properties (Zhou et al., 2015;Wei et al., 2018), our seasonal ecohydrological sensitivity prediction model can provide valuable information for understanding of the relative role of climate, vegetation and watershed characteristics of topography, soil and landscape in seasonal ecohydrological processes (Fig. 6). 370 Since many watersheds lack long-term monitoring data on climate, hydrology and vegetation, a quantitative assessment of hydrological response to vegetation change at the watershed scale is very challenging and time-consuming.
However, physical watershed data on climate, vegetation, watershed property can be easily derived from on-line climate datasets, remote sensing-based products, DEMs and soil databases. Development of a seasonal ecohydrological sensitivity prediction model can be an efficient tool for watershed managers to evaluate hydrological impact of vegetation restoration 375 programs with easily accessible data on climate, vegetation, topography, soil and landscape. Once seasonal ecohydrological sensitivity for different watersheds can be predicted quickly, future forest management can be implemented in a more sustainable way. We expect that the assessment framework from this study can be effectively applied to any watersheds where physical watershed data are available to support sustainable watershed planning and management.          (Hargreaves and Samani, 1985). It shows interactions between energy and water and indicates the water availability for vegetation growth.     and water-limited watersheds, respectively; RD is Rain-dominated.