Environ Earth Sci (2017)76:635 DOI 10.1007/s12665-017-6976-z

ORIGINAL ARTICLE

Assessment of land use land cover change impact on hydrological regime of a basin Vaibhav Garg1 • S. P. Aggarwal1 • Prasun K. Gupta2 • Bhaskar R. Nikam1 Praveen K. Thakur1 • S. K. Srivastav3 • A. Senthil Kumar3

•

Received: 24 May 2016 / Accepted: 8 September 2017  Springer-Verlag GmbH Germany 2017

Abstract The sustainability of water resources mainly depends on planning and management of land use; a small change in it may affect water yield largely, as both are linked through relevant hydrological processes, explicitly. However, human activities, especially a significant increase in population, in-migration and accelerated socioeconomic activities, are constantly modifying the land use and land cover (LULC) pattern. The impact of such changes in LULC on the hydrological regime of a basin is of widespread concern and a great challenge to the water resource engineers. While studying these impacts, the issue that prevails is the selection of a hydrological model that may be able to accommodate spatial and temporal dynamics of the basin with higher accuracy. Therefore, in the present study, the capabilities of variable infiltration capacity hydrological model to hydrologically simulate the basin under varying LULC scenarios have been investigated. For the present analysis, the Pennar River Basin, Andhra Pradesh, which falls under a water scarce region in India, has been chosen. The water balance components such as runoff potential, evapotranspiration (ET) and & Vaibhav Garg [email protected] 1

Water Resources Department, Indian Institute of Remote Sensing, Indian Space Research Organisation, Department of Space, Government of India, 4, Kalidas Road, Dehradun, Uttarakhand 248 001, India

2

Geo-informatics Department, Indian Institute of Remote Sensing, Indian Space Research Organisation, Department of Space, Government of India, 4, Kalidas Road, Dehradun, Uttarakhand 248 001, India

3

Indian Institute of Remote Sensing, Indian Space Research Organisation, Department of Space, Government of India, 4, Kalidas Road, Dehradun, Uttarakhand 248 001, India

baseflow of Pennar Basin have been simulated under different LULC scenarios to study the impact of change on hydrological regime of a basin. Majorly, increase in builtup (13.94% approx.) and decrease in deciduous forest cover (2.44%) are the significant changes observed in the basin during the last three decades. It was found that the impact of LULC change on hydrology is balancing out at basin scale (considering the entire basin, while routing the runoff at the basin outlet). Therefore, an analysis on spatial variation in each of the water balance components considered in the study was done at grid scale. It was observed that the impact of LULC is considerable spatially at grid level, and the maximum increase of 265 mm (1985–2005) and the decrease of 48 mm (1985–1995) in runoff generation at grid were estimated. On the contrary, ET component showed the maximum increase of 400 and decrease of 570 mm under different LULC change scenario. Similarly, in the base flow parameter, an increase of 70 mm and the decrease of 100 mm were observed. It was noticed that the upper basin is showing an increasing trend in almost all hydrological components as compared to the lower basin. Based on this basin scale study, it was concluded that change in the land cover alters the hydrology; however, it needs to be studied at finer spatial scale rather than the entire basin as a whole. The information like the spatial variation in hydrological components may be very useful for local authority and decision-makers to plan mitigation strategies accordingly. Keywords Hydrology and water resource  Land use land cover change  Runoff potential  Hydrological modelling  Models (physical)  Variable infiltration capacity model  Impact assessment

123

635

Page 2 of 17

Introduction The present analysis focuses on the impact of LULC change on hydrology of a water scarce river basin. A complex relationship exists between LULC and hydrology involving a large variety of surface–vegetation–atmosphere interactions at varying spatial and temporal scales (DeFries and Eshleman 2004; Dwarakish and Ganasri 2015). LULC has been considered as a driver to almost all hydrological processes such as evapotranspiration (ET), groundwater recharge and overland flow (Wilk and Hughes 2002; Mustard and Fisher 2004; Zhao et al. 2013). According to Chase et al. (2000), the LULC change affects the precipitation and temperature pattern, the fundamental driving forces of the hydrological cycle. Consequently, it alters the entire water balance that exists between evaporation, groundwater recharge and stream discharge of a region or basin (Sahin and Hall 1996; DeFries and Eshleman 2004). It is reported that the LULC change is the main driver in the increase of river discharge worldwide, since 1900 (Piao et al. 2007; Obahoundje et al. 2017). Among different categories of LULC change, it is observed that deforestation alone is the major cause of changes in various hydrological processes, such as ET, stream flow, accumulation and snowmelt processes (Bosch and Hewlett 1982; Gao et al. 2009; Dwarakish and Ganasri 2015; Obahoundje et al. 2017). It is also reported that developing countries, such as India, are likely to witness drastic change in the present century due to inadvertent urbanization (Gosain et al. 2006; Wagner et al. 2013). The consequent change in climate as a result of such changes is thus likely to be more severe in such countries due to poor adaptability (Gosain et al. 2006). Moreover, in the recent times, many of the river basins in the country are facing adverse hydrometeorological conditions, such as floods, droughts and cyclones repetitively (Gosain et al. 2006; Mall et al. 2007; Singh et al. 2010; Gosain et al. 2011). The frequent reoccurrence of such events indicates the shift in the hydrological response of the river basins (Aggarwal et al. 2012). The drivers responsible for changes in land use fall essentially in two categories: biophysical drivers and socioeconomic drivers (Lo and Yang 2002; Kindu et al. 2015). The biophysical drivers include characteristics and processes of the natural environment, such as elevation, slope, soil types and climatic variables, whereas socio-economic drivers mainly include demographic, social, economic, political and technological factors. Many studies have indicated that human actions, mediated by socio-economic drivers, are mainly responsible for land use change in the recent past which in turn has altered the distribution of water resources (Bosch and Hewlett 1982; Stednick 1996; Matheussen et al. 2000; Foley et al. 2004; Gao et al. 2009;

123

Environ Earth Sci (2017)76:635

Petchprayoon et al. 2010; Schilling et al. 2010). To understand the cause-and-effect relation between LULC change and hydrology, the following key questions need to be answered: (1) where LULC changes have taken place in the past; (2) what are the drivers or determinants of land use change, (3) how information on these drivers can be used to project future LULC patterns and (4) what will be the consequences of change in LULC (Lambin et al. 2003). However, the availability of reliable information about the past and present status of these changes in a spatially distributed form and at appropriate scale is one of the major issues in global change studies. In this regard, the capabilities of remote sensing can be utilized which provides synoptic coverage of earth at regular interval. Another problem confronted by water resources engineer is the selection of an appropriate hydrological model which can simulate impact of LULC change on hydrological components. The impact of such changes on hydrology can be best evaluated through a physically based, distributed hydrological model capable of simulating complex hydrological processes, governed by a large number of biophysical variables of the land surface and climatic forcing. Dwarakish and Ganasri (2015) have carried out comprehensive review of hydrological models suitable for the LULC change impact studies. The gridbased, Water Flow and Balance Simulation Model (WaSiM), which is a deterministic spatially distributed hydrological catchment model, has been applied for the purpose at meso-scale catchments (Niehoff et al. 2002; Verbunt et al. 2005; Shrestha et al. 2007; Krause et al. 2007; Merta et al. 2008; Cornelissen et al. 2013; Liu et al. 2013). The most favoured model for such analysis is Soil and Water Assessment Tool (SWAT) hydrological modelling (Schilling et al. 2008; Githui et al. 2009; Nie et al. 2011; Wang and Kalin 2011; Deng et al. 2014; Woldesenbet et al. 2017). SWAT works on hydrological response units (based on LULC, soil and slope) and runs at daily time step. However, the snowmelt runoff calculation of SWAT is temperature based. Hydrologic Engineering Center–Hydrological Modeling System (HEC-HMS) is another hydrological model widely used (Olang and Fu¨rst 2011). However, it has been developed for storm events and works on sub-basin scale. The MIKE-SHE modelling approach has also been attempted in this regard which works on subcatchment scale (Im et al. 2009). The high cost and the computational intensive nature of MIKE-SHE are its limitations. Except MIKE-SHE, none of the abovementioned models are compatible with gridded climatic parameter outputs of general circulation model (GCM). Therefore, one has to simulate two different models to study the impact of climate change and LULC change on hydrology. Moreover, the GCM outputs cannot be handled by micro-scale models, easily. In such a case,

Environ Earth Sci (2017)76:635

efficient downscaling technique has to be performed to produce future climatic parameters at desired scale. In recent past, soil–vegetation–atmosphere transfer schemes (SVATS) have been explicitly developed to represent the land surface partitioning of net radiation into latent, sensible and ground heat fluxes in climate and weather forecast models. ‘‘These models consider the role of vegetation in estimation of evapotranspiration. Furthermore, these models do surface energy balance by reiterating on effective temperatures’’ (Lettenmaier 2001). Moreover, these models generally emphasize on vertical/column processes also, such as extraction of soil moisture by vegetation and transpiration, those involve interaction of vegetation, soil moisture and surface atmosphere and their feedback.

Overview of variable infiltration capacity hydrological model The VIC model, which has also been developed as a SVATS by Liang et al. (1994), has been used for Indian Basin in the present study. It is a physically based, semidistributed hydrological model which quantifies the dominant hydrometeorological process taking place at the land surface atmosphere interface over large river basins to the entire globe (Aggarwal et al. 2013). The distinguishing features of VIC over other land surface models are as follows: •

•

•

•

•

• • •

it considers subgrid variability in land surface vegetation classes and explicitly represents its effects on water and energy budgets; the representation of soil moisture storage capacity as a spatial probability distribution at subgrid level, to estimate surface runoff (Zhao et al. 1980); its parameterization of drainage from the lower soil moisture zone (base flow) as a nonlinear recession (Dumenil and Todini 1992); incorporation of the topography which in turn considers orographic precipitation and temperature lapse rate, resulting in more realistic results for mountainous regions (Gao et al. 2009); the inclusion of both the saturation and infiltration excess runoff processes with a consideration of the subgrid-scale soil heterogeneity at model grid cell (Lohmann et al. 1998a, b; Liang and Xie 2003); the frozen soil processes consideration for cold climate conditions (Cherkauer and Lettenmaier 1999); it solves full surface energy and water balances (Gao et al. 2009); it has separate routing module based on a linear transfer function to simulate the stream flow;

Page 3 of 17

•

•

635

it also incorporates water management effects including reservoir operation, irrigation diversions and return flows (Haddeland et al. 2006a, b; Haddeland et al. 2007); it can be coupled with GCM and other climate models.

VIC model has been validated and adopted at a range of spatial scales, from large river basins to continental and global scales. Within North America, it has been applied at 1˚ spatial resolution to the Missouri (Wood et al. 1997), Arkansas-Red River (Abdulla et al. 1996), Columbia River Basins (Nijssen et al. 1997) and at 0.5 resolution to the Delaware River (Nijssen et al. 1997), Weser River Basin in Germany (Lohmann et al. 1998a, b). The grid network version of the two-layer VIC (VIC-2L) model together with a linear routing scheme was able to estimate daily, monthly and annual stream flow with reasonable accuracy. As a part of the Land Data Assimilation System project, Maurer et al. (2001a) carried out a detailed diagnosis of VIC model results of 50-year simulations over the central USA for model validation and parameterizations. Liang et al. (2004) adopted three-layer VIC (VIC-3L) model to investigate the impacts of spatially distributed precipitation and soil heterogeneity on modelling water fluxes of Blue River Watershed (1233 km2 area), Oklahoma, at different spatial resolutions. Yuan et al. (2004) applied VIC-3L land surface model to simulate stream flow for the Hanjiang River Basin in China. In this case study, the simulated runoff was routed and compared at six different outlets for the daily and monthly observed stream flow at these stations. It was reported that the model can simulate the observations with reasonable accuracy. Aggarwal et al. (2013) implemented the model for entire Indian landmass at 0.25 9 0.25 grid and studied the water balance of major river basins with very high accuracy. In a similar study, Garg et al. (2016) studied the water balance components of Godavari Basin, one of the major river basins in India at 5 9 5 km grid using VIC-3L model. The application of VIC model to simulate impact of LULC change on hydrology is in nascent stage. Matheussen et al. (2000) analysed two land cover scenarios (historical and a current) and its impact on hydrology of the Columbia River Basin at 0.25 spatial grid for the period of 1979–1990. An annual average increase in runoff in the sub-basins ranged from 4.2 to 10.7%, and decreases in ET ranged from 3.1 to 12.1%. The declination in forest cover found to increase stream flow as a result of reduced ET. This study provided a broad-scale framework for assessing the vulnerability of watersheds to altered hydrological regimes attributed to changes in LULC that occur over large geographical areas on long time frames. In a similar study, Dadhwal et al. (2010) carried out hydrological simulation of Mahanadi River Basin to study impact of

123

635

Page 4 of 17

LULC change on surface runoff using VIC model. An increase by 24.44 mm in the annual stream flow was observed at Mundali outlet of the basin from 1972 to 2003. It was concluded that a decrease in forest cover by 5.71% has caused the river flow to increase by 4.53%. Hurkmans et al. (2009) investigated the VIC model to study the effects of land use changes on average stream flow for various locations in the Rhine Basin for projected year 2030 under different scenarios. However, the model was forced with meteorological data of the period 1993–2003. Under each land use change scenario, an increase in stream flow was observed with magnitude varying from about 2% in different sub-basins in the upstream part of the Rhine to about 30% in the Lahn Basin. Due to diverse climatic and topographical features, the availability of water is highly erratic both spatially and temporally across India. The estimated annual precipitation including snowfall which the country receives is 4000 km3. However, the surface runoff potential in the rivers is about 1869 km3 as per the basinwise estimates of Central Water Commission (CWC), India (Garg and Hassan 2007; Mall et al. 2007; Mujumdar 2008). According to the distribution of water resources potential in the country, the national per capita annual availability of water was 1731 m3 as on 2004 (Mujumdar 2008). Per capita annual availability for rest of the country excluding Brahmaputra and Barak Basin works out to about 1345 m3. The Cauvery, Sabarmati, eastflowing rivers and west-flowing river basins fall into waterstressed category, where the availability of water is less than 1000 m3 per capita, which is considered as scarcity condition by international agencies (Gosain et al. 2006; Mujumdar 2008; CWC 2015). Moreover, the developmental activities going on in the country again put stress on future water availability. Considering all these aspects, there is a need to assess LULC change and its impact on hydrological regime of major basins in the country. However, each river basin is characterized by its typical topography, hydrology, climate and anthropogenic influence. In the present study, a detailed analysis of impact of LULC change on hydrological regime of the Pennar Basin has been carried out. Pennar River is one of the major eastflowing rivers and is considered as ‘‘water scarce’’ basin as per UN indicators (CWC 2015; Jain et al. 2009). The basin requires appropriate water management strategy, and the adapted methodology of the present study is described in the subsequent section.

Study area The Pennar River is one of the major east-flowing rivers in India, and it rises in the Thenanahesava hill of the Nandidurg range in Karnataka, drains into the Bay of Bengal

123

Environ Earth Sci (2017)76:635

near Nellore, Andhra Pradesh, and covers a total length of 597 km. The basin encompasses an area of 54,970 km2, and its location is shown in Fig. 1. The fan-shaped Pennar Basin is lying within 13150 –15500 N latitude and 77100 – 80100 E longitude. The major tributaries of the river are Jayamangali, Kunderu, Sagileru, Chitravathi, Papagani and Cheyyeru. The Somashila is the major hydro project in this river basin. There are 07 hydrological observation sites lying in the basin, namely Nagalamadike, Singavaram, Tadipattri, Kamalapuram, Alladupalli, Nandipalli and Nellore (CWC 2015). The elevation in basin varies from \5 to 1250 m. The basin receives rainfall in both the (southwest and northeast) monsoon. However, as the basin lies in a semiarid region, it experiences low rainfall. The rainfall in the basin varies from region to region, and the coastal areas of the basins experience heavier rainfall than western parts. The average annual rainfall of the basin is around 825 mm (CWC 2015). High humidity has been observed during monsoon period, whereas moderate during nonmonsoon. The annual mean temperature of the basin is around 27.17 C, whereas the annual average maximum and minimum temperature of the basin is around 32.71 and 21.63 C, respectively (CWC and NRSC 2014). The major part of the basin is used for agriculture (*53%) and secondly is covered by forest (*20%). Mostly, the basin is composed of fine- to medium-textured clayey soils.

Methodology Assessment of LULC change In order to assess LULC change, initially, the decadal LULC data developed under Indian Space Research Organisation Geosphere Biosphere Programme (ISROGBP): Land Use Land Cover dynamics and impact of Human Dimension in Indian river basins project for the years 1985, 1995 and 2005 at a 1:250,000 scale, were used (as shown in Fig. 3). The existing LULC database of the year 2005 at 1:50,000 scale, prepared under the Natural Resources Census Project of the Natural Resources Repository Programme of the Department of Space, Government of India, is harmonized at 1:250,000 scale following the International Geosphere Biosphere Programme (IGBP) classification scheme adopted by ISROGBP. This harmonized LULC map of 2005 at 1:250,000 scale forms the reference data, based on which changes in LULC units during 1995 and 1985 are mapped using spatially referenced satellite images of the corresponding periods. Firstly, the area of each LULC class for all the years has been identified in ArcGIS 10.1. Then, the area weighted percentage change in each LULC class has been

Environ Earth Sci (2017)76:635

Page 5 of 17

635

Fig. 1 Location of Pennar Basin in India (Source: India-WRIS), Location of gauging station on drainage map and generated 5 9 5 km grid map laid over DEM (DEM Source: USGS)

studied considering 1985 as a base year. Then, the simple subtraction of area has been taken as change of particular LULC class in a particular time period. Later, the decadal change between LULC has been analysed, considering LULC of year 1985 as base year. The details of the LULC change analysis can be found as LULC change matrices and figures in result section. Finally, the daily hydrological simulation of Pennar Basin has been carried out using the three soil-layered VIC (VIC-3L) hydrological models to study the impact of LULC change on its hydrological regime. The Pennar Basin has been transformed to 5 9 5 km scale grid as shown in Fig. 1. Masking of Pennar Basin and grid map was done to identify the active grids to be used as an input for the model. It was found that around 1865 grids are lying within the Pennar Basin boundary and are considered for hydrological simulations. The details of the model requirements and data used are presented in the subsequent section. Hydrological model parameterization As VIC requires land surface characteristics such as soil data, topography and vegetation parameters, the four major input files: soil parameter, vegetation parameter, vegetation library and forcing files were prepared for each grid. The soil parameter file used by VIC describes the unique soil properties for each grid cell in the model domain. In the present analysis, soil texture information and bulk densities

were derived from the National Bureau of Soil Survey and Land Use Planning (NBSS&LUP) digitized soil map as shown in Fig. 2. The soil texture class with highest reoccurrence (on area basis) within a grid has been designated as soil texture class for that particular grid. In this region, generally sand, loam, clay, clay skeletal and loamy skeletal textured soils are present. The file also identifies the grid cells to be simulated based on their latitudes and longitudes and to find the forcing files for the particular grid. The parameters such as porosity, saturated soil potential, saturated hydraulic conductivity and the exponent for each texture class under unsaturated flow condition were calculated for each grid using pedo-transfer functions suggested by Cosby et al. (1984). The parameters used to calibrate the model were: thickness of each soil layer, di; the exponent of the infiltration capacity curve, bi; and the three parameters in the baseflow scheme: Dsmax is the maximum baseflow velocity in mm/day, Ds is the fraction of maximum baseflow velocity, and Ws is the fraction of maximum soil moisture content of the third soil layer at which nonlinear baseflow occurs. The specific soil characteristics (e.g. field capacity, wilting point and saturated hydraulic conductivity) for each grid were obtained from algorithms developed by Cosby et al. (1984), Rawls et al. (1998) and Reynolds et al. (2000). The topographic parameters, namely elevation and slope, were derived from GTOPO30 (90-m resolution) digital elevation model (http://edc.usgs.gov/products/elevation/gtopo30/gtopo30. html) as shown in Fig. 1.

123

635

Page 6 of 17

Environ Earth Sci (2017)76:635

Fig. 2 Soil texture map of Pennar Basin (Data Source: NBSS & LUP)

The vegetation parameter and vegetation library files were prepared for the years 1985, 1995 and 2005 from LULC maps developed under ISRO-GBP Project at a 1:250,000 scale, as discussed earlier. From these data, the land cover types and their fraction of the grid cell occupied by each are identified, as described by Maurer et al. (2001a, b) in the model domain for generation of vegetation parameter file, which uses the same grid cell numbering as the soil file. Rooting depth for each land use class is also specified, which enables shorter crops and grasses draw moisture from the upper soil layers and tree roots from the deeper soil layer. This file cross-indexes each vegetation tile to the classes listed in the vegetation library (http://www.hydro.washington.edu/Lettenmaier/Models/ VIC/). Leaf area index (LAI) is the important characteristic of the land cover that affects the VIC model hydrological simulation. The monthly mean LAI corresponding to each vegetation class has been picked from the Moderate Resolution Imaging Spectroradiometer (MODIS) data product on LAI (http://ladsweb.nascom.nasa.gov/data/search.html) and stored in vegetation library file. The LAI values for a

123

particular LULC class will change seasonally; hence, intraannual variations in vegetation characteristics have been incorporated at monthly timescale using monthly LAI for respective classes. The additional parameters of vegetation library file, namely roughness length and displacement height (in m), architectural resistance (in s/m) and minimum stomatal resistance (in s/m) for each vegetation tile, were assembled based on Global Land Data Assimilation System database (http://ldas.gsfc.nasa.gov/gldas/GLDAS mapveg.php). Initially, the VIC-3L model was forced with daily observed surface meteorological data which include precipitation, minimum temperature, maximum temperature and wind speed for 34 years (1951–1984) at daily time step for calibration against long-term observed average discharge available at Global Runoff Data Centre (http:// www.bafg.de/GRDC/EN/01_GRDC/13_dtbse/database_ node.html). The forcing file of each grid for each year has been prepared using Indian Meteorological Department (IMD) gridded data (Pai 2014) on rainfall (0.25 9 0.25) and temperature (1 9 1).

Environ Earth Sci (2017)76:635

Model calibration As seen in the above section, alike other physically based hydrological models, the VIC model also has many parameters that must be specified. Most of these parameters can be derived from in situ measurement and remote sensing observation. Six of them are considered as calibration parameters, namely bi, d2, d3, Dsmax, Ds and Ws. bi describes the amount of available infiltration capacity as a function of relative saturated grid cell area and hence controls the partitioning of precipitation into infiltration and direct runoff as it. The high value of bi gives low infiltration and yields more surface runoff. The d2 and d3 are the soil layer thicknesses for second and third layers that control the availability of water for transpiration and baseflow, respectively. The thicker soil depths will dominate evapotranspiration, baseflow and lower down runoff response. On the other hand, it will retain soil moisture for long and results in higher baseflow in wet seasons. As in Pennar Basin, mostly deep soils are present, the d2 and d3 were kept constant as 6 and 0.5 m. However, the baseflow parameters, namely Dsmax, Ds and Ws, govern the water stored in the lowest soil layer and its release as baseflow (Liang et al. 1994). The higher the value of Ds, the higher the baseflow at lower water content in the lowest soil layer. A higher value of Ws will raise the water content required for baseflow and delay runoff peaks. Later, the calibrated model was forced with the same meteorological parameters for period (1980–2010) with vegetation parameter files of each year, i.e. 1985, 1995 and 2005, under consideration, separately. The meteorological forcing for period (1980–2010) was kept constant for entire simulations; however, only vegetation parameter file of years under consideration (i.e. 1985, 1995 and 2005) was changed in each simulation to study impact of LULC change on the water balance components.

Results and discussion Assessment of LULC change To study the LULC change, the LULC database developed under ISRO-GBP LULC Project for all the years 1985, 1995 and 2005 has been procured as shown in Fig. 3. Then, the decadal change in LULC has been analysed, considering LULC of year 1985 as base year as shown in LULC change matrices for the period 1985–1995 and 1985–2005 which are reported in Table 1 There is very less change observed in LULC between 1985 and 1995. However, a considerable change has been noticed between 1985 and 2005 (refer Fig. 4), and this may be attributed to developmental activities in the basin area in

Page 7 of 17

635

the last twenty-year time span. It was observed that there is a maximum increase in built-up class (13.94% approx.) and corresponding maximum decline in deciduous forest by 2.44% between years 1985 and 2005. Certainly, this change in LULC pattern will have an effect on hydrology of the basin. Impact of LULC change on hydrology To study the influence of LULC on hydrological regime of Pennar Basin, VIC-3L hydrological model has been utilized. Initially, the model was forced with daily meteorological parameters, namely rainfall, maximum temperature, minimum temperature and wind speed for period 1951–1984 (34 years). The NBSS&LUP soil texture data and LULC of year 1985 were used to derive soil parameter file and vegetation parameter files, respectively, as described in the previous section. The runoff estimated was routed at the basin outlet near Nellore, Andhra Pradesh, India (shown in Fig. 1) using VIC’s routing model as described in detail by Lohmann et al. (1996, 1998a). The calibration parameters as discussed in methodology section were adjusted through a trial-and-error procedure until acceptable match between estimated and observed discharge was found. The usual range and calibrated values of these parameters are provided in Table 2. As the long-term (1965–79) monthly average discharge data were available at GRDC, the estimated discharge between these periods was averaged in order to validate the model. The comparison of mean monthly estimated and observed discharge data is shown in Fig. 5. Later, the calibrated VIC-3L hydrological model has been investigated to study the impact of LULC change on hydrological regime of Pennar Basin. In this analysis, in order to study impact of LULC change, the meteorological forcing which was comprised of daily precipitation, daily minimum temperature, daily maximum temperature and daily wind speed (for period 1980–2010) has to be kept uniform for the years under consideration, i.e. 1985, 1995 and 2005. Only the vegetation parameter file corresponding to the particular year, i.e. 1985/1995/2005 was changed for hydrological simulation using VIC-3L. The simulation has been carried out for period 1980–2010, so that the model gets stabilized with good amount of warming period. The model generates separate flux files for each grid as an output which is comprised of precipitation, surface runoff, evaporation from canopy, evaporation from bare soil, transpiration from canopy, baseflow and soil moisture for each soil layer. The gridwise results of each simulation year, i.e. 1985, 1995 and 2005, were averaged to find out basin average hydrological components for the basin. The study indicates that all the water balance components are varying in accordance with general change in LULC; the

123

635

Page 8 of 17

Environ Earth Sci (2017)76:635

Fig. 3 LULC map of the basin for years 1985, 1995 and 2005 (Data Source: ISRO-GBP) Table 1 LULC change (km2) matrix for the period 1985–1995 and 1985–2005 1995 BU 1985

Total CL

FL

PL

DBF

MF

SL

WB

BL

WL

BU

322.50

0.88

0.00

0.00

0.00

0.00

0.00

0.00

0.00

0.00

323.38

CL

12.06

28791.50

614.69

3.31

0.00

8.38

69.88

64.56

2.88

0.63

29567.88

FL

10.88

615.25

3337.19

0.31

0.00

2.13

2.94

77.88

27.25

0.31

4074.13

PL

0.00

8.31

1.88

338.13

0.00

0.00

5.69

0.25

0.00

0.00

354.25 6982.81

DBF

0.00

0.81

0.00

0.00

6550.63

419.56

8.88

0.00

2.94

0.00

MF

3.38

7.56

1.69

0.00

51.38

4864.50

32.75

0.00

24.75

0.06

4986.06

SL

0.31

66.81

38.13

0.00

25.69

25.50

5404.38

12.69

19.00

0.00

5592.50

WB

1.38

16.38

5.38

0.06

0.00

1.38

1.19

2121.94

0.00

0.00

2147.69

BL

0.00

4.50

0.00

0.00

0.00

0.44

123.38

0.00

679.75

1.44

809.50 139.69

WL Total

0.00

0.06

0.00

0.00

0.00

0.00

0.00

0.00

0.00

139.63

350.50

29512.06

3998.94

341.81

6627.69

5321.88

5649.06

2277.31

756.56

142.06

2005 BU 1985

Total CL

FL

PL

DBF

MF

SL

WB

BL

WL

BU

322.44

0.38

0.19

0.00

0.00

0.13

0.19

0.06

0.00

0.00

323.38

CL

25.38

28328.44

610.19

15.75

33.50

66.50

161.38

279.75

41.94

5.06

29567.88 4074.13

FL

11.31

293.00

3609.19

8.00

8.31

4.13

26.19

83.06

29.38

1.56

PL

0.00

3.00

7.75

340.88

0.00

0.44

0.25

0.88

0.50

0.56

354.25

DBF

0.00

25.69

15.31

13.88

6575.06

290.31

33.94

26.44

2.19

0.00

6982.81

MF SL

3.38 1.19

49.63 118.69

27.06 41.81

5.31 3.81

165.63 16.50

4622.94 39.06

88.31 5309.25

2.38 35.19

21.38 26.50

0.06 0.50

4986.06 5592.50 2147.69

WB

3.44

202.94

40.88

5.00

13.50

21.75

35.81

1814.75

5.25

4.38

BL

1.31

6.63

23.25

0.00

0.00

1.88

16.13

1.00

757.88

1.44

809.50

WL

0.00

3.94

0.94

0.00

0.00

1.00

1.06

0.00

0.06

132.69

139.69

368.44

29032.31

4376.56

392.63

6812.50

5048.13

5672.50

2243.50

885.06

146.25

Total

The diagonal values are shown in bold, indicating the area which has remained unchanged between two time periods for respective LULC class BU built-up, CL crop land, FL fallow land, PL plantation, DBF deciduous broad leaf forest, MF mixed forest, SL shrub land, WB water body, BL barren land, WL waste land

value of estimated runoff potential is 210.21 mm in year 1985, 209.87 mm in 1995 and 201.42 mm in 2005. Similarly, the components, i.e. ET and baseflow, were also analysed as given in Fig. 6.

123

A very little variation in each of the water balance component estimates was observed. This null effect of considerable LULC change on hydrology of the basin was further analysed. The reason can be attributed to the fact

Environ Earth Sci (2017)76:635

Page 9 of 17

635

Fig. 4 LULC change map of the basin, location of major changes in LULC highlighted in red boxes

Table 2 VIC calibration parameters, their range and calibrated value Parameter

Value range

Calibrated value

bi

0–0.4

0.2

Ds

0–1

0.1

Dsmax

0–30

It has been assigned to each grid based on the soil texture class, saturated hydraulic conductivity and slope

Ws

0–1

0.9

that a very large basin has been simulated to study impact of LULC change. The spatial LULC change map, as shown in Fig. 4, was further analysed with respect to impact of particular change on hydrology, in general. The spatial area calculation has been done with regard to LULC change which might induce reduction or increase in runoff, as shown in Fig. 7. It was noticed that the area under LULC change that induces increase in runoff is around 1150 km2, whereas the area under LULC changes that induce reduction in runoff is about 1087 km2. It is evident from this

123

635

Page 10 of 17

Environ Earth Sci (2017)76:635

Fig. 5 Comparison of long-term mean observed and estimated discharge at basin outlet near Nellore, Andhra Pradesh Fig. 6 Plot of estimated basin average water balance components: runoff potential, baseflow and ET against rainfall

analysis that the area forcing increase and decrease in runoff generation is more or less similar. Therefore, the changes in LULC, which might result in generation of more runoff and changes which further may reduce the runoff production, get nullified over this large basin. It is to be noted that the same will be the case for other hydrological components, as the hydrological components depend on each other. For example, if built-up increases, runoff increases, whereas ET and baseflow decreases. On the contrary, if vegetation (forest/crop land) increases, ET and baseflow may increase with the decrease in runoff. However, the model used in the present study solves water

123

balance for each grid; the change in any hydrological parameter will have effect on the other. This study provided an impetus to study spatial variation in these water balance components with regard to LULC change. In this part of analysis, initially, a climatic normal of each water balance component, i.e. runoff, ET and baseflow, was derived by taking long-term (1951–1984) 34-year spatial mean of these components. The climatic normal of each water balance component is provided in Fig. 8. It was observed that the variation of runoff is from 0 to 364.3 mm, ET varies from 0 to 945 mm, and baseflow varies from 0 to 186 mm.

Environ Earth Sci (2017)76:635

Page 11 of 17

635

Fig. 7 Impact of LULC change on runoff generation

Later, each of these parameters under each LULC change scenario (1985/1995/2005) was analysed for meteorological forcing of year 1995, as the annual basin average rainfall in year 1995 is very close to long-term basin average rainfall of 825 mm. The difference or anomaly maps between long-term annual average and the annual sum of the particular year (with regard to forcing of year 1995) of each parameter were analysed. The difference maps of each parameter for each year under consideration are shown in Fig. 9. These maps depict that large spatial region is subjected to hydrological changes due to LULC change. The changes (increase or decrease) in each component have been observed for all the years under consideration

with regard to climatic normal. However, it was noticed that there is a very nominal change in hydrology of the basin between years 1985 and 1995 as there is very less variation in the LULC during this period. The runoff potential in some parts of the basin has increased maximum by around 48 mm and decreased by maximum around 173 mm during 1985 and 1995 with respect to climatic normal runoff. The effect of LULC change on runoff is predominant in the year 2005, where runoff potential has increased by 88 mm and decreased by 265 mm. However, the ET component has increased maximum by around 400 mm and decreased by 570 mm under each LULC scenario. Similarly, the baseflow is showing variation of 70-mm increase and 100-mm decrease throughout the

Fig. 8 Long-term climatic normal of basin runoff, ET and baseflow

123

635

Page 12 of 17

Environ Earth Sci (2017)76:635

Fig. 9 Spatial variation of basin average water balance components and their anomaly for respective years a basin annual runoff and runoff anomaly for each year under consideration, b basin annual ET

and ET anomaly for each year under consideration, c basin annual baseflow and baseflow anomaly for each year under consideration

analysis. The upper Pennar Basin is showing increasing trend in all the water balance components; however, the lower basin is showing decrease in these components. The grids with maximum increase and decrease in water balance components were analysed further for better

understanding. The location of these selected grids was marked on LULC change map (refer Fig. 4), and the corresponding change in LULC of these grids was analysed as presented in Table 3.

123

Environ Earth Sci (2017)76:635

Page 13 of 17

635

Fig. 9 continued

In the grids 2164, 469 and 1477; where, the increase in built-up and subsequent decrease in cropland/forest was observed, the general trend of runoff increase was noticed. This may be attributed due to slight increase in impervious area and reduction in vegetative surface. The baseflow showed an increasing trend as with the increase in built-up and subsequent decrease in vegetative surface, the transpiration through plants’ root reduced. However, the grid 1477 in year 1995 showed a slight decrease in runoff. It might be due to the reason that more area covered under water body in this particular grid, which may be retaining water to flow out. The decrease in runoff and increase in ET was observed, in the grids (Grid No. 1678, the year 2005), where, cropland/forest has decreased and water body area has increased. However, in such cases, generally, ET must reduce, but in the present analysis, slight increase in ET was observed. The reason could be the prevalence of large number of shallow fish ponds and storage tanks constructed in this coastal region of India, contributing more to evaporation. Similarly, baseflow was also showing slight increase, as the increase in water body would keep its surrounding saturated, so the lowest soil layer would remain saturated and contribute more to baseflow. In the grid 723, a large increase in cropland has been observed, which forced to reduce runoff generation, whereas the grid 1428 was showing shifting of deciduous forest to mixed forest, which showed general trend of increase in runoff. Similarly, the grid 1292 was also showing increase in runoff with the change of cropland to

fallow land. This might have attributed due to decrease in LAI and reduced transpiration from less vegetated surface. Based on these observations, it can be concluded that slight change in LULC can change hydrological regime of a basin entirely.

Summary and conclusions In the present study, the Pennar Basin has been simulated using VIC hydrological model for assessing the impact of LULC change on its hydrological regime successfully. Initially, the change in LULC between years 1985, 1995 and 2005 has been assessed. It was observed that almost all LULC classes show significant changes during these periods. Maximum increase in built-up LULC class has been observed, and maximum decline in deciduous forest class was observed. Later, the impact of these changes in LULC on water balance components of Pennar Basin was studied. In the present analysis, the VIC soil–vegetation–atmosphere transfer scheme has been investigated for Indian condition. Firstly, the model was simulated for meteorological forcing of period 1951–1984. The simulated model was then calibrated against the available average monthly discharge data for period 1965–1979 at GRDC site of Nellore outlet. The well-calibrated model was later utilized to study impact of LULC on hydrology of Pennar Basin. In order to study LULC change impact, the meteorological forcing (period 1980–2010) was kept constant for all three

123

635

Page 14 of 17

Environ Earth Sci (2017)76:635

Table 3 Change in LULC and corresponding change in water balance components Location

Fraction LULC Class Area

Runoff (mm)

1985

1995

2005

Grid no: 1678

CL: 0.53

CL: 0.53

CL: 0.0975

Lat: 14.525

SL: 0.355

SL: 0.355

SL: 0.355

Long: 79.075

WB: 0.115

WB: 0.115

WB: 0.5475

Grid no: 723

CL: 0.3575

CL: 0.975

CL: 0.975

Lat: 15.275

FL: 0.635

WB: 0.025

WB: 0.025

Long: 78.575

WB: 0.0075

Grid no: 1428 Lat: 79.175

CL: 0.0275 DBF: 0.955

CL: 0.0275 DBF: 0.545

CL: 0.0275 DBF: 0.545

Long: 14.725

SL: 0.0125

MF: 0.41

MF: 0.4075

WB: 0.005

SL: 0.0125

SL: 0.015

WB: 0.005

WB: 0.005

Grid no: 2164

CL: 0.805

BU: 0.0525

BU: 0.0925

Lat: 14.125

DBF: 0.0175

CL: 0.7525

CL: 0.7125

Long: 78.175

BL: 0.1775

MF: 0.0175

MF: 0.0175

SL: 0.075

BL: 0.1775

1995

ET (mm)

2005

1995

Baseflow (mm) 2005

1995

2005

0

-56.72

0

205.15

0

28.41

-1.26

-1.26

-5.54

-5.54

-0.04

0.38

1.91

1.67

22.47

23.09

0

0.59

0.21

10.31

-0.43

-30.16

0.22

0.25

-1.47

-1.64

0.5

2.02

-0.59

0.079

0

-0.212

31.58

27.12

BL: 0.1025 Grid no: 469

BU: 0.29

BU: 0.405

BU: 0.4175

Lat: 15.475

CL: 0.6875

CL: 0.5725

CL: 0.56

Long: 78.475

WB: 0.0225

WB: 0.0225

WB: 0.0225

Grid no: 1477

BU: 0.05

BU: 0.1525

BU: 0.17

Lat: 14.675

CL: 0.8375

CL: 0.7

CL: 0.505

Long: 78.475

FL: 0.0325 MF: 0.015

FL: 0.0525 MF: 0.0075

FL: 0.26 MF: 0.0075

WB: 0.065

WB: 0.0875

WB: 0.0575

Grid no: 1292

CL: 0.875

CL: 0.875

CL: 0.2125

Lat: 14.825

PL: 0.0225

PL: 0.225

FL: 0.67

Long: 78.675

SL: 0.0425

SL: 0.0425

PL: 0.0225

WB: 0.0375

WB: 0.0375

SL: 0.0425

BL: 0.0225

BL: 0.0225

WB: 0.03

-1.82

0

0

-0.038

0

1.103

-1.478

0.09

BL: 0.0225

simulations with vegetation parameters correspond to LULC of years 1985, 1995 and 2005. It was noticed that when water balance components were lumped over this large basin, a marginal change in hydrology of the basin was observed. Further, to verify the same, the LULC change map was analysed for changes that would induce runoff increase and changes those would reduce the same. It was observed that the area under each of these categories was almost the same, which would have hugely minimized the effect of LULC change on this large basin water balance components at basin scale to a negligible stage. Further attempts were made to analyse results with respect to impact of LULC change on spatial scale. The model was run for meteorological forcing of period 1951–1984 (34 yrs), and the climatic normal of each

123

parameter, i.e. runoff potential, ET and baseflow, was generated by averaging all the grids for this period. Later, the annual total of basin average of each of these parameters from the analysis of meteorological forcing (period 1980–2010) for year 1995 only was generated for each year under consideration, i.e. 1985, 1995 and 2005. Then, difference or anomaly maps were generated by subtracting average annual of each parameter from respective longterm climatic normal. It was observed that large area of the basin was unchanged with regard to hydrology. However, wherever, the changes in LULC over the period 1985–2005 were significant and the considerable change in basin’s water balance was observed. The grids where forest cover or crop land has declined, and the runoff has increased significantly. However, the regions of urban sprawl were

Environ Earth Sci (2017)76:635

showing general trend of increase in runoff potential and subsequent decrease in ET. The runoff has increased because impervious area has increased with the decrease in vegetative surface. The decline in vegetated surface further reduces transpiration, which has been depicted by decrease in overall ET. In this coastal region of India, large fish ponds and tanks were constructed; those are retaining water and contributing more to evaporation. The retention of water made soil to remain saturated; hence, increase in baseflow was observed in such cases. Therefore, it may be concluded that slight change in LULC influences the hydrological regime of river basin in larger way, but needs to be studied at finer spatial scale. Such studies may prove to be very important for decision-makers who require such information to evaluate mitigation and adaptation strategies under changing LULC scenarios. Further, it can support policy makers to allocate water judiciously among each sector: agriculture, ecosystems, domestic and industry. In the present study, the sensitivity of the Pennar hydrological system with respect to change in land cover has been studied over long time (decadal) frames. Moreover, climatic factors were kept constant to study the impact of LULC change over a large analysis period, this does not reflect reality, as simultaneously, climate might have also changed during the period of consideration. The nonconsideration of dynamics (in both land surface and climate) between such a long time duration may be one of the limitations of the present analysis. It is also to be noted that each region will behave differently with respect to LULC and climate change. Therefore, each of the regions needs to be studied separately, taking into account plausible LULC within a climate change context projecting the behaviour of the region in the future research. Acknowledgements The authors would like to thank the authorities of Indian Space Research Organisation for providing financial grant for this research work. The work has been done under the ISRO-GBP Project on ‘‘Land Use Land Cover dynamics and impact of Human Dimension in Indian river basins’’. The authors thank IMD for providing daily gridded data on rainfall and temperature and the VIC hydrological model team for their help during the course of the study. Thanks are due to LULC team for their efforts in generating LULC map for entire India and for providing the data for this particular basin. Authors would like to extend their gratitude to Ms. Asfa Siddiqui for reviewing the manuscript and correcting it for the language.

References Abdulla FA, Lettenmaier DP, Wood EF et al (1996) Application of a macroscale hydrologic model to estimate the water balance of the Arkansas-Red river basin. J Geophys Res 101(D3):7449–7459 Aggarwal SP, Garg V, Gupta PK et al (2012) Climate and LULC change scenarios to study its impact on hydrological regime. Int Archives of the Photogramm Remote Sens Spatial Inf Sci

Page 15 of 17

635

(ISPRS), XXXIX-B8, 147–152. doi:10.5194/isprsarchivesXXXIX-B8-147-2012 Aggarwal SP, Garg V, Gupta PK et al (2013) Runoff potential assessment over Indian landmass: a macro-scale hydrological modeling approach. Curr Sci 104(7):950–959 Bosch JM, Hewlett JD (1982) A review of catchment experiments to determine the effect of vegetation changes on water yield and evapotranspiration. Jour Hydrol 55:3–23 Chase TN, Pielke RA Sr, Kittel TGF et al (2000) Simulated impacts of historical land cover changes on global climate in northern winter. Clim Dyn 16:93–105 Cherkauer KA, Lettenmaier DP (1999) Hydrologic effects of frozen soils in the upper Mississippi River basin. J Geophys Res 104(D16):19599–19610 Cornelissen T, Diekkru¨ger B, Giertz S (2013) A comparison of hydrological models for assessing the impact of land use and climate change on discharge in a tropical catchment. J Hydrol 498:221–236. doi:10.1016/j.jhydrol.2013.06.016 Cosby BJ, Hornberger GM, Clapp RB et al (1984) A statistical exploration of the relationships of soil moisture characteristics to the physical properties of soils. Water Resour Res 20(6):682–690 CWC (2015) Integrated hydrological data book (non-classified river basins). Hydrological data directorate, information system organisation, water planning, projects wing, central water commission, ministry of water resources, river development & Ganga Rejuvenation, Government of India, New Delhi CWC, NRSC (2014) Pennar basin report version 2.0. Ministry of Water Resources, Government of India Dadhwal VK, Aggarwal SP, Misra N (2010) Hydrological simulation of Mahanadi River Basin and impact of landuse/landcover change on surface runoff using a macro scale hydrological model. In: W Wagner, B Szekely (eds.) Proceedings of ISPRS TC VII symposium–100 years ISPRS. Vienna, Austria, July 5–7, ISPRS, XXXVIII (7B), pp. 165–170 DeFries R, Eshleman KN (2004) Land-use change and hydrologic processes: a major focus for the future. Hydrol Process 18:2183–2186 Deng Z, Zhang X, Li D, Pan G (2014) Simulation of land use/land cover change and its effects on the hydrological characteristics of the upper reaches of the Hanjiang Basin. Environ Earth Sci 73(3):1119–1132 Dumenil L, Todini E (1992) A rainfall-runoff scheme for use in the Hamburg climate model. In: P O’Kane (ed.) Advances in theoretical hydrology, a tribute to James Dooge. European geophysical society series on hydrological sciences 1, pp. 129–157 Dwarakish GS, Ganasri BP (2015) Impact of land use change on hydrological systems: a review of current modeling approaches. Cogent Geosci 1(1):1115691. doi:10.1080/23312041.2015. 1115691 Foley JA, Kucharik CJ, Twine TE et al (2004) Land use, land cover, and climate change across the Mississippi Basin: impacts on selected land and water resources. In: RS Defries, GP Asner, RA Houghton (Eds.) Ecosystems and land use change, American Geophysical Union, Washington, DC. doi: 10.1029/153GM19 Gao Z, Xhang Z, Zhang X (2009) Responses of water yield to changes in vegetation at a temporal scale. Front Forestry China 4(1):53–59 Garg NK, Hassan Q (2007) Alarming scarcity of water in India. Curr Sci 93(7):932–941 Garg V, Dhumal IR, Nikam BR et al (2016) Water resources assessment of Godavari River Basin, India. In: Proceedings of 37th Asian conference on remote sensing (ACRS-2016) on promoting spatial data infrastructure for sustainable economic development, Oct 17–22, 2016, Colombo, Sri Lanka

123

635

Page 16 of 17

Githui F, Mutua F, Bauwens W (2009) Estimating the impacts of land-cover change on runoff using the soil and water assessment tool (SWAT): case study of Nzoia catchment. Kenya Hydrol Sci J 54(5):899–908 Gosain AK, Rao S, Basuray D (2006) Climate change impact assessment on hydrology of Indian river basins. Curr Sci 90(3):346–353 Gosain AK, Rao S, Arora A (2011) Climate change impact assessment of water resources of India. Curr Sci 101(3):356–371 Haddeland I, Lettenmaier DP, Skaugen T (2006a) Effects of irrigation on the water and energy balances of the Colorado and Mekong river basins. J Hydrol 324(1–4):210–223 Haddeland I, Skaugen T, Lettenmaier DP (2006b) Anthropogenic impacts on continental surface water fluxes. Geophys Res Lett 33:L08406. doi:10.1029/2006GL026047 Haddeland I, Skaugen T, Lettenmaier DP (2007) Hydrologic effects of land and water management in North America and Asia: 1700-1992. Hydrol Earth Syst Sci 11(2):1035–1045 Hurkmans RTWL, Terink W, Uijlenhoet R et al (2009) Effects of land use changes on streamflow generation in the Rhine basin. Water Resour Res 45:W06405. doi:10.1029/2008WR007574 Im S, Kim H, Kim C, Jang C (2009) Assessing the impacts of land use changes on watershed hydrology using MIKE SHE. Environ Geol 57:231–239 Jain AK, Rao BMM, Rao MSR, Swamy MV (2009) Groundwater scenario in Andhra Pradesh. Centre for economics and social studies, WASHcost report, Hyderabad, India. (Available online: http://www.cess.ac.in/cesshome/wp/WP-3-Groundwater%20Sce nario%20in%20AP.PDF, Accessed on Jan 18 2016) Kindu M, Schneider T, Teketay D, Knoke T (2015) Drivers of land use/land cover changes in Munessa-Shashemene landscape of the south-central highlands of Ethiopia. Environ Monit Assess 187(7):452. doi:10.1007/s10661-015-4671-7 Krause S, Jacobs J, Bronstert A (2007) Modelling the impacts of landuse and drainage density on the water balance of a lowland– floodplain landscape in northeast Germany. Ecol Modelling 200(3–4):475–492 Lambin EF, Geist HJ, Lepers E (2003) Dynamics of land-use and land-cover change in tropical regions. Annu Rev Environ Resour 28:205–241 Lettenmaier DP (2001) Macroscale Hydrology: Challenges and Opportunities. In: Matsuno T, Kida H (eds) Present and future of modeling global environmental change: toward integrated modeling. TERRAPUB, Tokyo, pp 111–136 Liang X, Xie ZH (2003) Important factors in land-atmosphere interactions: surface runoff generations and interactions between surface and groundwater. Global Planet Change 38(1–2):101–114 Liang X, Lattenmaier DP, Wood EF et al (1994) A simple hydrologically based model of land surface, water, and energy flux for general circulation models. J Geophys Res 99(D7):14415–14428 Liang X, Guo J, Leung LR (2004) Assessment of the effects of spatial resolutions on daily water flux simulations. J Hydrol 298(1–4):287–310 Liu T, Fang H, Willems P et al (2013) On the relationship between historical land-use change and water availability: the case of the lower Tarim River region in northwestern China. Hydrol Process 27(2):251–261 Lo CP, Yang X (2002) Drivers of land-use/land-cover changes and dynamic modeling for the Atlanta, Georgia Metropolitan area. Photogramm Eng Remote Sens 68(10):1073–1082 Lohmann D, Nolte-Holube R, Raschke E (1996) A large scale horizontal routing model to be coupled to land surface parameterization schemes. Tellus 48A:708–721 Lohmann D, Raschke E, Nijssen B et al (1998a) Regional scale hydrology: I. formulation of the VIC-2L model coupled to a routing model. Hydrol Sci J 43(1):131–141

123

Environ Earth Sci (2017)76:635 Lohmann D, Raschke E, Nijssen B et al (1998b) Regional scale hydrology: II. application of the VIC-2L model to the Weser river, Germany. Hydrol Sci J 43(1):143–158 Mall RK, Bhatla R, Pandey SN (2007) Water resources in India and impact of climate change. Jalvigyan Sameeksha 22:157–176 Matheussen B, Kirschbaum RL, Goodman IA et al (2000) Effects of land cover change on streamflow in interior Columbia River Basin (USA and Canada). Hydrol Process 14:867–885 Maurer EP, O’Donnell GM, Lettenmaier DP et al (2001a) Evaluation of the land surface water budget in NCEP/NCAR and NCEP/ DOE reanalyses using an off-line hydrologic model. J Geophys Res 106(D16):17841–17862 Maurer EP, O’Donnell GM, Lettenmaier DP et al (2001b) Evaluation of NCEP/NCAR reanalysis water and energy budgets using macroscale hydrologic model simulations. In: V Lakshmi, J Albertson, J Schaake (eds) Land surface hydrology, meteorology, and climate: observations and modeling. Water Sci. Appl. 3, AGU, Washington, pp. 137–158. doi:10.1029/WS003p0137 Merta M, Seidler C, Bianchin S et al. (2008) Analysis of land use change in the Eastern Ore Mts. regarding both nature protection and flood prevention. Soil and Water Res 3:S105–S115 (Special Issue 1) Mujumdar PP (2008) Implications of climate change for sustainable water resources management in India. Phys Chem Earth 33:354–358 Mustard J, Fisher T (2004) Land Use and Hydrology. In: Gutman G et al (eds) Land change science: observing monitoring and understanding trajectories of change on the earth’s surface. Kluwer Academic Publishers, Dordrecht, pp 257–276 Nie W, Yuan Y, Kepner W et al (2011) Assessing impacts of landuse and landcover changes on hydrology for the upper San Pedro watershed. J Hydrol 407(1–4):105–114 Niehoff D, Fritsch U, Bronstert A (2002) Land-use impacts on stormrunoff generation: scenarios of land-use change and simulation of hydrological response in a macro-scale catchment in SWGermany. J Hydrol 267:80–93 Nijssen BN, Lettenmaier DP, Liang X et al (1997) Streamflow simulation for continental-scale river basins. Water Resour Res 33(4):711–724 Obahoundje S, Ofosu EA, Akpoti K, Kabo-bah AT (2017) Land use and land cover changes under climate uncertainty: modelling the impacts on hydropower production in Western Africa. Hydrology 4(1):2. doi:10.3390/hydrology4010002 Olang LO, Fu¨rst J (2011) Effects of land cover change on flood peak discharges and runoff volumes: model estimates for the Nyando River Basin. Kenya Hydrol Process 25:80–89 Pai DS, Sridhar Latha, Rajeevan M et al (2014) Development of a new high spatial resolution (0.25 9 0.25) Long period (1901–2010) daily gridded rainfall data set over India and its comparison with existing data sets over the region. Mausam 65(1):1–18 Petchprayoon P, Blanken PD, Ekkawatpanit C et al (2010) Hydrological impacts of land use/land cover change in a large river basin in central–northern Thailand. Int J Climat 30(13):1917–1930. doi:10.1002/joc.2131 Piao S, Friedlingstein P, Ciais P et al (2007) Changes in climate and land use have a larger direct impact than rising CO2 on global river runoff trends. PNAS 104(39):15242–15247 Rawls WJ, Gimenez D, Grossman R (1998) Use of soil texture, bulk density, and slope of the water retention curve to predict saturated hydraulic conductivity. Trans ASABE 41(4):983–988 Reynolds CA, Jackson TJ, Rawls WJ (2000) Estimating soil waterholding capacities by linking the food and agriculture organization soil map of the world with global pedon databases and continuous pedotransfer functions. Water Resour Res 36(12):3653–3662 Sahin V, Hall MJ (1996) The effects of afforestation and deforestation on water yields. J Hydrol 178:293–309

Environ Earth Sci (2017)76:635 Schilling KE, Jha MK, Zhang Y-K et al (2008) Impact of land use and land cover change on the water balance of a large agricultural watershed: historical effects and future directions. Water Resour Res, 44:W00A09. doi:10.1029/2007WR006644 Schilling KE, Chan KS, Liu H et al (2010) Quantifying the effect of land use land cover change on increasing discharge in the Upper Mississippi River. J Hydrol 387(3–4):343–345 Shrestha RR, Bardossy A, Rode M (2007) A hybrid deterministicfuzzy rule based model for catchment scale nitrate dynamics. J Hydrol 342:143–156 Singh S, Srivastav SK, Srivastava VK et al (2010) Human dimensions of climate change: a few initiatives. Bulletin of national natural resources management system. NNRMS (B) 35:118–126 Stednick JD (1996) Monitoring the effects of timber harvest on annual water yield. Jour Hydrol 176:79–95 Verbunt M, Groot Zwaaftink M, Gurtz J (2005) The hydrologic impact of land cover changes and hydropower stations in the Alpine Rhine basin. Ecol Modelling 187(1):71–84 Wagner PD, Kumar S, Schneider K (2013) An assessment of land use change impacts on the water resources of the Mula and Mutha Rivers catchment upstream of Pune. India Hydrol Earth Syst Sci 17:2233–2246 Wang R, Kalin L (2011) Modelling effects of land use/cover changes under limited data. Ecohydrol 4:265–276. doi:10.1002/eco.174

Page 17 of 17

635

Wilk J, Hughes DA (2002) Simulating the impacts of land-use and climate change on water resource availability for a large south Indian catchment. Hydrol Sci J 47(1):19–30 Woldesenbet TA, Elagib NA, Ribbe L, Heinrich J (2017) Hydrological responses to land use/cover changes in the source region of the Upper Blue Nile Basin, Ethiopia. Sci Total Environ 575:724–741. doi:10.1016/j.scitotenv.2016.09.124 Wood EF, Lettenmaier D, Liang X et al (1997) Hydrological modeling of continental-scale basins. Annu Rev Earth Planet Sci 25:279–300 Yuan F, Xie Z, Liu Q et al (2004) An application of the VIC-3L land surface model and remote sensing data in simulating streamflow for the Hanjiang river basin. Can J Remote Sens 30(5):680–690 Zhao R-J, Zhang YL, Fang LR et al (1980) The Xinanjiang model. In: Proceedings hydrological forecasting oxford symposium, IAHS, International Association of Hydrological Sciences Press, Wallingford, UK (1980), pp. 351–356 Zhao R, Chen Y, Shi P et al (2013) Land use and land cover change and driving mechanism in the arid inland river basin: a case study of Tarim River, Xinjiang. China Environ Earth Sci 68(2):591–604

123