Environ Monit Assess (2016) 188:690 DOI 10.1007/s10661-016-5705-5

Assessment of impacts of land use changes on surface water using L-THIA model (case study: Zayandehrud river basin) M. Mirzaei

&

E. Solgi & A. Salmanmahiny

Received: 6 July 2016 / Accepted: 17 November 2016 # Springer International Publishing Switzerland 2016

Abstract Land use changes in a basin are the most important factors affecting its hydrology and water quality. A hydrological model is an effective tool in assessing the effects of land use change on surface water. In this study, the effects of land use changes in the Zayandehrud basin are estimated using long-term hydrologic impact assessment model. This model is applicable using long-term data on climate, soil hydrological groups, and land use maps. The study covered three land uses across 18 years (from 1997 to 2015), and we used data on 30 years of precipitation (from 1985 to 2015) in the model. The results of modeling revealed that the average runoff volume increased from around 5,765,034 m3 in 1997 to 8,894,525 m3 in 2015. The results also showed an increase in runoff depth. Land use changes over the study period showed an increase of residential areas, bare land, and agricultural lands and a decrease of pasture and forests. The results can be used to make decisions and monitor changes in land use to

M. Mirzaei (*) : E. Solgi Department of Environmental Science, Malayer University, Malayer, Hamedan, Iran e-mail: [email protected] E. Solgi e-mail: [email protected] A. Salmanmahiny Department of Environmental Sciences, Gorgan University of Agriculture Sciences and Natural Resources, Gorgan, Golestan, Iran e-mail: [email protected]

control the depth and volume of runoff. Using output maps helps in delimitation of the areas that have high runoff average and in implementation of the management plans for controlling the amount of runoff in these areas. Appropriate land use design can decrease impacts of land use changes including hydrologic effects. Keywords Land use changes . Zayandehrud river basin . Hydrological modeling . L-THIA model

Introduction Water is one of the most precious resources and its quantity and quality is essential for sustainable development (Vörösmarty et al. 2005). The rivers are very important due to their vital role in securing water for urban and rural areas (Kashefipour and Tavakolizadeh 2008). Rivers have been places of human concentration from the dawn of history. Cities, agricultural and industrial centers, are usually near the rivers to use their water resources. Over time and with development of these communities and consequently increasing use of water resources, manipulation and abnormal changes in water quality in rivers have increased (Enrique et al. 2007). Surface water bodies are the potential recipients of the contaminations contained in surface runoff from their catchments (Jarvie et al. 1998). Therefore, surface water quality management is extremely important. Land use change has ecological and socioeconomic ramifications in declining water quality in watersheds or rivers (Riki 2000). Knowledge of land use and land cover for many

690

Page 2 of 19

of the planning activities is important as an essential element in natural systems (Lillesand and Ralph 2000). Land use changes are one of the most important problems in the world. Change of land use patterns occurs mainly because of socioeconomic benefits. However, this process has negative effects on the environment. Change in hydrological conditions, reduction of water resource, and their quality are important effects resulting from land use change )Bhadori et al. 2000; Tang et al. 2005a; Li et al. 2007; Galdavi et al. 2012). Land use change is a key example of the impact of humans on the environment (Khakpour et al. 2007; Oluseyi 2006). Water is one of the most precious national resources. Because water is a renewable resource and many economic and social benefits are derived from its correct use, appropriate land use is very important. Population growth, coupled with a lack of planning for land productivity, is a cause that lead to conversion of forests and pastures into agricultural land. As a result, less

Fig. 1 Zayandehrud basin with the main and secondary rivers

Environ Monit Assess (2016) 188:690

water in the upstream river penetrates to earth and thus flows faster overland. In this way, floods become more frequent, more severe, and more sudden and most people suffer from severe flooding (Barkhordari and Khosroshahi 2007). Land use in the general concept refers to the use of land in the existing situation which includes all land uses in various sectors of agriculture, natural resources, and industry. In other words, these include farming activities (rainfed and irrigated), residential areas, forests, pastures, and mining and industrial facilities (Ahmadi 1995). Land use/ land cover is one of the effective variables on surface water quality (Mirzaei 2013). Conversion of agricultural lands, forests, grasslands, and wetlands to urban areas with rapid and widespread increase in impervious surfaces can change natural hydrological conditions in the watershed. As a result of this change, increase in volume and the amount of surface runoff together with reduced water storage underground, base flow, water quality degradation in rivers and shallow waters occurs (Engle et al. 2003; Li

Environ Monit Assess (2016) 188:690

Page 3 of 19 690

Table 1 Classification of land use maps based on the model classes in the study area Code of land use classes

Land use classification based on L-THIA model

Land use classes

7000

Forest

Forest

6000

Pasture/grassland

Pasture

3000

Agriculture

Agriculture

4000

High-density residential areas

Bare land

1000

Water

Water

4000

High-density residential areas

City

6000

Pasture/grassland

Green space

Fig. 2 Classified land use map in 1997

et al. 2007 and Lin et al. 2007) and ultimately leads to more and larger number of local flood events, reduction of water resources in urban and rural areas, and reduction of the base flow in the water channels during the dry season. On a global scale, extensive changes in land use and land cover is going on. These changes include deforestation, increasing agricultural activities, drainage of wetlands, road construction, and urbanization and will bring many effects including hydrological effects. The hydrological effects are visible on seasonal and annual flows, flood, water quality, and erosion (Weng 2001). Also, conversion of rangeland and forest to farming lands has led to the change of watering regime in these areas

690

Page 4 of 19

(Khalighi et al. 2005). On the global, regional, and local scales, important effects of human activities on the

Fig. 3 Classified land use map in 2002

Environ Monit Assess (2016) 188:690

hydrological system have been created due to land use change. If this type of important changes occurs in the

Environ Monit Assess (2016) 188:690

hydrological balance of a watershed or sensitive parts of a watershed, these can have short- and long-term effects

Fig. 4 Classified land use map in 2015

Page 5 of 19 690

including an increase in downstream flooding and reduction of long-term depth and groundwater

690

Environ Monit Assess (2016) 188:690

Page 6 of 19

availability. Low water levels could change wetland systems and could create drying periodic systems during the drought period (Bhadori et al. 2000). In aquifers, conversion of non-urban to urban land leads to increase of impervious surfaces and runoff (Shi et al. 2007). Then, urban drainage systems increase flood peak and cause damage to water quality. Urban areas are important sources of non-point pollution as well. Flooding of urban areas absorbs a variety of contaminants such as nutrients, sediment, oil, grease, heavy metals, pesticides, and pathogens that cause water quality degradation (Tang et al. 2005b). Land use change in addition to the impact on regional climate and soil also affects the quality of surface water and the water cycle (Liang et al. 2004). Optimal management of natural resources in a region needs an understanding of the impact of land use/cover changes on the hydrologic cycle in that region (Scanlon et al. 2005). Land use changes, overexploitation of the soil, and the constant use of surface water resources have significant negative effects on the environment (Vito et al. 2003). In the present study, the long-term hydrological impact assessment (L-THIA) model was used to evaluate the effects of land use change on surface water. Maps of land use and soil hydrological groups together with long-term climate data are required to run the model. This model has been successfully used to evaluate the effects of urban development on surface water resources in terms of quantity and quality by Bhadori et al. 2000, Tang et al. 2005, Perry and Nawaz 2008, Yang et al. 2008, Wilson and Weng, 2010, and Mahiny et al. 2012. The results of all the above studies show increase of the average volume and depth of surface runoff over time with increasing changes in natural land use.

great, and spectacular river in Iran that has faced with drought in recent years. Much of this watershed has an arid and semi-arid climate. The river contributes to development of agriculture, water supply in industries, and economic activities. Zayandehrud basin and its main and secondary rivers are shown in Fig. 1. Method Hydrological models can be used for assessment of the effects of land use change on water quality and quantity (Lim et al. 2001). Many hydrological models are hard to apply due to their complexity and the required data. Initial assessment of the hydrological effects of land use change requires a simple model that can provide a primary assessment of absolute and relative effects in watershed with available input data and identify the requirements for more advanced models (Bhadori et al. 2000). The Long-Term Hydrologic Impact Assessment model (L-THIA) is a useful tool for modeling non-point source pollution and evaluation of the potential impacts of land use change on surface runoff and water pollution (Mahiny et al. 2012). The L-THIA model has been developed as an easy and intuitive analysis tools by the Purdue University (Ma 2004). This model has many advantages than other models. The L-THIA is a simple model that is linked with GIS and estimates direct runoff from the input data. This model needs less data in comparison to other models and is easier to apply. It is possible to use this model in different countries with different weather patterns and topography. Input data required for this model are land use maps, soil maps (soil hydrological groups), text file for rainfall of the last 30 years, and a text file of CNtable. The L-THIA model calculates non-point source

Materials and methods

Table 2 Hydrologic Soil Groups

Study region

Soil Description group

USDA soil texture

The study was carried out in Zayandehrud basin (latitude 31° 12' N, longitude 50° 02' E) in Iran. Zayandehrud watershed extends over a total area of 41,500 km2, and the Zayandehrud river flows in this watershed with a length of about 350 km from west to east. This river is one of the most important rivers in the Iranian plateau that is originated from the Zagros Mountains in the West of Isfahan Province and ends in Gavkhuni wetland. Zayandehrud river is a historical,

A

Deep soils with high permeability rate

Sand, loamy sand, or sandy loam

B

Relatively fine to relatively coarse-textured soils with moderate permeability Fine textured soils with low permeability rate Soils with very low transfer and permeability rates

Silt loam or loam

C D

Sandy clay loam Clay loam, silt clay loam, sandy clay, silt clay, or clay

Environ Monit Assess (2016) 188:690

pollution based on the volume of runoff and EMC (event mean concentration). EMC is a key analytical parameter, which refers to a flow-weighted average concentration in the whole process of a rainfall-runoff event, defined as the total pollution load mass divided by the total runoff volume (Bertrand-Krajewski et al. 1998), and this can be used to evaluate the effects of rainfall runoff on the water quality of the receiving waters. Using the CN equation in L-THIA model is a simple alternative to complex hydrologic models that

Fig. 5 Hydrologic soil group map

Page 7 of 19 690

need a lot of data that is often not available in most areas. The L-THIA model uses data that are available easily and provides an overall estimation of long-term hydrologic effects through land use change. This model estimates daily runoff for different values of CN using daily rainfall data and CN value. Then, daily values are summed and annual runoff is provided that is considered as the output of the model. The L-THIA applies no calibration of the model data with real data of the area. In the L-THIA model, CN default values are

690

Page 8 of 19

prepared for land uses and the hydrologic soil groups. Also, there is a range of values for each land use and composition of the hydrologic soil groups that is dependent on soil moisture conditions, land cover types, and land management (Galdavi et al. 2012). To prepare land use maps, satellite images of the three time periods were

Fig. 6 Annual precipitation in 1997

Environ Monit Assess (2016) 188:690

used. These images included Landsat TM and ETM+ images of the years 1997, 2002, and 2015. Preparation of the required data The primary data sources for the land cover classification was Landsat Thematic Mapper (TM) of the years

Environ Monit Assess (2016) 188:690

1972 and 2002 and Landsat Enhanced Thematic Mapper Plus (ETM+) of the year 2015. Image georegistration was performed using road vector map and aerial photographs. Resampling procedure was carried out using the nearest neighbor interpolation. Spectral correction of imagery was performed to create a distinct phenomenon and high quality images and to eliminate

Fig. 7 Annual precipitation in 2002

Page 9 of 19 690

the adverse effects of light and atmosphere. Then, by the correlation between bands, false color composites of the bands 4, 3, and 2 (NIR, red, and green) for the years 1972 and 2010 was created and supervised classification was performed using maximum likelihood method. This model considers eight classes of land uses, including high and low density residential land,

690

Environ Monit Assess (2016) 188:690

Page 10 of 19

Fig. 8 Annual precipitation in 2015

Table 3 The results of satellite data classification accuracy

agricultural areas, range or pasture, forest, commercial land use, water zone, and industrial areas. To run the model, land use map of the study area was reclassified. Also, if some of the classes are different with classes of the L-THIA model, they should be placed in a category that most closely

Land use maps

Kappa criteria

Total accuracy

1997

0.9703

0.9812

2002

0.9801

0.9875

2015

0.9402

0.9572

Environ Monit Assess (2016) 188:690

Page 11 of 19 690

Table 4 Area of different land uses during the studied years Land use

1997 area (km2) 2015 area (km2)

HD residential and bare lands 3679.8

8574.05

Agricultural lands

7602.8

8636.8

Range lands

26,041.8

21,285.5

Forest

1499.8

930.54

Water

2675.8

2073.11

Fig. 9 CN map in 1997

resembles the characteristics of runoff. For example, bare land class is absent in the L-THIA model, so it was placed in a high-density residential area. Two land uses are somewhat similar in terms of runoff production (Galdavi et al. 2012). Land use maps from the years 1997, 2002, and 2015 were re-classified to run the model. Then, for any land use class, codes from 1000 to 7000 were

690

Page 12 of 19

allocated using land use classification module in L-THIA GIS/NPS program and land use maps were converted into raster layers to run the model. Table 1 shows classification of land use map based on classes of the L-THIA model. Figures 2, 3, and 4 show land use maps for the years 1997, 2002, and 2015.

Fig. 10 Runoff depth map in 1997

Environ Monit Assess (2016) 188:690

Preparing the soil map The soil in the L-THIA model is classified into four hydrological groups including A, B, C, and D based on potential probability in runoff production. This classification is that of the United States Soil Conservation Service based on water movement in

Environ Monit Assess (2016) 188:690

Page 13 of 19 690

Fig. 11 Runoff volume map in 1997

the soil and water permeability properties (USDA 2007). Group A consists of coarse-grained soils, sand, and with good drainage that has the highest penetration rate and the lowest runoff potential. Soils of group D are heavy clay soils with poor drainage with the lowest penetration rate and the highest runoff potential. Soil groups C and B are

Table 5 The results of the annual runoff volume for the studied years The total amount of runoff volume (m3)

Years of study

5,765,033.68 6,220,366 8,894,524.64

1997 2002 2015

690

Page 14 of 19

between groups A and D. Detailed classification of the four groups is presented in Table 2, and its map is shown in Fig. 5. To determine the

Fig. 12 Increased runoff volume from 1997 to 2002 (m3)

Environ Monit Assess (2016) 188:690

hydrologic soil group, Bland use suitability^ and Bland use type^ were used. To prepare the soil hydrological map, using the maps of slope,

Environ Monit Assess (2016) 188:690

topography, and satellite imagery, land map with very low slope was produced. Then, the low slope of the land maps were overlaid with geological map and land use and hydrologic soil groups were

Fig. 13 Increased runoff volume from 2002 to 2015 (m3)

Page 15 of 19 690

determined according to distribution of soil texture and geomorphologic map. The hydrologic soil groups (HSG) were derived from the land type data by reclassifying the soil textures (Fig. 5).

690

Environ Monit Assess (2016) 188:690

Page 16 of 19

Table 6 Observed annual total base flow and annual direct runoff over the study period Year

Annual total base flow (m3)

Annual direct run off (m3)

1997

960

190

2002

890

210

2015

601.51

280

Preparation of rainfall data

Results and discussion Satellite data classification accuracy Accuracy assessment was accomplished using ground control points. For this, the classified image is compared with an image containing ground control points. This procedure was performed for each land use map, and kappa values and overall accuracy were obtained in the range of 0–1. The closer this number is to one, the higher is the accuracy of the classified map. The results of this assessment are presented in Table 3.

Climatic data Land use changes The L-THIA hydrological model needs daily rainfall data for 30 years for long term impact assessment. In this model, rainfall data is used as a text file with Btxt^ format. For the preparation of this file, the data was converted to an acceptable format in excel software. Finally, with inputs of land use, soil, and rainfall, the L-THIA was run and annual runoff volume and depth maps were extracted in GIS. Maps of annual precipitation from 1997 to 2015 are shown in Figs. 6, 7, and 8.

Land use changes across 18 years from 1997 to 2015 are shown in Table 4. According to Table 4, the residential areas have increased from 3679.8 to 8574.05 km2 and also agricultural lands have increased from 7602.8 to 8636.8 km2 during 18 years and the amount of range lands has decreased from 26,041.8 to 21,285.5 km2. Also, forest and water have decreased in the area of 569.26 and 602.69 km2, respectively.

Fig. 14 Changes to the values of annual runoff volume in the period under study

Environ Monit Assess (2016) 188:690

Page 17 of 19 690

Model implementation The L-THIA model was run for the years 1997, 2002, and 2015 using land use maps, hydrological soil group map, and rainfall data for 30 years. The precipitation data from 30 years was obtained from Isfahan regional water authority (from 1985 to 2015). To prepare annual runoff depth and annual runoff volume maps, CN map must first be prepared which was conducted so using land use and hydrological groups of soil maps for each year separately. Figure 9 shows a CN map for the year 1997. After preparation of CN map, annual runoff depth and volume were prepared using L-THIA. Figures 10 and 11 show annual runoff depth and annual runoff volume in 1997, respectively. Table 5 shows the results of the investigation of runoff volume. The results of the L-THIA model showed that depth and volume of annual runoff have increased over time in the studied period. The increased runoff volume from 1997 to 2002 is shown in Fig. 12 and also the increased runoff volume from 2002 to 2015 is shown in Fig. 13. According to Figs 12 and 13, the maximum increased runoff volume is in the regions that HD residential is developed. Also, the amount of observed annual total base flow and annual direct runoff over the study period are presented in Table 6. Table 6 illustrates that base flow is reduced over time i.e., the penetration is reduced and amount of surface runoff is increased that is consistent with the evidence in this study. Land use change has important and direct impacts on hydrological conditions in the basin; one of which is quantitative changes in surface waters. Development of residential areas in the region plays an important role in increasing the depth and volume of surface runoff. The residential areas have spread from the years 1997 to 2015 in the study area. This has led to the creation of impervious surfaces and increase in the depth and volume of runoff. Also, degradation of pasture lands and their conversion into residential areas or bare lands over Table 7 Average annual precipitation and the annual calculated runoff depth Average annual precipitation Annual runoff depth (cm) (cm)

Years of study

27.44

464.18

1997

25.66

501.37

2002

20.34

547.42

2015

time has had an important role in changing hydrological conditions, confirmed by the results of the model and the depth and volume of runoff obtained. Changes to the values of annual runoff volume in the period under study are shown in Fig. 14. Average annual precipitation and the annual calculated runoff are indicated in Table 7. According to Table 7, increase in runoff depth and volume seems not highly correlated with climate and precipitation and the reason for the increased runoff can be attributed to land use.

Conclusion Land use is one of the most important factors that alters local and regional hydrology, especially in rivers. With increasing population growth, residence in cities, and establishment of industrial units, land use patterns change, and so does runoff from rainfall and urban wastewater discharges, leading to an increase in the amount of nutrients and other pollutants released into rivers and surface water resources (Hara et al. 2004). Evaluation of the long-term hydrologic impacts of land use change is important for controlling runoff and nonpoint source pollution. In this study, changes in runoff volume and depth were calculated by L-THIA model for 3 years in Zayandehrud river basin. Zayandehrud basin is the most important basin in Iran that has undergone urbanization in recent decades. Maps of multi-temporal land use and soil hydrological groups along with precipitation data were used in the model. Our application of L-THIA model was helpful, albeit in a relative manner, in assessing the qualitative and quantitative effects of land use change on surface water resources. The results showed an increase in total depth and volume of runoff in the studied period which confirmed the effects of land use changes on surface runoff in the study area. We demonstrated a likely change of 75.40 cubic meters of runoff per square kilometer over 18 years in the area. Runoff depth is also increased in the studied period. The precipitation data showed that average annual precipitation has decreased; thus, we could safely postulate that long-term precipitation trend is not responsible for the modeled and observed increase in runoff and land use change is the main cause of runoff increase in the study area. In Zayandehrud basin, residential lands have increased in the past 18 years expanding the impermeable zones and increasing

690

Page 18 of 19

surface runoff. Furthermore, rangeland in Zayandehrud basin has decreased and unfortunately most of these lands have been converted to bare lands. Our application showed that the maximum increased runoff volume happened in the regions with development of highdensity residential areas. The L-THIA model estimated an increase of 154% in average annual direct runoff while that of the observed annual runoff demonstrated an increase of 147% in the studied period. This indicates that our modeling data is close to true field data pointing to the applicability of the model in real situations when only relative results are required and when there is a lack of precise and complete data for other demanding models. We also found that L-THIA model is a useful tool for identifying environmentally sensitive areas for nonpoint source pollution potential and assessment of land use changes scenarios for management of this pollution. The results showed that urban sprawl can cause an increase in the non-point source pollution. We believe that appropriate design in land use can reduce these effects. The results can be helpful in showing the requirements for proper implementation of programs to control land use change and maintain ecological balance. Accurate physical design in land use can decrease land use change impacts including the hydrologic effects. The results can also be useful in prioritizing subbasins and implementation of watershed management in vulnerable sectors. The results of this study indicated that implementation of better land use plans is necessary for controlling land use changes and preservation of ecological balance in Zayandehrud basin. The results as such can be linked to the current land use planning practice, currently undergoing for Isfahan Province, towards a more sustainable future.

References Ahmadi, R. (1995). The role of land use in inducing or increasing forest mass movements. M.S C thesis, faculty of humanities, Tarbiat Modarres University, 124 pages. Barkhordari, J., & Khosroshahi, M. (2007). Investigation of land cover changes impacts and climate on river flow (case study: Minab basin). Journal of Research and development on natural resources, 77, 191–199. Bertrand-Krajewski, J. L., Chebbo, G., & Saget, A. (1998). Distribution of pollutant mass vs volume in stormwater discharges and the first flush phenomenon. Journal of Water Resources, 32(8), 2341–2356.

Environ Monit Assess (2016) 188:690 Bhadori, B., Harbor, J., Engel, B., & Grove, M. (2000). Assessing watershed-scale, long-term hydrologic impacts of land-use change a GIS-NPS model. Environmental Management, 26, 643–658. Engle, B. A., Choi, J. Y., Harbor, J., & Pandey, S. (2003). Webbased DSS for hydrologic impact evaluation of small watershed land use changes. Computers and Electronics in Agriculture, 39, 241–249. Enrique, S., Colmenarejo, M. F., Vicente, J., Rubio, A., Garcia, M. G., Travieso, L., & Borja, R. (2007). Use of the water quality index and dissolved oxygen deficit as simple indicators of watersheds pollution. Ecological Indicators, 7(2), 315–328. Galdavi, S., Mohammadzadeh, M., Salman mahini, A., & Najafipour, A. (2012). Application of L-THIA model in assessment of impacts of land use change on water resources. Environment and Development, 3(6), 35–46. Hara, Y., Takeuchi, K., & Okubo, S. (2004). Urbanization linked with past agricultural land use patterns in the urban fringe of a deltaic Asian mega-city: a case study in Bangkok. Landscape and Urban Planning, 73(1), 16–28. Jarvie, H. P., Whitton, B. A., & Neal, C. (1998). Nitrogen and phosphorus in east coast British rivers: speciation, sources, and biological significance. Science of the Total Environment, 210, 79–109. Kashefipour, S., Tavakolizadeh, A. (2008). Hydrodynamic and qualitative model (FASTER) and its application in river engineering. Iran-Water Resources Research, 3(3), 56–68. Khakpour, B., Velayati, S., & Kiannejad, G. H. (2007). Pattern of land use change in Babol city, Iran. Journal of Geography and regional development, 18, 46–64. Khalighi, S. H., Mahdavi, M., & Saghafiyan, B. (2005). The effect of land use changes on flooding by using NRCS model (case study: Barandooz Chay basin in Azarbayejan, Iran). Iranian Journal of Natural Resources, 58(10), 733–742. Li, N., Xu, Y. P., & Guo, H. C. (2007). Analysis of long-term impact of urbanization on surface run-off in Xitiaoxi river basin. Environmental informatics archives, 5, 346–353. Liang, T., Hao, W., Hsiang-te, K., & Chao-sheng, Z. (2004). Agriculture land-use effects on nutrient losses in West Tiaoxi watershed, China. Journal of the American Water Resources Association (JAWRA), 40(6), 1499–1510. Lillesand, T. M., & Ralph, W. K. (2000). Remote sensing and image interpretation (4th ed.). new York: Wiley ISBN 0 – 471 – 25515 – 7. Lim, K. J. Engel, B. A., Kim, Y., Bhadori, B., Harbor, J. (2001). Development of the long term hydrologic impact assessment (LTHIA) WWW systems. This paper was peer- reviewed for scientific content. Pages 1018–1023. In: D. e. Stott, R.H. Mohtar and G.C. Steinhardt (eds). , conservation organization meeting held May 24–29, 1999 at Purdue University and the USDA- ARS national soil erosion research laboratory. Lin, Y. P., Hong, N. M., Wu, C. F., & Veburg, P. H. (2007). Impacts of land use change scenarios on hydrology and land use patterns in the Wu-Tu watershed in northern Taiwan. Landscape and Urban Planning, 80, 111–126. Ma, Y. (2004). L-THIA: a useful hydrological impact assessment model. Nature and sciences journal, 2, 68–73. Mahiny, A. S., Hosseinnia, A., Ghasempoori, S. M., & Tavasoli, A. (2012). Long term hydrological impact assessment of land use change on surface annual runoff at the catchment scale. Geography and Development Journal, 10, 33–36.

Environ Monit Assess (2016) 188:690 Mirzaei, M. (2013). Empirical modeling of the relationship between surface water quality parameters and measures of landscape using adaptive neuro-fuzzy inference system (case study: Mazandaran province), M.S C thesis, Tarbiat Modarres University, Faculty of Natural Resources, 115 p. Oluseyi, O. F. (2006). Urban land use change analysis of a traditional city from remote sensing data: the case of Ibadan metropolitan area, Nigeria. Humanity and social sciences journal, 1, 42–61. Perry, T., & Nawaz, R. (2008). An investigation into the extent and impacts of hard surfacing of domestic gardens in an area of Leeds, United Kingdom. Landscape and Urban Planning, 86, 1–13. Riki, M. (2000). Evaluation of groundwater quality in Khash plain. M.S C thesis in the field of groundwater, Department of Irrigation engineering. Faculty of Natural Resources, Tehran University. Scanlon, B., Reedy, R., Tonestromw, D., Prudicz, D., & Dennehy, K. (2005). Impact of land use and land cover change on groundwater recharge and quality in the southwestern US. Global Change Biology, 11, 1577–1593. Shi, P. J., Yuan, Y., Zhang, J., & Wang, J. A. (2007). The effect of land use/cover change on surface runoff in Shenzen region, China. Catena, 69, 31–35. Tang, Z., Engel, B. A., Lim, K. J., Pijianowski, B. C., & Harbor, J. (2005a). Minimizing the impact of urbanization on long term runoff. Journal of the American Water Resources Association, 41, 1347–1359.

Page 19 of 19 690 Tang, Z., Engel, B. A., Pijianowski, B. C., & Lim, K. J. (2005b). Forecasting land use change and its environmental impact at a watershed scale. Journal of Environmental Management, 76, 35–45. US Department of Agriculture, Natural Resource Conservation Service (2007). National Soil Survey Handbook, title 430-VI, available online: http://soils.usda.gov/technical/handbook/. Vito, F. U., Raffaele, G., & Nicola, L. (2003). A fuzzy knowledgebased decision support system for groundwater pollution risk evaluation. Environmental Management, 73(3), 197–189. Vörösmarty, C. J., Lévêque, C., Revenga, C., Bros, R., & Caudill, C. (2005). In R. Bos, C. Caudill, J. Chilton, E. M. Douglas, M. Meybeck, D. Prager, P. Balvanera, S. Barker, M. Maas, C. Nilsson, T. Oki, & A. Reidy (Eds.), Fresh water in millennium assessment. Volume 1: conditions and trends working group report (pp. 165–207). Washington: Island Press. Weng, Q. (2001). Modeling urban growth effects on surface runoff with the integration of remote sensing and GIS. Environmental Management, 28, 737–748. Wilson, C., & Weng, Q. (2010). Assessing surface water quality and its relation with urban land cover changes in the Lake Calumet Area, Greater Chicago. Environmental Management Journal, 45, 96–111. Yang, L., Ma, K., Guo, Q., & Bai, X. (2008). Evaluating long-term hydrological impacts of regional urbanization in Hanyang, China, using a GIS model and remote sensing. International Journal of Sustainable Development & World Ecology, 15, 350–356.