Background: The spatiotemporal analysis of urban land use/land cover change (LULCC) helps to understand the dynamics of the changing environment of green infrastructure (GI) on the basis of sustainable city development. There are important links between spatiotemporal land use/land cover and GI change in urban areas. Therefore, the main objective of this study was to examine the spatiotemporal trends of urban land use/land cover and GI changes in Bahir Dar and Hawassa cities for the last four decades (1973–2015). Three different sets of Landsat satellite data were procured from EMA for Bahir Dar and Hawassa from 1973, 2000 and 2015 using Landsat 4 MSS, 7 TM and 8 OLI respectively. Based on this, using ERDAS Imagine (ver. 9.2) and Arc GIS ( Ver.10.3) five LULCC classes were identified for analysis purpose. Result: The results show that vegetation decreased by 30 and 14% in Bahir Dar and Hawassa respectively for the period 1973–2015, while built‑ up areas expanded by 10 and 24% respectively in the two cities. These land use changes have significant impacts on spatiotemporal trends of GI in urban areas. GI has increased in Bahir Dar and Hawassa in association with built‑ up area expansion and deliberate activity of city administrations with effective implementation of spatial plans of corresponding cities. Conclusions: There is a growing concern about GI in cities. Policy makers and stakeholders should also decide on how to use the land at present and in the future. LULCC policymaking processes should aim to balance GI and other types of land use/land cover for sustainable urban development. Urban LULCC has important effects on the urban GI system. Keywords: City planning, Green infrastructure, GIS, Landsat image, Land use land cover change, Remote sensing, Spatiotemporal 2014). The process of urbanization involves the growth of Background urban population and built-up areas. The share of world With a little more than 50% of the human population liv- urban population is expected to increase to 66% by 2050, ing in urban areas, urbanization is now recognized as a and of this about 90% will be concentrated in Africa and major phenomenon (UN 2014; Zhang et al. 2013). Social Asia (UN 2014). This population increase will lead to fast scientists, urban planners, and geographers have inves- growth of built-up areas that consumes the surrounding tigated the unprecedented urban concentration from productive land and encroaches on the necessary ecosys- many perspectives, including the geography, demograph- tems. At the same time, the horizontal rapid expansion ics, economies and spatial evolution of cities (McIntyre of built-up areas will lead to discontinuous suburbs with et al. 2000) as well as urban green infrastructure (Mell low density and uneven pattern (Tewolde and Cabral 2011; Varshney 2013). *Correspondence: Kg19me@gmail.com As urbanization increases and urban areas continue Department of Geography and Environmental Studies, University to grow fast, there is a concern on urban environment of Gondar, P.O. Box 196, Gondar, Ethiopia Full list of author information is available at the end of the article © The Author(s) 2018, corrected publication May 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 2 of 15 and its quality because the quality of urban environment open spaces, cemeteries, nursery sites, and green roofs directly influences the social and economic development and walls (MoUDH 2015). of the city (Masser 2007; He et al. 2010). Many scholars Green space is sometime synonymous with green infra- hold the view that the urban environment represents a structure, though the latter is more inclusive than the highly complex area depicting a continuum of different former. Green space helps reduce heat effects of build - spatial, temporal and spectral variability in land use and ings, provides shadow to pedestrians and ground and has land cover (Haregeweyn et al. 2012). Spatial variability the ability to improve air quality and the environment arises due to the varied landscape: temporal variations (Noor et al. 2013). The development of urban green infra - are attributed to periodic seasonal changes over the year structure planning and management practices requires while spectral variability is due to the great variety of important information from LULCC studies (Yang et al. materials and structures associated with the urban area 2014). Previous studies implied that traditional investiga- (Zoran 2007). It is thus necessary to analyze the spati- tion of urban environment was not considered GI (Miller otemporal patterns of LULCC in order to understand the and Hobbs 2002). The urban green infrastructure, how - urban ecology (McIntyre et al. 2000; Abebe and Megento ever, enables urban residents to experience outdoors 2016). visually and kinetically. Green infrastructure network in Land use/land cover change (LULCC) refers to the any urban area is significant because it attempts to pro - earth’s territorial surface modification by human activi - vide optimal experiential qualities to urban residents and ties (Anderson et al. 1976; Meyer and Turner 1992; Lu to overcome the negative effects of living in the urban et al. 2004; Muriuki et al. 2011; Ayalew et al. 2012). The built environment (Mansor et al. 2010). Moreover, based process of LULCC affects biodiversity, climate, soil, and on a deeper understanding of the relationships between air in particular, and the ecosystem, in general, and it has the LULCC and GI change require that the underlying become the greatest environmental concern for human mechanisms, patterns, and processes of land conversion beings to date (Long et al. 2007; Tsegaye et al. 2010; as well as the response of urban environment should be Hailemariam et al. 2016). LULCC is useful to understand addressed throughout official decision-making processes environmental changes because it can provide a tool to (Zhang et al. 2013). The planners and decision-makers assess ecosystem changes and their environmental impli- could fully evaluate the consequences of different land cation at various temporal and spatial scales (Anderson development scenarios and therefore have scientific et al. 1976; Haregeweyn et al. 2012). basis with which to improve future planning and regula- Urban space consists of built-up areas that include vari- tions of GI. In terms of GI, the spatiotemporal analysis ety of land uses in commercial, institutional and residen- of LULCC can help to understand the dynamics of the tial areas. It also consists of non-built area that is mostly changing environment of GI and form the basis for sus- dominated by greenery and open spaces (Moroney and tainable development and provide a fundamental piece of Jones, 2006; Tzoulas et al. 2007; Mansor et al. 2010). information required for policy making and planning (Hu Previous researches (Kong and Nakagoshi 2006; Phan et al. 2008; Teferi et al. 2016). and Nakagoshi 2007; Byomkesh et al. 2012) indicated Though LULCC is not a recent phenomenon in Ethio - that urban green spaces are those lands that are covered pia (Hailemariam et al. 2016), it is exacerbated by the with natural or man-made vegetation but are present in scale, speed and long-term nature of urbanization and built-up areas. However, the universally agreed defini - modernization (Msoffe et al. 2011). Existing studies on tion is still arguable. Most developed countries have LULCC in Ethiopia have focused on land degradation their own definition (Byomkesh et al. 2012). Therefore and associated consequences due to expansion of cultiva- this research used as its working definition stated by the tion and deforestation (Taddese 2001; Feoli and Vuerich Ethiopian Ministry of Urban Development and Housing 2002; Amsalu et al. 2007; Meshesha et al. 2010; Tsegaye (MoUDH): green infrastructure typologies to include et al. 2010). There is little effort to understand LULCC in parks, sports fields, roadside and squares, plazas and fes - relation to green infrastructure changes in urban areas. tive areas, river and riverside areas, lakes and lakeside This study highlights the important links between spa - areas, watershed areas, urban agriculture development, tiotemporal land use/land cover and green infrastructure woodlots and green belts (inside and surrounding for- change in urban areas. In this research, green infrastruc- ests), private compounds and surroundings, institutional ture is taken as one category of land use/land cover that compounds and surroundings (both governmental and is an interconnected network of multifunctional, pre- non-governmental), communal housing compounds and dominantly un-built, spaces that support ecological and surroundings (condominiums, real estate, etc.), religious social activities (Kambites and Owen 2006; Tzoulas et al. institutions compounds and surroundings, neighborhood 2007; MoUDH 2015). The transformation of land use/ land cover types leads to a change in the structure and Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 3 of 15 Fig. 1 Location map of study areas function of green infrastructure services (Lei and Zhu Dar is located at 11° 36′ North and 37° 23′ East and has 2017). The need to balance economic, social and eco - an average elevation of 1801 m above sea level. It has a logical ecosystems is becoming increasingly urgent city administration with special zone status and nine sub- because LULCC is in the direction of rapid urbaniza- cities which have district status (Fig. 1). It is also the seat tion (Song et al. 2016). This study aims to investigate the for the Bahir Dar Zuria district. Hawassa is located at 07° rapid urban expansion on LULCC and GI, and its devel- 03′ North and 30° 28′ East with an average elevation of opment and planning. Our research focused on (1) The 1708 m above sea level. Similar to Bahir Dar, Hawassa has rates of LULCC in Bahir Dar and Hawasa between 1973 a city administration status and has eight sub-cities with and 2015, (2) LULCC trends during the 1973–2000 and district status (Fig. 1). Hawassa is also a seat for Sidama 2000–2015 periods for both Bahir Dar and Hawassa, administrative zone. (3) Which land-cover types were most affected by the According to the National Meteorological Agency change process, and (4) The rates of changes and con - (NMA), Bahir Dar has an average annual temperature version from other land cover types to green spaces and and precipitation of 19.6 °C and 1419 mm respectively urban areas over the period 1973–2015. (NMA 2013). It is situated in the woina-dega agro-eco- logical zone and experiences uni-modal rainfall over a Methods 3-month period from mid-June to mid-September. Study area Hawassa has an average annual temperature and precipi- Bahir Dar and Hawassa are the capital cities of Amhara tation of 20.8 °C and 993.4 mm respectively (NMA 2013). National Regional State and Southern Nations, Nationali- It is one of the major urban areas of Ethiopia located ties and Peoples’ Regional State respectively. The former inside the greater Ethiopian rift valley. It experiences uni- is located at 565 km to Northwest of Addis Ababa, on modal rainfall over a 3-month period from mid-June to the southern shore of Lake Tana, the source of the Blue mid-September and has woina-dega agro ecological zone. Nile (Abay) river, while the latter is located at 275 km to the south of Addis Ababa near the eastern shore of Lake Hawassa, one of the rift valley lakes in Ethiopia. Bahir 1 Woina-dega is a local term that defines mid altitude climate. Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 4 of 15 Table 1 Land use land cover change (LULCC) classification Ethiopian mapping agency (EMA) in GeoTIFF file format schemes used in this study projected in UTM projection and WGS 84 datum univer- sal transverse Mercator (UTM), Zone 37° North coordi- LULC class Description nate system. The three Landsat satellite images with 30 m Urban built‑up area Includes areas with all types of artificial resolution were acquired for January 1973, January 2000 surfaces, including residential, com‑ and January 2015. A study by Sadidy et al. (2009) pointed mercial, and industrial land uses as well as transportation infrastructure out that Landsat images with varying resolution are Vegetation Includes areas with dense vegetation among the most widely used data sources in order to gain cover, such as those covered with important input for mapping and planning projects. The shrubs forming closed canopies, trees Landsat images were geo-referenced to the digitized map and other vegetation that is relatively tall and dense, as well as areas covered of the corresponding area using first-order polynomial with both indigenous and exotic trees transformation and nearest neighborhood resembling Water body Includes lakes, rivers, ponds (Yuan et al. 2005; Murat et al. 2006). Green spaces in built‑up areas An area of grass, trees, or other veg‑ etation set apart for recreational or Data analysis aesthetic purposes inside urban built environment. It includes urban parks, There are many change detection approaches for greenery, roundabouts, public squares remotely sensed images (Yuan et al. 2005). Among these, and plaza, open Spaces, medians and the post-classification comparison method is particularly sport fields attractive due to its nature of clearly identified change Crop land Includes grazing areas, cultivated lands, (Hung and Wu 2005; Muttitanon and Tripathi 2005; Yuan community open lands and areas along the lake shore that are used et al. 2005). This study employs the post-classification for agricultural purposes when the method to detect changes. lake level retreats following the long LULC maps for both Bahir Dar and Hawassa for 1973, dry‑season. Information obtained from local residents indicates that the units 2000 and 2015 were prepared for the study areas by inde- categorized in this category can gener‑ pendently supervised classifications using a maximum ally be used in one way or another for likelihood algorithm classifier. Hence, the five land-cover agricultural purposes classes are as follows: urban built-up, vegetation, water body, green spaces, and crop land were mapped. These five land use/land cover classification (Table 1) These two cities are among the largest and the fast - schemes were chosen considering the standard classes est growing urban centres in Ethiopia. The population defined by the National Aeronautics and Space Admin - of Bahir Dar city grew from 96,140 in 1994, the second istration (NASA) and the US Geological Survey (USGS) census period, to 155,428 in 2007, the third census period as well as the study detail and objectives (Mohan et al. (CSA 2007). The rate of growth between the two cen - 2011). Some studies (Thompson 1996) also outlined suses periods was 3.7%. According to the CSA (2017), the the need to contextualize LULCC classification systems population of Bahir Dar is estimated to be 348,429. The for the local situation. The reason is that no universally population of Hawassa was 69,169 in 1994 and it grew to accepted classification system exists as it is influenced 157,139 in 2007 showing a growth rate of 6% (CSA 1994, by specific users’ objectives and also often by geo - 2007). The CSA estimated the population of Hawassa in graphic location. ERDAS Imagine (ver. 9.2) and Arc GIS 2017 to be 315,267 (CSA 2017). (Ver.10.3) were used to perform LULCC classification in Bahir Dar and Hawassa cities were selected as research a multi-spectral approach. Satellite images with remote sites for this study in addition to rapid population sensing techniques are used to show spatiotemporal increase is that both are lakeside cities, regional capitals, trends of LULCC of the study areas. In order to deter- fast growing cities and have relatively better availabil- mine the quality of the information derived from the ity of green infrastructure as compared to other cities data, accuracy assessment of classification was made for and towns in the country. According to Municipality of 1973, 2000 and 2015 images. We used the most obvious Hawassa (2015) and Municipality of Bahir Dar (2015) method of change detection (Lu et al. 2004; Lu and Weng Hawassa and Bahir Dar has 21.96 and 17.44% GI cover- 2007; Butt et al. 2015) which involves a comparative anal- age respectively. ysis of spectral classifications for times t and t produced 1 2 independently (Mas 1999). The percentages of change Data detection of LULCC were calculated using the following This study uses three different sets of Landsat satellite equation: data for Bahir Dar and Hawassa over four decades (1973– 2015). These satellite images were procured from the Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 5 of 15 A − A Green (2009). The formula for computing producer accu - 2 1 = ∗ 100 (1) racy, user accuracy, overall accuracy, and Kappa index coefficient is given as follows: where, Δ is the land use/land cover change percentage, nii and A and A are initial and final. producer’s accuracy i = (3) 1 2 Gii LULCC areas respectively. In this equation, the positive values suggest a gain, whereas negative values imply a nii loss in extent. The LULCC rate was also computed using User’s accuracy i = (4) Cii the formula suggested by Puyravaud (2003): nii i=1 r = ∗ 100 (2) (5) Over all accuracy = �t ln(A − A ) 2 1 where, r is the annual rate of change in %, Δt is the time k k interval in years during the LULCC being assessed; ln is nii − (GiCi) i=1 i=1 k = (6) the base of the natural logarithm function; and A and A k 1 2 n2 − (GiCi) i=1 are initial and final LULCC areas respectively. In the present study, each image of Landsat 4 MSS, where, i is the class number, n is the total number of clas- Landsat 7 TM and Landsat 8 OLI for both cities were sified pixels that are being compared to ground truth, nii independently classified for the three-time periods (1973, is the number of pixels belonging to the ground truth 2000 and 2015). The ground referenced data were gath - class i, that have also been classified with a class i, Ci is ered by combining Google Earth data and GPS points the total number of classified pixels belonging to class i during the field survey and the resulting samples were and Gi is the total number of ground truth pixels belong- imported to the ERDAS Imagine software and the inter- ing to class i. section files were generated. A nonparametric Kappa index is a measure of agree- LULCC can be summarized in a unique change statistic ment between predefined producer ratings and user that quantifies the proportion of pixels that have changed assigned ratings (Foody 2002). Using formulas, 3, 4, 5 and in the overall study area independent of their classes. 6, the kappa index results indicated that all of the images Field-based information supports the interpretation of met the minimum of 85% accuracy in LULCC analysis to the process of LULCC (Fig. 2a, b). In this study, super- each classified object that matches (intersects) a given vised classification was carried out using the multi-date reference object (Table 2). images to classify the images into clusters and to identify the type of potential changes. Post-classification com - Urban expansion analysis parison is used to detect LULCC among three images in The extent and direction of the cities’ expansion for the Bahir Dar (Fig. 2a) and Hawassa (Fig. 2b). Object based years 1973, 2000, and 2015 were analyzed by superimpos- supervised classification (Zhou and Troy 2008; Radoux ing the different time-series images and by calculating the et al. 2011; Robertson and King 2011; Chen et al. 2012; corresponding areas in GIS software. The annual rates of Hussain et al. 2013) was carried out using maximum urban area expansion (UAE) for the periods: 1973–2000, likelihood algorithm method (MLC) for each image sep- 2000–2015, and 1973–2015 are calculated for Bahir Dar arately to test the accuracy assessment of the classifica - and Hawassa using the following relationship (Valdkamp tion. Stratified random method is used for land use/land et al. 1992; Mohan et al. 2011) in a modified form: cover representation extracted from satellite images over UAx + n − UAx classes of the area. Past and recent studies have identi- ∗ 100 (7) n ∗ UAx fied image differences as being the most accurate change detection technique. where: UAx + n and UAx are the urban area in Ha at time The accuracy assessment was done based on ground x + n and x, respectively, and n is the interval of the cal- truth data and visual interpretation using 100 points culating period (in years). (Butt et al. 2015). Statistical error matrices together with In this study we also used, land consumption rate nonparametric Kappa index were used for comparison (LCR) as an index to evaluate the progressive spatial of reference data and classification result (Robertson expansion of urban areas. The land consumption rate and King 2011; Dabboor et al. 2014). The producer accu - (LCR) is computed using the following formula (Fanan racy, user accuracy, overall accuracy, and Kappa coeffi - et al. 2011; Sharma et al. 2012): cient were calculated for the classified map of 1973, 2000 UA and 2015 based on the formula given by Congalton and (8) P Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 6 of 15 Fig. 2 a Land use/land cover patterns of Bahir Dar city (1973–2015). b Land use/land cover patterns of Hawassa city (1973–2015) where: UA is area of the city (ha) and P is the population. The classification scheme was created on the basis of the cover types in the study areas that were present in large Results quantities. These were the classes that were extracted as Land use/land cover change and urban expansion thematic classes from the images and for which area sta- The major land cover areas presented in the images for tistics were generated at local situation. The matrix indi - Bahir Dar and Hawassa cities are given in Table 3a, b. cates the amount of land in hectares and percentage of Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 7 of 15 Table 2 Accuracy assessment for classified images of Bahir Dar and Hawassa Bahir Dar Hawassa Reference year Classified image Overall classifi‑ Overall kappa Reference year Classified image Overall classifi‑ Overall kappa cation accuracy coefficient cation accuracy coefficient (%) (%) 1973 Landsat 4 MSS 86.75 0.75 1973 Landsat 4 MSS 87.37 0.85 2000Landsat 7 TM 95.5 0.93 2000 Landsat 7 TM 96.0 0.91 2015 Landsat 8 OLI 98.0 0.96 2015 Landsat 8 OLI 95.45 0.92 Multi‑spectral Scanner Thematic Mapper Operational Land Imager Table 3 LULCC pattern and change in Bahir Dar (1973–2015) and Hawassa (1973–2015) LULC type 1973 2000 2015 Change Ha % Ha % Ha % 1973–2000 2000–2015 1973–2015 Ha % Ha % Ha % Bahir Dar Water body 3742.3 14.73 3710.0 14.60 3700.2 14.56 − 31.7 − 0.12 − 10.2 − 0.04 − 42.1 − 0.16 Vegetation 10,111.8 39.79 6447.1 25.37 2583.2 10.166 − 3664.7 − 14.42 − 3863.9 − 15.21 − 7528.6 − 29.63 Green space in built‑up area 98.7 0.4 1762.9 6.9 2841.5 11.2 1664.2 6.55 1078.6 4.24 2742.8 10.79 Cropland 11,415.3 44.9 12,096.9 47.6 13,504.6 53.1 681.6 2.68 1407.7 5.54 2089.3 8.22 Urban built‑up area 85.3 0.3 634.6 2.5 2739.6 10.8 549.4 2.16 2105.0 8.28 2654.3 10.44 Total 25,411.4 100.0 25,411.4 100.0 25,411.4 100.0 0.0 0.0 0.0 0.0 0.0 100.0 Hawassa Water body 1512.6 36.37 1514.9 36.43 1561.4 37.5 2.3 0.06 46.5 1.16 48.8 1.17 Vegetation 744.6 17.91 389.9 9.38 169.5 4.1 − 354.7 − 8.53 − 220.5 − 5.30 − 575.1 − 13.83 Green space in urban area 343.0 8.2 457.5 11.0 1049.6 25.2 114.5 2.75 592.2 14.24 706.6 16.99 Cropland 1281.9 30.8 1020.8 24.5 117.1 2.8 − 261.1 − 6.28 − 903.7 − 21.73 − 1164.8 − 28.01 Urban built‑up area 276.5 6.6 775.5 18.6 1261.0 30.3 499.0 11.99 485.5 11.67 985.0 23.67 Total 4158.6 100.0 4158.6 100.0 4158.6 100.0 0.0 0.0 0.0 0.0 0 0.0 land use/land cover class changed to other type. The val - no significant change (Table 3a, b). The changes in green ues were presented in terms of hectares and percentages spaces and urban area expansion are presented in detail. as stated in formula 1 and 2. The persistence values are the values which mean unchanged amount. The gain values computed by sub The data presented in Table 3a shows that in 1973, - the vegetation cover in Bahir Dar was 40% and this was tracting the persistence value from the total area of final reduced to 25% in 2000 and 10% in 2015. On the other year and the loss value also computed by negative of hand, crop land increased from 45% in 1973 to 48% in subtracting the persistence value from the total area of 2000 and further to 53% in 2015. This change in land the initial year. Table 4 presents the persistence, gains, cover could indicate a shift from vegetation to crop - losses and net changes of different land use and land land use. Table 3b depicts that in Hawassa both veg- cover. Accordingly, in Hawassa, water body has shown a etation and crop land showed a decline. The vegetation higher persistence accounting for 75.7% while cropland has shown a higher loss (55%). In addition, the land cover cover declined from 18% in 1973 to 9% in 2000 and 4% type which persisted least is vegetation (4.8%) and the in 2015. Similarly crop land declined from 31% in 1973 land cover with least loss is water body (0.1%). In Bahir to 25% in 2000 and 3% in 2015. It should be noted that Dar, the land cover type with the highest persistence is the water body located in the study areas namely Lake cropland (60%) and that with the highest loss is vegeta Tana in Bahir Dar and Lake Hawassa in Hawassa showed - tion (70%). Built up area has shown low persistence and Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 8 of 15 Table 4 Percentage of LULCC in (ha) in Hawassa and Bahir Dar during (1973–2015) LULC types Hawassa Bahir Dar Persistence Gains Losses Net change Persistence Gains Losses Net change Ha % Ha % Ha % Ha % Ha % Ha % Ha % Ha % Urban built up area 175.5 8.8 1085.6 50.16 − 101.0 − 4.7 985.5 45.5 42.5 0.3 2697.2 22.51 − 42.9 − 0.4 2654.3 22.2 Vegetation 96.6 4.8 72.9 3.368 − 648.0 − 29.9 − 575.1 − 26.6 1730.1 12.9 853.1 7.119 − 8381.7 − 69.9 − 7528.6 − 62.8 Water body 1510.2 75.7 51.2 2.364 − 2.4 − 0.1 48.7 2.3 3512.8 26.2 229.3 1.914 − 187.5 − 1.6 41.8 0.3 Green space in urban area 116.5 5.8 933.1 43.12 − 226.5 − 10.5 706.6 32.7 55.6 0.4 2785.9 23.25 − 43.1 − 0.4 2742.9 22.9 Cropland 95.6 4.8 21.4 0.99 − 1186.2 − 54.8 − 1164.8 − 53.8 8086.9 60.2 5417.6 45.21 − 3328.0 − 27.8 2089.6 17.4 Total 1994.4 47.96 2164.2 52.04 − 2164.2 − 52.04 0.0 0.0 13,428.0 52.84 11,983.2 47.16 − 11,983.2 − 47.16 0.0 0.0 Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 9 of 15 Table 5 Horizontal urban expansion of Bahir Dar and Hawassa (1973–2015) Year Bahir Dar Hawassa Urban area (ha) Change (Ha) Urban area (ha) Change (Ha) 1973–2000 2000–2015 1973–2015 1973–2000 2000–2015 1973–2015 1973 184 2214 619 613 2000 2398 3181 1232 1079 2015 5579 5395 2311 1692 −1 % Change (ha year ) 44.56 8.84 69.81 3.67 5.84 6.51 Horizontal urban expansion includes urban built up area and green spaces with in built‑up area (see Fig. 3a, b) losses but higher gain. In Bahir Dar and Hawassa cit- than time gap within 2000–2015 (15 years). Similarly, ies the land cover types which gained more are built-up the data in Table 5 show that the rates of urban expan- area (50%) and crop land (45%) respectively (Table 4 and sion in Hawassa for the periods 1973–2000, 2000–2015, Fig. 2a, b). and 1973–2015 were 4, 6, and 7% per year per hectares In general, the results show that 53% of Bahir Dar and respectively, implying that annual rates of urban expan- 48% of Hawassa land use/land cover remained unchanged sion is much higher for the period 2000–2015 than for over the 1973–2015 periods. On the other hand, 47% the period 1973–2000. The intensification of built-up of Bahir Dar and 52% of Hawassa land use/land cover area in Bahir Dar and Hawassa for the past four dec- changed during 1973–2015. This indicates that there is a ades (1973–2015) was 70 and 7% per year per hectares higher change of LULCC in Hawassa than in Bahir Dar in respectively. This shows that 5395 and 1692 ha of land the last four decades (Table 4 and Fig. 2a, b). were converted to urban uses from other land cover types The driving factors for this rapid LULCC are the rapid in Bahir Dar and Hawassa respectively. Bahir Dar has a growth of urban population and the horizontal expan- much higher average annual rate of urban expansion than sion of urban areas (see below). In line with this, the Hawassa. This could be because Bahir Dar has a larger population in Bahir Dar has more than tripled between boundary than Hawassa, and this might have prompted 1994 and 2017 (96,140 in 1994 and 348,429 in 2017) and the rapid conversion of other land uses to urban land use quadrupled in Hawassa between 1994 and 2017 (69,169 (Fig. 3a and b). On the other hand, although the politi- in 1994 and 315,267 in 2017). The Landsat images analy - cal boundary is small, Hawassa is relatively close to Addis sis reveals, however, that land cover change is faster since Ababa, the capital city, and has higher potential to attract 2000 than it was during the 1990s. The following discus - businessmen and investments which are very important sion focuses on two types of land use changes namely the contributors to the fast growth of urban areas. urban expansion and the green space in both cities. Using the technique presented in formula 8, the LCR result for Bahir Dar is 0.002, 0.003 and 0.015 for the years Urban expansion 1973, 2000 and 2015 respectively. Likewise, the LCR for Following the technique in formula 7, the annual rates of Hawassa is 0.023, 0.005 and 0.009 for the years 1973, urban expansion are analyzed from two perspectives. The 2000 and 2015 respectively. It can be seen that the LCR first is the expansion in LULCC as a result of the sprawl result is in accordance with LULCC result and is higher of each city, which is the horizontal expansion while the for Bahir Dar except for the year 1973 and 2000. second is the changes in LULCC that occurred within the 1973 boundaries of the cities during the period 1973– Changes in green space in built‑up area 2015. This type of change is referred to as intensification Table 1 defined green space in urban area as an area of increases in the density of dwellings and other infrastruc- grass, trees, or other vegetation set apart for recreational ture within existing built-up areas. or aesthetic purposes inside urban built environment. The data presented in Table 5 show that the annual It includes urban parks, greenery, roundabouts, public rates of urban expansion for Bahir Dar in the period squares and plaza, open spaces, medians and sport fields. 1973–2000, 2000–2015, and 1973–2015 were 45, 9, It is clear that this type of land use is created by the city and 70% per year per hectares respectively. It is thus government as part of its land use planning schemes. clear that urban expansion was much higher during the The data presented in Table 3a, b clearly show that period 1973–2000 than 2000–2015. This could be due green spaces in Bahir Dar and Hawassa have increased to the time gap within 1973–2000 (27 years) is longer significantly between 1973 and 2015. In Bahir Dar, Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 10 of 15 Fig. 3 a Horizontal expansion of Bahir Dar city (1973–2015). b Horizontal expansion of Hawassa city (1973–2015) Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 11 of 15 green space increased by 2742.8 ha while it increased by land use/land cover classes 1973–2015 due to intensifica - 706.6 ha in Hawassa between the years 1973 and 2015. tion of the built-up area. The result shows that built-up The percentage increase is much higher for Bahir Dar increased by 22% in Bahir Dar, whereas the cropland land than for Hawassa. This is because of both the small base increased by net-change of 17% at the expense of vegeta- and the higher additions of green spaces in Bahir Dar tion which showed a decrease by net-change of 63%. A than in Hawassa. Hawassa, however, has a higher percent similar analysis for the periods 1973–2015 in Hawassa increase for the period 2000–2015 than Bahir Dar. The revealed that net-change in built-up area is 45%. During land cover types mostly changed to green infrastructure the entire study period (1973–2015), crop land and veg- are vegetation and cropland. etation decreased by net-change of 54 and 27% respec- In comparing green space and built-up area expansion, tively (Table 4). Green space is also one of the land use it can be seen that built-up area in Bahir Dar increased by categories that showed a rapid increase with a higher- 10% in 1973–2015 and this is proportional to green space level gain than loss in both cities. increment of 11% of green space for the same period. The built-up area in Hawassa however increased (24%) Discussion more than green space (17%) though both have a rising This study showed that land use/land cover is imperative tendency. for understanding the GI conditions of urban areas. Land A comparative perspective between Hawassa and Bahir use/land cover can be used for planning and monitoring Dar shows that green space increment in Hawassa (33%) the status of GI. On the other hand, GI study requires is by far more than Bahir Dar (23%) (Table 4). The reason land use/land cover change detection in order to under- could be that Hawassa has witnessed a decline in both stand GI within the setting of other land use/land covers. vegetation and crop land which must have contributed to Some researches for instance, Li et al. (2015) indicated built-up and green spaces in the city. On the other hand, that LULCC can be an important indicator to link GI and in Bahir Dar though vegetation has decreased, crop land human activities in urban ecosystems. Liu et al. (2014) has increased there by competing with the increase in also examine LULCC and urbanization effect on urban built up and green space. Therefore care should be taken environment. Although, there are studies on farm land to conserve these lands. effects of LULCC or urbanization (Pauleit et al. 2005), this study has made the first attempt to explore the com - Land use land cover types most affected bined effect of LUCC and GI under the rapid urbaniza - The land cover proportions obtained from the succes - tion on farm land in fast growing cities of Bahir Dar and sive (enhanced) classifications revealed that in 1973, Hawassa 1973–2015. Bahir Dar was dominated by crop land (45% of total The green infrastructure concept has come into the area), followed by vegetation (40%) (Table 3a). However, table of discussion in the last few decades and is used for after a quarter of a century, in 2000, vegetation occupied urban green environment improvement (Tzoulas et al. only 25% of the total area, and crop land increased and 2007). In this study, the foregoing data showed that green occupied 48%. The change was further intensified after spaces in both study cities have increased. The increasing 2000 as vegetation was reduced to 10% and crop land trend of green infrastructure in Bahir Dar and Hawassa was increased to 53%. Moreover, as presented in Table 4 during the period 1973–2015 was due to continued and urban built-up area increased by 45% ha per year during drastic increment of built-up areas at the cost of other the period 1973–2015. land cover types (vegetation and cropland). Some stud- The data presented in Table 3b indicates that in 1973, ies (Noor et al. 2013) indicated that the issue of green Hawassa was dominated by crop land (31%) and veg- infrastructure has become major concern throughout the etation (17%). However, in 2000 or after a quarter of world particularly among developing countries due to the a century, vegetation occupied only 9% and crop land obvious negative impacts which occurred as the result decreased and covered 25%. The urban built-up area on of loss of green infrastructure in terms of visual qual- the other hand increased alarmingly by 30%, with an ity, environmental quality and health quality with in fast expansion of 7% ha per year during the periods of 1973– growing cities and towns. 2015. This implies that in both cities, urban-built up area A study by Luck and Wu (2002) recognized that is the land use type that showed marked increase while urbanization is one of the most important driving forces crop land and vegetation land have different trends. The behind LULCC in Jinan city (China). Kong and Nak- Landsat images analysis confirmed that the major land agoshi (2006) also reported that the driving forces are cover conversions were from the vegetation cover classes the policies that affect the development and management to crop land and built-up classes (Fig. 2a, b). This is fur - of urban GI. However, Byomkesh et al. (2012) noted, the ther confirmed by examination of net changes of the five causes of changes in GI, among other things, are rapid Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 12 of 15 population growth driven by rural–urban migration, eco- 2011). Horizontal expansion has strong effects on other nomic development and a lack of awareness among city urban land uses, such as crop/agricultural land, green managers and city dwellers. In addition, the use of politi- space, and forest lands (Mohan et al. 2011). cal power to influence the illegal conversion and leasing The results from the analysis of the Landsat images of GI, a lack of appropriate rules and regulations to pro- show that for the two cities the land cover types that have tect urban GI, limited budget for the management and significantly contributed to gains for built-up area and maintenance of urban GI, are also factors that contrib- the corresponding green space are crop land and veg- ute to GI change. In their study about analysis of LULCC etation. These two-land use/land cover types have fed and urban expansion of Nairobi city, Mundia and Aniya built-up areas and green spaces while they show a drastic (2005) noted the analysis of LULCC shows that the big- decline in their coverage. This is consistent with previous gest challenge to city planners’ is perhaps to maintain an research (Xu et al. 2000) crop lands are under great pres- internal balance between economic activity, population sure from rapid urban expansion. As it is explained in growth, infrastructure and services not limiting impacts tables above in both cities the expansion rate is fast and on the natural environment. In order to maintain eco- many agricultural lands are changed into urban areas. logical balance and proper functioning of ecosystems, Severe arable land loss will have a significant impact on comprehensive GI planning and management strategy the county’s further agricultural development. Obvi- needs to be formulated. Educating the people to increase ously, dynamical monitoring of the expansion of urban their awareness about the role and importance of GI for areas is valuable for the sustainable development of the healthy environment is necessary. It is worth mentioning country. Rana (2011) noted rapid urbanization is always that comprehensive land use planning would contribute characterized by spatial extension in the periphery, which to enhance GI and sustainability and livability of urban leads to exploitation of forest and crop land. This could development. be because of limited capacity of planning. The city This study has presented the LULCC and dynamics of authorities are facing huge lack in skilled manpower and urban expansion which is demonstrated by the interplay sufficient resources to reach the detail plan stages (Islam between biophysical, location site and socio-economic 2002; Shafi 2003). characteristics in shaping the growth of both cities. The It is important to note that urban expansion and the spatial expansion of both Bahir Dar and Hawassa cities is loss of crop land have impacts on the surrounding farm- very rapid during the last 10–15 years. The driving forces ers and the nearby water bodies in the study area. With to this urban expansion resulted from population growth, regard to famers, loss of crop land and the associated economic reform and industrialization (Meyer and urban expansion give rise to changes in the livelihood Turner 1992; Morrisette 1992; Rockwell 1992; Sander- of farmers as they derive reduced income from farming son 1992). Increase in investment brings fast economic (Haregeweyn et al. 2012). In relation to water bodies, the development which leads accelerated urban area expan- small decrease noted in both cities between 1973 and sion because the development of the industrial parks 2015 is associated with a retreat of the lakes caused by (Grubler 1992) in both cities. Industrial development is siltation and the subsequent use of this land for built- a major driving force for urbanization (Xu et al. 2000). up areas. Field observation and satellite images analysis Expansion direction is also necessary in city management verify this because there are clear indicators of the retreat and study of LULCC. The direction of urban expansion is of Lake Tana and Lake Hawassa. Other related studies importantly controlled by the topographical and physical (Gashaw and Fentahun 2014; Wondrade and Tveite 2014; factors. Urban expansion directions and land use conver- Teshale and Bantider 2015; Minale and Belete 2017) con- sions analysis indicates that deliberate planning is largely ducted in different parts of Ethiopia reported that the important in Bahir Dar and Hawassa urbanization pro- life of both artificial and natural lakes is threatened by cess. Bahir Dar city expands towards South, West and a high sedimentation rate, with the sediment primarily North-East but no more expansion towards the North being delivered from agricultural watersheds. At the city because of Lake Tana and towards the South-East due to level, the results revealed that different anthropogenic bezawit ridge. Hawassa city expands towards East, North activities had significantly affected the urban green infra - and South-East but not towards the West and South structure composition and configuration within the inner because of Lake Hawassa and amoragedel ridge respec- cities of both Bahir Dar and Hawassa. tively. These areas of both cities are considered as green Our findings are helpful for policy makers to bet - belts of corresponding cities. The horizontal expansion of ter understand and address these complex relationships urban and suburban areas requires more land and drives between urbanization, LUCC, and GI. It is important to the conversion of surrounding rural areas to urban land develop improved land-use policies that balance LUCC, use/land cover (Farooq and Ahmad 2008; Mohan et al. GI proportion and urbanization. The findings of this Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 13 of 15 Author details research have not only important policy implications Department of Geography and Environmental Studies, University of Gondar, for urban GI design and management, but also provide P.O. Box 196, Gondar, Ethiopia. Department of Geography and Environmental important information for other research areas such as Studies, Addis Ababa University, P.O. Box 1176, Addis Ababa, Ethiopia. urban environment and ecology. Acknowledgements We would like to thank the anonymous reviewers and the editor for their Conclusions genuine comments and corrections which helps the paper to be in its present form. Special thanks to Ethiopian Mapping Agency (EMA) for accessing the The LULCC dynamics largely depend on dynamic rela - satellite imageries. tionships not only natural factors but also among popu- lation and policy/institutional factors. In this study we Competing interests The authors declare that they have no competing interests. noted the spatiotemporal trends of urban land use/land cover and an aspect of green infrastructure change. Availability of data and materials Change detection is important to understand the magni- Not applicable. tude and direction of change in any land use/land cover Ethics approval and consent to participate category in general and in green spaces in particular. Not applicable. Our result revealed that green infrastructure defined Consent for publication as urban parks, open spaces, greenery, roundabouts, Not applicable. public squares and plaza, medians and sport fields have increased in both cities during the period 1973–2015. Funding The authors would like to thank Addis Ababa University for financial support Such increase is believed to be associated with urban for this research for both researchers and University of Gondar for financial expansion since the latter have increased in both cities. support to the first researcher. The mechanism is the implementation of land use plan - ning at city level in order to cope up with the increasing Publisher’s Note urban expansion. The two cities could thus be taken as Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. exemplary to other cities and towns in Ethiopia since the increase in green spaces is closely related to sustainable Received: 19 January 2018 Accepted: 24 April 2018 urban development. In recent years there has been growing concern among planners about the green infrastructure in cities. In addi- tion, policy makers and stakeholders should also decide References on how to use the land at present and in the future. Abebe MT, Megento TL (2016) The city of Addis Ababa from ‘Forest City’ to Therefore, LULCC policymaking processes should aim to ‘Urban Heat Island’: assessment of urban green space dynamics. J Urban balance green infrastructure and other types of land use/ Environ Eng 10(2):254–262 Amsalu A, Stroosnijder L, de Graaf J (2007) Long‑term dynamics in land land cover for sustainable urban development. Generally, resource use and the driving forces in the Beressa watershed, highlands we can imply that urban land use/land cover have impor- of Ethiopia. J Environ Manag 83:448–459 tant effects on the urban green infrastructure system. Anderson JR, Hardy EE, Roach JT, Witmer RE (1976) A land‑use and land‑ cover classification system for use with remote sensor data. US Geological In conclusion, LULCC dynamics and GI analyses are Survey Professional Paper 964, Washington, DC imperative for understanding the landscape ecological Ayalew D, Kassahun D, Woldetsadik M (2012) Detection and analysis of land‑ conditions of urban environment. This study revealed use and land‑ cover changes in the Midwest escarpment of the Ethiopian Rift Valley. J Land Use Sci 7(3):239–260 that the vegetation and crop land are decreasing over Butt A, Shabbir R, Ahmad SS, Aziz N (2015) Land use change mapping and the course of time due to the increasing pace of built-up analysis using Remote Sensing and GIS: a case study of Simly watershed, area. This activity is causing the destruction of landscape Islamabad, Pakistan. Egypt J Remote Sens Space Sci 2015(18):251–259 Byomkesh T, Nakagoshi N, Dewan AM (2012) Urbanization and green space ecological processes and the biodiversity in urban areas. dynamics in Greater Dhaka, Bangladesh. Landsc Ecol Eng 8:45–58 Therefore, a comprehensive urban land use planning Chen G, Hay GJ, Carvalho LMT, Wulder MA (2012) Object‑based change detec‑ and GI management strategy should be implemented for tion. Int J Remote Sens 33(14):4434–4457 Congalton RG, Green K (2009) Assessing the accuracy of remotely sensed data: proper functioning of the urban environment. principles and practices, 2nd edn. Taylor & Francis, Baco Raton CSA (1994) Statistical report. Addis Ababa, Ethiopia Authors’ contributions CSA (2007) Statistical report. Addis Ababa, Ethiopia KG has conceived of the study and made contributions in the design, data col‑ CSA (2017) Statistical abstract. Addis Ababa, Ethiopia lection and analysis, interpretation of results and revisions of the manuscript. Dabboor M, Howell S, Shokr M, Yackel J (2014) The Jeffries–Matusita distance TG‑E has participated in the sequence alignment and critical commenting of for the case of complex Wishart distribution as a separability criterion for the draft manuscript. He also participated in its design and coordination, and fully polarimetric SAR data. Int J Remote Sens 35(19):6859–6873 helped to draft and edits the manuscript. Both authors read and approved the Fanan U, Dlama KI, Olusey IO (2011) Urban expansion and vegetal cover loss in final manuscript. and around Nigeria’s Federal Capital City. J Ecol Nat Environ 3(1):1–10 Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 14 of 15 Farooq S, Ahmad S (2008) Urban sprawl development around Aligarh city: Mell C (2014) Aligning fragmented planning structures through a green infra‑ a study aided by satellite remote sensing and GIS. J Indian Soc Remote structure approach to urban development in the UK and USA. Urban For Sens 36:77–88 Urban Green 13(2014):612–620 Feoli E, Vuerich L (2002) Processes of environmental degradation and Meshesha D, Tsunekawa A, Tsubo M (2010) Continuing land degradation opportunities for rehabilitation in Adwa, Northern Ethiopia. Landsc Ecol and its cause‑ effect in Ethiopia’s Central Rift Valley. Land Degrad Dev 17(4):315–325 23(2):130–143 Foody GM (2002) Status of land covers classification accuracy assessment. Meyer WB, Turner BL (1992) Human population growth and global land‑use/ Remote Sens Environ 80:185–201 cover change. Annu Rev Ecol Syst 23(1):39–61 Gashaw T, Fentahun T (2014) Evaluation of land use/land cover changes in east Miller J, Hobbs R (2002) Conservation where people live and work. Conserv of lake Tana, Ethiopia. J Environ Earth Sci 4(11):49–53 Biol 16:330–337 Grubler A (1992) Technology and global change: land‑use, past and present. Minale AS, Belete W (2017) Land use distribution and change in lake Tana sub In: Meyer WB, Turner BL II (eds) Global land‑use/land‑ cover change. basin. In: Stave K, Goshu G, Aynalem S (eds) Social and ecological system Boulder, AIES dynamics. AESS interdisciplinary environmental studies and sciences Hailemariam SN, Teshome S, Teketay D (2016) Land use and land cover change series. Springer, Cham in the Bale Mountain Eco‑Region of Ethiopia during 1985 to 2015. Land Mohan M, Pathan S, Narendrareddy K, Kandya A, Pandey S (2011) Dynamics 5:41 of urbanization and its impact on land use/land cover: a case study of Haregeweyn N, Fikadu G, Tsunekawa A, Tsubo M, Tsegaye DM (2012) The megacity Delhi. J Environ Prot 2:1274–1283 dynamics of urban expansion and its impacts on land use/land cover Moroney J, Jones D (2006) Biodiversity space in urban environments: implica‑ change and small‑scale farmers living near the urban fringe: a case study tions of changing lot size. Aust Plan 43(4):22–27 of Bahir Dar, Ethiopia. Landsc Urban Plann 106(2012):149–157 Morrisette PM (1992) Developing a political typology of global patterns of He C, Shi P, Xie D, Zhao Y (2010) Improving the normalized difference built ‑up land and resource use. In: Meyer WB, Turner BL II (eds) Global land‑use/ index to map urban built‑up areas using a semiautomatic segmentation land‑ cover change. Boulder, AIES approach. Remote Sens Lett 1(4):213–221 MoUDH (2015) Ethiopia national urban green infrastructure standard, Addis Hu H, Liu W, Cao M (2008) Impact of land use and land cover changes on eco‑ Ababa, Ethiopia system services in Menglun, Xishuangbanna, South west China. Environ Msoffe FU, Kifugo SC, Said MY, Neselle MO, Gardingen PV, Reid RS, Ogutu JO, Monit Assess 146:147–156 Herero M, de Leeuw J (2011) Drivers and impacts of land‑use change in Hung M, Wu Y (2005) Mapping and visualizing the Great Salt Lake landscape the Maasai Steppe of northern Tanzania: an ecological, social and political dynamics using multi‑temporal satellite images, 1972–1996. Int J Remote analysis. J Land Use Sci 6(4):261–281 Sens 26:1815–1834 Mundia CN, Aniya M (2005) Analysis of land use/cover changes and urban Hussain M, Chen D, Cheng C, Wei H, Stanley D (2013) Change detection from expansion of Nairobi city using remote sensing and GIS. Int J Remote remotely sensed images: from pixel‑based to object ‑based approaches, Sens 26(13):2831–2849 ISPRS. J Photogramm Remote Sens 80(2013):91–106 Murat H, Selçuk R, Mustafa A (2006) Detection of spatial‑temporal changes of Islam N (2002) The Bangladesh urban environment (Editorial Notes). CUS bul‑ development potential of Aksaray city using remote sensing and GIS. In: letin on urbanization and development, vol 43. Center for urban studies, Shaping the change, XXIII FIG congress, Munich, Germany, October 8–13 Dhaka Muriuki G, Seabrook L, McAlpine C, Jacobson C, Price B, Baxter G (2011) Land Kambites C, Owen S (2006) Renewed prospects for green infrastructure plan‑ cover change under unplanned human settlements: a study of the ning in the UK. Plan Pract Res 21(4):483–496 Chyulu hills squatters, Kenya. Landsc Urban Plan 99(2011):154–165 Kong F, Nakagoshi N (2006) Spatial‑temporal gradient analysis of urban green Muttitanon W, Tripathi N (2005) Land use/cover changes in the coastal zone spaces in Jinan, China. Landsc Urban Plan 78:147–164 of Ban Don bay, Thailand using Landsat 5 TM data. Int J Remote Sens Lei C, Zhu L (2017) Spatio‑temporal variability of land use/land cover change 26:2311–2323 (LULCC) within the Huron river: effects on stream flows. Climate Risk NMA (2013) National meteorological station report. Federal democratic repub‑ Manag 19(2018):35–47 lic of Ethiopia, Addis Ababa Li W, Bai Y, Zhou W, Han C, Han L (2015) Land use significantly affects the Noor NM, Abdullah A, Manzahari H (2013) Land cover change detection analy‑ distribution of urban green space: case study of Shanghai, China. J Urban sis on urban green area loss using GIS and remote sensing techniques. J Plan Dev 141(3):A4014001 Malays Inst Plan 2013(11):125–138 Liu Y, Huang X, Yang H, Zhong T (2014) Environmental effects of land‑use/ Pauleit S, Ennos R, Golding Y (2005) Modeling the environmental impacts cover change caused by urbanization and policies in Southwest China of urban land use and land cover change—a study in Merseyside, UK. Karst area—a case study of Guiyang. Habitat Int 44(2014):339–348 Landsc Urban Plan 71(2–4):295–310 Long H, Tang G, Li X, Heilig GK (2007) Socio‑ economic deriving forces of land‑ Phan DU, Nakagoshi N (2007) Analyzing urban green space pattern and eco‑ use change in Kunshan, the Yangtze River Delta economic area of China. network in Hanoi, Vietnam. Landsc Ecol Eng 3:143–157 J Environ Manag 83(3):351–364 Puyravaud JP (2003) Standardizing the classification of annual rate of defor ‑ Lu D, Weng Q (2007) A survey of image classification methods and tech‑ estation. For Ecol Manag 177(1–3):593–596 niques for improving classification performance. Int J Remote Sens Radoux J, Bogaert P, Fasbender D, Defourny P (2011) Thematic accuracy assess‑ 28(5):823–870 ment of geographic object‑based image classification. Int J Geographical Lu D, Mausel P, Brondízio E, Moran E (2004) Change detection techniques. Int J Inf Sci 25(6):895–911 Remote Sens 25(12):2365–2401 Rana MP (2011) Urbanization and sustainability: challenges and strategies Luck M, Wu J (2002) A gradient analysis of urban landscape pattern: a case for sustainable urban development in Bangladesh. Environ Dev Sustain study from the Phoenix metropolitan region, Arizona, USA. Landsc Ecol 13:237–256 17:327–339 Robertson LD, King DJ (2011) Comparison of pixel‑ and object ‑based classifi‑ Mansor M, Said I, Mohamad I (2012) Experiential contacts with green infra‑ cation in land cover change mapping. Int J Remote Sens 32(6):1505–1529 structure’s diversity and well‑being of urban community. Asian J Environ Rockwell R (1992) Culture and cultural change as driving forces in global Behav Stud 49:33–48 land‑use/cover changes. In: Meyer WB, Turner BLI II (eds) Global land‑use/ Mas JF (1999) Monitoring land‑ cover changes: a comparison of change detec‑ land‑ cover change. Boulder, AIES tion techniques. Int J Remote Sens 20(1):139–152 Sadidy J, Firouzabadi P, Entezari A (2009) The use of radar sat and land sat Masser I (2007) Managing our urban future: the role of remote sensing and image fusion algorithms and different supervised classification methods geographic information systems. Habitat Int 25:503–512 to use map accuracy—case study. Sari lain‑Iran. http://www.isprs.org/ McIntyre N, Knowles‑ Yánez K, Hope D (2000) Urban ecology as an interdiscipli‑ procedding/XXXVI/5‑ C55/papers‑Sadidy_javad.pdf . Accessed 19 Oct nary field: differences in the use of urban between the social and natural 2015 sciences. Urban Ecosyst 4:5–24 Sanderson S (1992) Institutional dynamics behind land use change. In: Meyer WB, Turner BL II (eds) Global land‑use/land‑ cover change. Boulder, AIES Gashu and Gebre‑Egziabher Environ Syst Res (2018) 7:8 Page 15 of 15 Shafi SA (2003) Use of planning tools as a guide to balanced urban develop ‑ Valdkamp E, Weitz A, Staritsky I, Huising E (1992) Deforestation trends in the ment. CUS Bulletin on urbanization and development, vol 44. Center for Atlantic zone of Costa rica: a case study. Land Degrad Rehabil 3:71–84 urban studies, Dhaka Varshney A (2013) Improved NDBI differencing algorithm for built ‑up regions Sharma L, Pandey PC, Nathawat MS (2012) Assessment of land consumption change detection from remote‑sensing data: an automated approach. rate with urban dynamics change using geospatial techniques. J Land Remote Sens Lett 4(5):504–512 Use Sci 7(2):135–148 Wondrade N, Tveite DH (2014) GIS based mapping of land cover changes Song X, Chang K, Yang L, Scheffran J (2016) Change in environmental benefits utilizing multi‑temporal remotely sensed image data in Lake Hawassa of urban land use and its drivers in Chinese cities, 2000–2010. Int J Envi‑ Watershed, Ethiopia. Environ Monit Assess 186(3):1765–1780 ron Res Public Health 13:535 Xu H, Wang X, Xiao G (2000) A remote sensing and GIS integrated study on Taddese G (2001) Land degradation: a challenge to Ethiopia. Environ Manag urbanization with its impact on arable lands: Fuqing city, Fujian province, 27(6):815–824 China. Land Degrad Dev 11:301–314 Teferi E, Bewket W, Belay S (2016) Eec ff t of land use land cover on selected Yang J, Huang C, Zhang Z, Wang L (2014) The temporal trend of urban green soil quality indicators in the head water area of the Blue Nile basin of coverage in major Chinese cities between 1990 and 2010. Urban For Ethiopia. Environ Monit Assess 2016(188):1–12 Urban Green 13(2014):19–27 Teshale RR, Bantider A (2015) Land use land cover dynamics in Hawassa tabor Yuan F, Sawaya KE, Loeffelholz BC, Bauer ME (2005) Land cover classifica‑ and alemura ridge and its surroundings in the case of SNNPR, Ethiopia, tion and change analysis of the twin cities (Minnesota) metropolitan Doctoral dissertation, Haramaya university area by multitemporal Landsat remote sensing. Remote Sens Environ Tewolde MG, Cabral P (2011) Urban sprawl analysis and modeling in Asmara, 98(2–3):317–328 Eritrea. Remote Sens 2011(3):2148–2165 Zhang H, Qi ZF, Ye XY, Cai YB, Ma WC, Chen MN (2013) Analysis of land use/ Thompson M (1996) Standard land cover classification scheme for remote land cover change, population shift, and their effects on spatiotemporal sensing application in South Africa. S Afr J Sci 92:34–42 patterns of urban heat islands in metropolitan Shanghai, China. Appl Tsegaye D, Moe S, Vedeld P, Aynekulu E (2010) Land‑use/cover dynam‑ Geogr 44:121–133 ics in Northern afar rangelands, Ethiopia, agriculture. Ecosyst Environ Zhou W, Troy A (2008) An object‑ oriented approach for analyzing and 139(2010):174–180 characterizing urban landscape at the parcel level. Int J Remote Sens Tzoulas K, Korpela K, Venn S, Yli‑Pelkonen V, Ka´zmierczak A, Niemela J, 29(11):3119–3135 James P (2007) Promoting ecosystem and human health in urban Zoran M (2007) Urban environmental quality assessment by satellite and areas using green infrastructure: a literature review. Landsc Urban Plan in situ monitoring data. In: AIP conference proceedings 899, 407 81(2007):167–178 UN (2014) World urbanization prospect, department of economic and social affairs, New York
Environmental Systems Research – Springer Journals
Published: May 8, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera