Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Quantification and mapping of the spatial landscape pattern and its planning and management implications a case study in Addis Ababa and the surrounding area, Ethiopia

Quantification and mapping of the spatial landscape pattern and its planning and management... GEOLOGY, ECOLOGY, AND LANDSCAPES 2021, VOL. 5, NO. 3, 161–172 INWASCON https://doi.org/10.1080/24749508.2019.1701309 RESEARCH ARTICLE Quantification and mapping of the spatial landscape pattern and its planning and management implications a case study in Addis Ababa and the surrounding area, Ethiopia a,b a c Asfaw Mohamed , Hailu Worku and Mengistie Kindu Ethiopian Institute of Architecture, Building Construction and City Development, Addis Ababa University, Addis Ababa, Ethiopia; b c Department of Geography and Environmental studies, Hawassa University, Hawassa, Ethiopia; Institute of Forest Management, Department of Ecology and Ecosystem Management, TUM School of Life Sciences Weihenstephan, Technische Universität München, Freising, Germany ABSTRACT ARTICLE HISTORY Received 22 May 2019 Rapid urbanization in Addis Ababa and the surrounding area resulted in the rapid land use/ Accepted 27 November 2019 land cover (LULC) change that affected landscape structures and ecological functions. This study aimed at quantifying and mapping the spatial patterns of landscape structure for KEYWORDS a sustainable city region landscape conservation planning and management. GIS and statistical LULC; Landsat; remote tools were used to compute important landscape metrics. Pearson Correlation and factor sensing; city region; green analysis were also applied to reduce redundant indices and identify underlining factor of the space landscape structure by network of hexagonal area. The analysis depicted four landscape and four class-level underlined metrics. Accordingly, as the region overall landscape was character- ized by patch size and density, shape and texture (interspersion) index, the forest class also attributed by patch size and density, and shape metrics. The result shows that the region landscape planning and management schemes must emphasize on the level of patch frag- mentation and landscape complexity to maintain the natural land cover habitat functioning, the amount of ecological process and extent of human intervention. This research will help scientific base decision-making in conservation planning and management of the tropical highland urban landscape in general, and the study area in particular. 1. Introduction acceptable quality of life for their residents (Simson, 2017). Concurrently, studies were dealing more on the Urban areas, nowadays, are growing at an alarming rate identification and quantification of the ecosystem ser- and becoming more complex. Urban built-up, agricul- vices and goods generated from the city region’svegeta- ture and other anthropogenic uses take over areas tion. For instance, Livesley, McPherson, and Calfapietra which are previously covered by natural vegetation, (2016)review different researches about the urban for- particularly forest and other green spaces (Yared, est impact on the urban heat island effect, human ther- Heyaw, Pauleit, & Mengistie, 2019). Such mal comfort, soil and water pollution, urban catchment a degradation of the city regions’ vegetation and the hydrology and air pollution. Likewise, Nowak et al. rate of their ecosystem functions caused by uncon- (2008) measure urban forest contribution to pollution trolled urban growth and other anthropogenic activities removal and carbon storage and sequestration. Dobbs, have not yet given much emphasis. Researches also have Escobedo, and Zipperer (2011) also indicated soil PH shown the danger of ignoring the city region (metro- and organic matter as the most influential ecosystem politan area) vegetation. For example, Davies, services and goods generated from urban trees. Hernebring, Svensson, and Gustafsson (2008)pre- A mix of remotely sensed data, GIS and the con- sented urbanization and the changing climate as the ceptual basis of landscape ecology is helpful to assess main factor for more frequent and severe flooding, and monitor the large-scale ecology (Herold, Scepan, heat and air pollution. In fact, the notion of the sustain- & Clarke, 2002; Lausch & Herzog, 2002; Mengistie, able city brought the relationship of the social and Schneider, Döllerer, Teketay, & Knoke, 2018). The biophysical environment by highlighting the enormous concept of Landscape ecology is an interdisciplinary benefits of the urban vegetation and the challenge of its subject concerned with the interaction between the governance and management (Brown et al., 2013). The spatial pattern and ecological process (Li, Ling, idea of urban green infrastructure thus appears in rela- Cheng, & Xiao, 2001), and it is also an important tion to urban forestry considering the landscape struc- method to measure the status of the city regions’ ture planning of cities and city regions to provide an LULC configuration and related ecosystem functions. CONTACT Asfaw Mohamed asfethiop@yahoo.com Ethiopian Institute of Architecture, Building Construction and City Development, Addis Ababa University, Po. Box. 518, Addis Ababa, Ethiopia © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON). This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 162 A. MOHAMED ET AL. At the same time, the advancement of remotely sensed ecological footprint which could be even more than data and GIS technology gives rise to the concerned their immediate hinterlands. In order to cater to the experts’ ability to measure the LULC pattern and to urban immediate economic need, a huge amount of have better knowledge about its spatial configuration energy and construction material resources was and landscape structure (Mengistie, Schneider, extracted and utilized. This is also responsible for the Teketay, & Knoke, 2013). Currently, various landscape degradation of the natural support system and irrever- metrics that measure both the structural and func- sible damage and loss of critical ecosystem functions tional characteristics of the landscape were recognized (Tadesse, 2010). On the other hand, Mpofu (2013) (Malinowska & Szumacher, 2013). Different showed the sub-Saharan African context, uncontrolled researches depicted a wide range of its application. expansion of cities is followed by the degradation of Walz (2011) utilized the metrics to investigate, evalu- the ecosystem services like sinking or recycling of ate and measure the relationship between landscape polluted substances. This deterioration is mainly due structure and biodiversity. Also, Kim and Stephan to the complexity of managing and regulating the (2005) used it to show the level of species richness, system as it tends to lay across administrative bound- patterns of species dispersal and the extent of ecologi- aries. Similarly, the study area, Addis Ababa and the cal stability of the different landscapes and Jones, surrounding Oromiya Special Zone peri-urban vege- Swanson, Wemple, and Snyder (2000) analyze the tation covers and its landscape functions are suffering ability of landscape metrics to predict nutrient and with such critical problem. sediment yield to streams. Therefore, having a clear Studies were conducted on the landscape patterns knowledge and understanding of the structure of the quantification and mapping, but the urban landscapes city regions landscape pattern is an important require- of the Afro-mountain region are not well discussed. ment for management decisions on its conservation Some planning document in Ethiopia has lately and development. brought up the notion of landscape in relation to the The Ethiopian vegetation covers suffered from green infrastructure development (Ministry of Urban a wide range of deforestation and some small forest Development and Housing) (MUDHo, 2016). Mesfin patches remained in inaccessible and marginal areas (2009) also made other attempts in measuring the (Simonian et al., 2010; Fikirte, Kumelachew, landscape metrics for built-up class of Addis Ababa Mengistie, & Thomas, 2017; Habtamu et al., 2013, city only. This research, however, takes into account Temesgen, Cruz, Kindu, Turrión, & Gonzalo, 2014). the comprehensive city region (metropolitan area), At the same time, the deforestation and degradation and involves the landscape-level and class-level land- contribute high for the depletion of the area vegetal scape metrics of ecologically valuable land covers such cover ecosystem regulating service (Temesgen et al., as forest. Such quantification and mapping of the 2018). Terefe, Feyera, and Moges (2016) have shown urban landscape pattern therefore have enormous that the decline of overall ecosystem services by 68%, importance in the ecosystem conservation process of mainly because of deforestation in the study con- the city region. As the case study of Afro-mountain ducted in the central highlands of Ethiopia. region landscape, Addis Ababa and the surrounding Therefore, the protection of these valuable vegetation area characterized by rapid built-up growth and dete- covers has a significant impact on the extent of eco- riorating ecosystem conditions, thus the landscape system functions provided by the system. Wolde analysis enables to design the framework for sustain- (2013) for example, has shown the wide difference able environmental planning and management. The between a closure and communal grazing land ecosys- main objectives of this study are to (1) analyze the tem functioning in terms of carbon, total soil nitrogen study area landscape and class-level metrics and map stock and the availability of phosphorus stock particu- zonal characteristics of the landscape structure, (2) larly in northern highland areas of Ethiopia. describe the underlined landscape structure showing Specifically, in the study area, Addis Ababa and the Addis Ababa and the surrounding area overall land- surrounding Oromiya region, the problem is more scape and forest class-level structure, and (3) identify conspicuous. The city and the overall city region vege- the landscape planning and management schemes for tation and its ecosystem service’s degradation are sustainable landscape functioning. mainly caused by the huge dependence of urban and rural areas demand of their product particularly for their fuel wood consumption. For example, Solomon 2. Materials and methods (1985) relates the continuation of Addis Ababa as the 2.1. Study area capital of the country with the introduction of Eucalyptus tree after long years of wandering on the The selected study area encompasses the biophysical surrounding hilltops for their fuel wood need. The environment found within the administrative bound- implication of the cities’ physical expansion also aries of Addis Ababa and the surrounding districts of beyond its population size as they have a large Oromiya that covers about 4400 sq.km area. In fact, GEOLOGY, ECOLOGY, AND LANDSCAPES 163 the total area of the regions is 5400 sq.km while the Forest LULC type is essentially forested patches and study was conducted on an area of 4400 sq.km, where mixed woodland comprises shrubs and woody grass- the proposed hexagonal zones fully overlay on. This land. Cultivated land also is a land use with crops and area encompasses the major forest patches and other harvested land. Besides, bare land is an area that ecologically valuable LULC at the center of the coun- includes soils, quarries, and exposed rocks, and built- 0 0 try. It is geographically located between 8 38 0N to up cover mainly represents roads, parking lots and 0 0 0 0 0 0 00 buildings. Accordingly, this research applied 9 30 20N latitude and 38 28 4E to 39 6 0 E longitude a confusion matrix technique to assess the accuracy (Figure 1). The major source and potentials of Addis of the classification. The producer, user, overall accu- Ababa green spaces and other ecological functions are racy, and the Kappa coefficients were calculated to from the forest patches found in the peri-urban part of check the image classification and signature selection the city and the surrounding Oromiya districts. dependability. 2.2. Data used and LULC classification 2.3. Hexagonal area delineation and zonal During the last decade, the study of landscape metrics landscape metrics statistical computation has been used different types of remotely sensed data (Kirstein & Netzband, 2001). A number of landscape The landscape metrics evaluation was undertaken based structure studies were conducted using Landsat on the prepared LULC map. To measure the zonal images (Bruce et al., 2001; Lu & Qihao, 2004; Seto & landscape metrics, the overplaying hexagonal area Michail, 2005). We generated the LULC map of the map was prepared as it was more suitable for analyzing study area through GIS and remote sensing techni- the ecological modeling chosen by various studies ques. The main data source to classify the LULC map (Adamczyk & Tiede, 2017; Birch, Oom, & Beecham, of this study was Landsat-8 OLI, which scanned on 2007; Schindler, Poirazidis, & Wrbka, 2008). This 23rd of December 2015 with a row 168 and path 54. research, therefore, used Arc GIS patch analyst spatial The LULC classification was made by a supervised statistics by region tool that enables users to main class classification technique through a maximum likeli- and landscape-level analysis on a raster layer for indi- hood algorithm. For training samples about 60 sample vidual polygon regions like the overlying hexagonal sites of LULC type were chosen. To sharpen the sig- zones. This function also permits the analysis of the nature selection further, normalized difference vegeta- relationship between the structural characteristics of tion index (NDVI) based unsupervised classification the landscape and some identified functions or values was also used. Then, six end members (LULC types) (Rempel, Kaukinen, & Carr, 2012). Of the total area of were selected; these are water bodies, forest, mixed 54,000 sq.km, 178 full hexagonal zones were taken to woodland land, cultivated land, bare land, and built- support the equal area representation of the selected ups. Water bodies mainly include reservoirs and dams. zones statistical analysis. The network of the hexagon Figure 1. Study area map of Addis Ababa and the surrounding Oromiya. 164 A. MOHAMED ET AL. has 25 sq.km and this size designated as an optimal permit near-natural disturbance regimes (Kim & representation of the necessary details and the viability Stephan, 2005). The variation of the patch size and of the data. Thus, the entire 178 hexagonal zones were density as represented in this study by TLA, MPS, applied for landscape-level analysis and 66 of them used PSSD and PSCov, measuring the level of fragmenta- for class-level forest landscape metrics. In fact, as tion and fractal of a certain LULC type in each zone of Adamczyk and Tiede (2017) have shown, the entire the categorical map (Shi et al., 2008; Wu et al., 2000). landscape metrics may not calculate well in such zonal For instance, mean patch size (MPS) is a function of level evaluation since the specific conditions for calcu- the total area of the landscape and the patch number lating landscape metrics within statistical zones that and it is an indicator of grain. Smaller values the index considered as a small subset of the landscape. showed a higher fragmentation of the landscape Therefore, out of the 42 landscape metrics, 31 of them (Kirstein & Netzband, 2001). applied for forest class and 35 of them applied for The grid shape of the LULC arrangement reflects an landscape-level measurement. anthropogenic intervention. However, the shape land- After the patch analyst landscape metrics computa- scape metrics mainly measure how irregular a certain tion, the output further evaluated statistically. Some of LULC is, and as the habitats shape becomes more a highly correlated metrics having the coefficient of irregular it can also comprise more plant species. greater than or equal to 90%, thus excluded from Walz (2011) thus indicated how the shape affected further analysis to remove redundancy (Griffith, the number of species. Therefore, the shape complex- Martinko, & Price, 2000). As a result, 17 class-level ity used to asses LULC data as an index for species forest landscape metrics and 16 landscape-level richness. MPFD, MSI, and AWMSI as a measure of metrics were depicted for further analysis. The statis- landscape shape showed the level of LULC type irre- tical treatment of the data hase performed the normal- gularity and complexity in this research. For example, ity tested to get information about relationship MSI calculates to mean patch shape complexity and behavior between landscape metrics factor analysis when all patches are square and increase without limit using varimax normalized rotation through Shapiro- as patch shape becomes more irregular. In fact, it is the Wilk normality test. To identify the number of com- simplest and most straightforward measure of overall ponents first the principal component analysis (PCA) shape (McGarigal et al., 2002). However, it appears was evaluated. Though different criteria were sug- more meaningful when it is related to patch density gested, in this case, the number of components deter- and size measures like MPS. Since the patch area and mined by taking the engine-value greater than one as perimeter length were used to produce the shape criteria used by Kevin, Robert, and Tait (1986). Based indices. Concurrently, AWMSI calculates the com- on the calculated number of components factor ana- plexity of the patches in the landscape according to lysis was executed. Factor analysis as a data reduction their size. Circular patches have an AWMSI value technique applied for a long list of landscape metrics close to one and it increases with patch shape irregu- evaluation (Herzog et al., 2001; Lausch & Herzog, larity (Olivier, Rolo, & van Aarde, 2017). Larger 2002; Schindler et al., 2008). Thus, the highest loading patches enter more strongly in the calculation than of each factor was taken as a representative metric smaller ones do, thus greater values index is the indi- (Herzog et al., 2001; Schindler et al., 2008). In general, cator for a higher form variety of the surfaces (Kirstein for the LULC and landscape metrics computation, Arc & Netzband, 2001). GIS tools were applied, and for statistical measure- Diversity interspersion or configuration metrics are ment, R studio programing language was used. another essential measures. It is basically a spatial characteristic that reflects the spatial distribution of a certain LULC type (Farina, 2000). The analysis 2.4. Major landscape metrics categories highly dependent on the resolution of the data and the size of the study area. IJI is an example of Although many landscape metrics have been calcu- a configuration or interspersion measure, calculated lated, some of the main metrics computed for this from the relationship between the length of edge type study area have been highlighted. Accordingly, the and a total edge of the landscape, divided by the patch density and size are the important factors that number of land-use type. As a result, when the value influence a number of basic ecological processes in close to 0, the distribution of adjacencies among patch a landscape. Generally, large patches provide large types becomes increasingly uneven, and when the benefits and small patches have small supplemental value equals to 100 means all patch types are equally benefits (Cook, 2002). In addition, large patches of adjacent to all other patch types (Herzog et al., 2001). natural vegetation are an important measure of the In addition, the edge metrics reflect the level of landscape’s ecological integrity as they protect aquifers human interference or it eventually measures how and interconnected stream networks. They also enable the natural state of the environment is (Chmielewski, to bear viable populations of particular species in Kułak, & Malwina, 2014). Edge metrics again interior habitats, which provide core habitats and GEOLOGY, ECOLOGY, AND LANDSCAPES 165 considered as a representation of landscape configura- The average total core area of landscape-level mea- tion. TE is an absolute measure of the total edge length sures was about 1038 hectare and it appears about 373 of a particular patch type at the class level or of all hectare for forest class analysis. patch types at the landscape level (Hakan & Nil, 2014). Some of these landscape metrics depicted the Since this study computes the metrics based on the redundancy of a certain dimensions, thus the correla- hexagonal region, it introduced an artificial edge of tion matrix helps to remove such repetitive measure- each hexagonal area and the analysis was executed by ment. The analysis correlates the metrics and the cutting the layer with the statistical unit. The main landscape level using 35 metrics with 595 correlation landscape metrics considered in this study are shown pairs and forest class-level measures using 30 metrics in Table 1. with 435 correlation pairs and. The correlation coeffi- cient matrix of the forest class- and landscape-level metrics portrayed by Figure 3 that showed the rela- 3. Results tionship between the computed landscape metrics. To make the data suitable for factor analysis and useable 3.1. The LULC classification and statistical for further interpretation it removes some of highly analysis of landscape metrics correlated metrics that have the correlation coefficient A total of six LULC types were extracted with an greater than or equal to 0. 9. overall accuracy of 89% with a kappa coefficient of In addition, the factor analysis result depicted that 0.87 (Table 2). From the extracted 4400 sq.km area, the level of the relationship of the component factors Figure 2 depicts that 0.35% occupied by water, 7% of with each landscape metrics. The first four compo- the area is forest, 38% of the area is mixed woodland, nents of the overall landscape and forest class metrics 26% of the area is cultivated land, 20% of the area is were identified. The factor loading of each metric bare land, and the remaining 8% of the area covered by corresponded with the components, and the ortho- built-up. gonally rotated component matrix has shown in The 39 landscape metrics summary statistics are Table 4. Regarding the landscape-level analysis, while shown in Table 3, and it shows the mean, standard the first-factor component was highly correlated with division and coefficient of variation for the hexagonal PSSD, MPS, Nump and TE, the second component area. The summary gives the general pictures of the factor has a high correlation with AWMSI and PSCoV. quantitative result of the landscape metrics at forest The third component factor also was correlated with class level and, the overall landscape-level analysis. In MSI and MPFD and the fourth factor was correlated this section, the study tries to see at least some of the highly with IJI. The first four highest factor loading of representative measures of the broader landscape orthogonally rotated landscape-level metrics again metrics categories (Area, Patch Size and Density, depicted in the component matrix and the first axis Edge, Shape, Diversity and Interspersion, and Core depicted PSSD, the second axis identified AWMSI, the Area), although the measures are not applied at the third one picked MPFD and the fourth axis took IJI. landscape level, an average total landscape area of Of these, two of them were from shape landscape forest- and landscape-level metrics is equal to 2500 metrics category (AWMSI and MPFD), one from hectare, as it is considered the equal-sized hexagonal patch size and density (PSSD) and the other one is area. Also, the percentage of landscape area of forest from diversity interspersion or texture category (IJI). was about 19% with average coefficient of variation Accordingly in the forest class-level metrics, the first (90%). The average number of patches of landscape- component axis has a high relationship with MPS, TCA level analysis is 1336 and for forest class-level analysis and MP and the second factor highly correlated with was about 34. The average edge density is relative to an PSSD and AWMSI. Likewise, the third factor has a high area, and the average edge density of the landscape- correlation with only MPFD and the fourth also factor level measures was about 198 meter/hectare and aver- highly correlated with MSI and Nump. Consequently, the age edge density was the forest class was about 22 highest component loading is considered to depict one meter/hectare and. Concerning, shape index the aver- representative metrics, and the analysis helps to select age landscape shape index (LSI) of the landscape level MPS, PSSD, MPFD and MSI representing thefirst, second, was about 25 and it became about 4 for forest class- third and the fourth component axis, respectively. Addis level evaluation. Interspersion Juxtaposition Index Ababa and the surrounding area forest landscape, thus, (IJI), on the other hand, measures how even the are highly characterized by shape and patch size and patch adjacencies is, thus, in the landscape level ana- density class-level metrics. Of these four chosen metrics lysis it became more even, which is about 76, where as two of them are from the shape (MPFD and MSI) and the in the forest class level analysis, it was at the middle other two are from patch size and density (MPS and level evenness and it is about 53. The total size of the PSSD) category of the landscape metrics. computed core patches evaluation showed different Furthermore, the factor analysis data reduction values for landscape and forest class-level analysis. technique helps to identify the underlined landscape 166 A. MOHAMED ET AL. Table 1. The Landscape metrics selected for detailed landscape- and class-level analysis. Metrics Acronyms Descriptions Area Metrics Total Landscape Area TLA Sum of areas of all patches in the landscape. Example: Landscape Area (TLA) = 46,872.719 + 359,047.844 + . . . + 62,423.574, TLA = 184.11 hectares Patch Density & Size Metrics Mean Patch Size MPS Example: Mean Patch Size of Conifer Patches (Class Level), MPS = (359,047.844 + 139,531.484 . . . + 65,819.984)/4, MPS = 17.42 hectares Example: Mean Patch Size of Patches (Landscape Level), MPS = (46,872.719 + 359,047.844 + . . . + 62,432.574)/14, MPS = 13.15 hectares Patch Size Coefficient of PSCoV Example: Coefficient of Variation of Conifer patches (Class Level) Variance PSCoV = PSSD/MPS = (11.05 hectares/17.42 hectares) *100 = 63 Example: Coefficient of Variation of all patches (Landscape Level) PSCoV = (9.51 hectares/13.15 hectares)*100 = 72 Patch Size Standard PSSD Example: Patch Size Standard Deviation of Conifer Patches (Class Level) Deviation PSSD = 11.05 hectares Example: Patch Size Standard Deviation of all patches (Landscape Level) PSSD = 9.51 hectares Edge Metrics Total Edge TE Example: Total Edge Conifer (Class Level) TE = Sum of perimeter of all conifer patches. TE = 10,858.88 meters Units are expressed in native maps units. Example: Total Edge all patches (Landscape Level) TE = Sum of perimeter of all patches TE = 28,607.27 meters Shape Metrics Mean Shape Index MSI MSI is equal to 1 when all patches are circular (for polygons) or square (for rasters (grids)) and it increases with increasing patch shape irregularity. MSI = sum of each patch’s perimeter divided by the square root of patch area (in hectares) for each class (when analyzing by class) or all patches (when analyzing by landscape), and adjusted for circular standard (for polygons), or square standard (for raster (grids)), divided by the number of patches. Area Weighted Mean AWMSI AWMSI is equal to 1 when all patches are circular (for polygons) or square (for rasters (grids)) and it increases with Shape Index increasing patch shape irregularity. AWMSI equals the sum of each patch’s perimeter, divided by the square root of patch Mean Patch Fractal MPFD Mean patch fractal dimension (MPFD) is another measure of shape complexity. Mean fractal dimension Dimension approaches one for shapes with simple perimeters and approaches two when shapes are more complex. Diversity & Interspersion Metrics Mean Proximity Index MPI Mean proximity index is a measure of the degree of isolation and fragmentation of a patch. MPI uses the nearest neighbor statistic. The distance threshold default is 1,000,000. If MPI is required at specific distances, select Set MPI Threshold from the main Patch pull-down menu and enter a threshold distance. Both MNN and MPI use the nearest neighbor statistic of similar polygons in their algorithm. Interspersion IJI Measure of evenness of patch adjacencies, equals 100 for even and approaches 0 for uneven adjacencies Juxtaposition Index Table 2. Accuracy assessment result by study period. the landscape structure and holds majority of other metrics information. LULC Type PA UA Water 100 100 Forest 95 96.6 Mixed Woodland 83.3 84.7 3.2. The pattern of the underlining landscape Cultivated Land 81.7 80.3 metrics Bareland 78.3 82.5 Builtup 91.7 91.7 The result of the factor analysis is helpful to show the Overall accuracy 89 Overall kappa statistics 87 spatial pattern of the representative underlining metrics that configure the comprehensive picture of the area landscape structure. The landscape-level spa- structure of the study area and depicted the surrogate tial structure reflects a mosaic of the overall LULC metrics to give a special emphasis on the landscape types. The most important dimension of the study monitoring and conservation planning. Therefore, the area landscape structure pattern explained by PSSD, landscape-level factor analysis identified PSSD, AWMSI, MSI and IJI. The patch size at the landscape- AWMSI, MPFD and IJI as they contributed the high- level metrics score highly variable at the center of the est loading on the four component axis, and the forest region where an area mainly covered by an urban built class-level computation depicted MPS, PSSD, MPFD up and comprise divers LULC type. PSSD, as and MSI contributing the highest loadings on the four a measure of patch size deviations, is getting higher components. As a result, these eight landscape metrics in this center and somehow in the northwestern part were representing best as the underlined dimension of of the study area. The two shape metrics do not have GEOLOGY, ECOLOGY, AND LANDSCAPES 167 Figure 2. The Land use-land use/land cover map of the study area. the same pattern, and a very high value of AWMSI central, northeast central and northern part of the found in the southwestern part and some dispersed region. spots of high score was seen in the southeast periph- eral pat. However, the simple MSI high value mainly 4. Discussion found in the northeast and northwest part of the study area were relatively less affected natural environment. Although a number of landscape metrics were identi- The low value of MSI and AWMSI, conversely, was fied it is very important to understand the theoretical concentrated at the center of the study region that is and the empirical essence of them. For instance, an area in which high urban built up is dominated. Cushman, McGarigal, and Neel (2008) indicate that A high IJI also observed at the southwest and south- most of these landscape metrics are theoretically central part of the region shows that seams all patch related and many others are empirically associated types are equally adjacent to all other patch types. due to the consistent integration of different aspects Most of the other part appears increasingly uneven of structure in real landscapes. In fact, few representa- (Figure 4). tive metrics could be adequate to capture the land- Figure 5 is the map showing the pattern of the scape pattern. Griffith et al. (2000), for instance, landscape metrics based on the factor loading scores analyze the Kansas area landscape structure by using of the selected forest class-level landscape metrics. The different data reduction and quantification techniques, first two are from patch size and density landscape Schindler et al. (2008), Herzog et al. (2001) and Lausch metrics (MPS and PSSD) that have an almost similar and Herzog (2002) also used factor analysis. The cor- pattern, a high value of the score was dominantly relation analysis in this study showed that the mere visible around the southwestern and north-central relationships of the metrics in describing the Addis part of the study area, and it extends from south- Ababa and the surrounding city region’s landscape at west to north-eastern side continuously at a different class level (forest) and the overall landscape level. The scale. The third and fourth represent the shape cate- correlation also helps to reduce highly associated gory of landscape metrics represented by MPFD and metrics and make the data suitable for factor analysis, MSI. Though these metrics high score area have the which enables to choose the underlining landscape same pattern and extends southwest to northeast metrics for monitoring landscape situation in terms direction, and a very high score value of these metrics of land-use pattern and structure. When compared are dispersed different part of the region with with studies conducted in other ecosystems a varying level of gray tone spots. However, this high (Schindler et al., 2008) it has a similar amount of the gray ton spots are mainly visible in the west central, overall variance explained by the first four factors of 168 A. MOHAMED ET AL. Table 3. Summary statistics of 39 computed metrics for hex- variance. Sufficient research works were not made in agonal regions at landscape and class levels. quantifying the landscape structure and patterns in Landscape Level Summary Forest Class Level summary Ethiopia and most of the African tropical landscape. statistics statistics Nonetheless, various assessment methods and results Metrics Mean SD CoV Mean SD CoV produced in other parts of the world can help signifi- MPI 520.4 235.4 45.2 1800.7 2395.9 133.1 cantly to refer particularly for a warm and cool tem- MNN 44.8 10.1 22.6 60.9 50.9 83.7 IJI 76 14.9 19.6 52.8 21.1 40 perate highland landscape of the tropics like the study TLA 2492.2 13.3 0.5 2498.3 6.2 0.2 area Addis Ababa and the surrounding. NumP 1336.2 358.5 26.8 34.1 44.7 131 MPS 2 0.8 37.6 29.4 41.6 141.4 The landscape-level landscape structure analysis of PSCoV 1600.8 575 35.9 325.2 193.4 59.5 the study area mainly characterized by patch density, PSSD 33.1 17.8 53.8 88.2 104.2 118.1 TE 493,516.2 123,886.7 25.1 55,856.4 45,066.5 80.7 shape and interspersion categories of landscape ED 198.1 50.2 25.3 22.4 18 80.6 metrics which are represented by PSSD, AWMSI, MSI 1.4 0.1 5.3 1.4 0.2 13.2 AWMSI 9.5 3.7 39.3 3.8 2 52.4 MSI, and IJI. The size and patch density metrics help MPFD 1.1 0 0.7 1.1 0 1.8 to evaluate the various attributes of patches within the AWMPFD 1.3 0 2.4 1.2 0.1 5.3 ecosystem. As Cook, Yao, Foster, Holt, and Patrick SDI 1.1 0.2 19.4 SEI 0.8 0.2 18.6 (2005) indicated that large patches have a large benefit TCA 1037.8 343.7 33.1 373.5 381.1 102 and small patches have small supplementary benefit. CAD 19.4 7.2 36.9 0.9 0.9 103.8 MCA 3.2 4.3 133.7 44.4 98.5 221.9 The study area landscape mainly characterized by CASD 39.2 37.5 95.7 81.6 109 133.5 shape indices of AWMSI and MSI. Patch size and CACoV 4000.3 1683.4 42.1 283.5 188.6 66.5 TCAI 41.6 13.7 32.9 63.9 25.6 40.2 shape influence a number of basic ecological processes CACV1 2267.6 747.8 33 353.4 210.7 59.6 in a landscape in a particular decrease with increasing CASD1 21.4 16.7 78.1 74.7 95.8 128.3 LPI 41.9 20.4 48.7 15.6 17.1 109.4 human land use (Kim & Stephan, 2005). The AWMSI LSI 24.7 6.2 25.2 3.8 2.1 55.2 calculates the complexity of the patches in the land- MCAI 1.9 0.6 31.5 12.5 11 87.9 scape according to their size and it is a good predictor MCA1 0.9 0.7 79.6 24.7 38.9 157.5 NCA 482.9 177.9 36.8 22 22.9 103.9 of birds and species richness pattern (Schindler, von MSIDI 1 0.3 26.7 Wehrden, Poirazidis, Wrbka, & Kati, 2013). IJI also PR 3.9 0.8 19.8 PRD 0.2 0 19.8 measures the distribution of adjacencies among differ- SHEI 0.8 0.2 18.6 ent patch types or configuration. Shape with the SIEI 0.8 0.2 18.5 MSIEI 0.7 0.2 27.1 arrangement of patches of different land-use and CA 480.5 436 90.8 cover types is used together with areal statistics-to %LAND 19.2 17.4 90.7 C%LAND 14.9 15.2 102 quantify the landscape structure and composition. DLFD 1.3 0.1 9.4 The forest class-level landscape structure of Addis Ababa and the surrounding city region is highly char- acterized by patch size and density and shape land- landscape-level analysis. However, other studies scape metrics. The analysis depicted MPS and PSSD as (Griffith et al., 2000) that contain greater than four- the main representative of patch size and density cate- factor component had explained the very little gory. Shi et al. (2008) describe the MPS as an indicator Figure 3. Correlation coefficient of all computed landscape metrics (a) landscape-level correlation matrix and (b) forest class-level correlation matrix. GEOLOGY, ECOLOGY, AND LANDSCAPES 169 Table 4. Factor varimax rotated matrix for the selected class- and landscape-level metrics. Landscape level metrics Forest class level metrics F1 F2 F3 F4 F1 F2 F3 F4 MPI 0.846 0.361 MPI 0.325 −0.425 0.475 MNN MNN 0.772 −0.537 IJI IJI 0.915 TLA TLA 0.331 0.542 NumP 0.31 0.919 NumP −0.855 −0.367 MPS 0.896 −0.342 MPS 0.929 PSCoV 0.731 0.341 PSCoV 0.36 −0.862 PSSD 0.997 PSSD 0.937 TE 0.769 0.533 TE −0.839 0.385 MSI 0.979 MSI 0.341 0.908 AWMSI 0.865 AWMSI −0.911 MPFD 0.881 MPFD −0.314 0.842 TCA 0.868 0.364 SDI −0.672 0.564 MCA 0.683 −0.425 SEI −0.715 0.337 0.462 TCAI 0.628 MCAI 0.799 MCAI −0.351 0.704 PR 0.31 −0.417 −0.757 DLFD 0.596 F = factor loadings. Factor loadings smaller than 0.3 have been removed. Figure 4. Landscape-level landscape metrics pattern. Figure 5. Forest class-level landscape metrics pattern. 170 A. MOHAMED ET AL. of the level of landscape fragmentation or grain, as it 5. Conclusions gets lower with increased disturbance and human Landsat and other similar remotely sensed satellite ima- intrusion. Lausch and Herzog (2002) also discussed geries offer valuable information helps to map the land- MPS as an important element of forest patches that scape configurations that raised the interests of determine habitat functioning, particularly if it is com- environmental planners and managers. Alongside, the bined with core area and neighborhood indices. digital mapping and statistical quantification packages Likewise, the shape metrics help to examine the most brought the LULC maps with a wide range of landscape important dimension of the spatial structure of the metrics that gives-rise to a new conception and techni- landscape in many previous studies (Griffith et al., ques of processing in the field of landscape ecology and 2000; Schindler et al., 2008). This study identified the remote sensing. The rapid development in these tech- shape metrics of MSI and MPFD that determined the nologies, thus, enable to measure the sophisticated landscape shape complexity and the patch fractal levels of landscape structure, ecological functioning dimension. Therefore, it is a vital dimension to predict and attempts were also made to monitor or manage birds and plant species richness pattern (Olivier et al., a very large, multi-jurisdictional landscapes more often 2017). The landscape pattern map of the area again now. The landscape pattern mapping and quantifica- helps to prioritize the specific zone or locality and tion, currently, appears one of the important urban monitor the functions for proper conservation plan- landscape planning and management instrument that ning and management that helps to maintain the helps to measure the extent of fragmentation, network- stated benefits of the landscape analysis. ing, species richness and other functions of the given Generally, landscape metrics are a useful tool for ecosystem or LULC types. The potential of landscape incorporating ecological knowledge in planning and metrics yields information on landscape pattern to management. One of the most important concepts is develop the environmental LULC indicators for differ- that landscape pattern strongly influences the ecological ent parts of the earth. process and characteristics. Moreover, it describes how Ethiopian highlands were one of the important the spatial structure influences the most fundamental examples to study Afro-mountain urban landscape ecological process, and how landscape planning and structures. This study thus has a significant role in the management intern influences the landscape structure. Addis Ababa and the surrounding area’slandscape As discussed above, four class levels and four land- ecology to provide comprehensive knowledge for the scape-level metrics were distinguished pertaining to city region decision-making organ. The analysis mainly the area landscape structure based on their factor used Arc GIS patch analyst, statistical tools to compute loading. The metrics were the best representative of 39 metrics for 178 hexagonal zones of landscape-level a large information generated by other metrics and and 66 hexagonal zones of forest class metrics. Out of form a core set of the structural characteristics of which the correlation matrix computes 35 landscape- landscape monitoring, and they are then helpful to level and 30 forest class-level metrics scoring 0.9, and address the main requirements to apply to the land- the above coefficient of variation was omitted to remove scape structure analysis. repetition. The multivariate factor analysis measured 17 The conservation planning, management schemes class- and 16 landscape-level metrics and the analysis, thus focusing on the forest LULC and the overall then identified 4 forest class- and 4 landscape-level landscape mosaic. As this city region hosts some of metric that has a higher component loading. the indispensable mountain forest and other green Accordingly, it depicted PSSD, AWMSI, MPFD and spaces that support the main landscape services and IJI for landscape-level, and MPS, PSSD, MPFD and goods needs, particularly through nutrient up taking MSI for forest class-level analysis that have to be given and biodiversity maintenance entails the special special emphasis for any landscape monitoring scheme. emphasis in this research. Therefore, any changes in Therefore, the landscape conservation planning and composition and configuration of the landscape monitoring endeavors should focus on the level of demand analysis of a different evaluation technique. patch fragmentation that determines the habitat func- At the same time, the knowledge generated in this tioning level on natural land covers like forest, the analysis also is useful to confront the challenge growth of dispersed built-up and sprawling in urban through different intervention technique. The moun- areas. In addition, it also concerned on maintaining the tain forest and other green spaces in the city region natural landscape complexity that determines the num- have to be maintained, and efforts of land-use man- ber of ecological processes and the extent of human agement should be supported by the landscape struc- intervention. The maps showing the pattern of the ture analysis, and this analysis also helps to measure underlining landscape metrics also portrayed the level the improvements due to the attempts of conservation of priority of space that has to be given for a set of not only in this study area but also the highland area of conservation mechanism and monitoring scheme. the tropical city region’s landscape. GEOLOGY, ECOLOGY, AND LANDSCAPES 171 On the whole, the Addis Ababa and the surrounding experiment and simulation in ecology. Ecological Modelling, 206(3–4), 347–359. city-region landscape structure analysis and mapping of Brown, D., Band, L. E., Green, K. O., Irwin, E. G., Jain, A., its spatial pattern are helpful to maintain the region’s Lambin, E. F., ... & Verburg, P. H. (2013). Advancing landscape ecology as they have a valuable role in biodi- Land Change Modeling: Opportunities and Research versity conservation and maintaining species richness, Requirements. Washington, D.C.: The National increasing nutrient up taking capacity and ecosystem Academies Press. Bruce, J. K., Neale, A. C., Nash, M. S., Van Remortel, R. D., connectivity, etc. In a nutshell, the spatial landscape Wickham, J. D., Riitters, K. H., & O’Neill, R. V. (2001). structure metrics have a vital role to analyze, monitor Predicting nutrient and sediment loadings to streams and providing a comprehensive understanding of from landscape metrics: A multiple watershed study a given landscape pattern. This study also explored the from the United States Mid-Atlantic region. Landscape underlined landscape structure metrics and their impli- Ecology, 16(4), 301–312. cation on the sustainable ecological conservation plan- Chmielewski, T. J., Kułak, A. A., & Malwina, M.-Ś.(2014). How to evaluate and forecast changes in landscape image: ning and management of the tropical highland urban The case of a Small River Valley in Poland. Landscape landscape in general and the study area in particular. Research, 40(4), 466–475. Some of the important concerns of future studies in Cook, E. A. (2002). Landscape structure indices for assessing the area could be stated as follows. Using a very-high urban ecological networks. Landscape and Urban spatial resolution satellite imageries less than the Planning, 58, 269–280. Cook, W. M., Yao, J., Foster, B. L., Holt, R. D., & 30 m helps to identify a comprehensive urban land Patrick, L. B. (2005). Secondary succession in an experi- uses. Urban areas typically exhibit spatially heteroga- mentally fragmented landscape : Community patterns mous LULC. This study considers the whole urban across space and time. Ecology, 86(5), 1267–1279. built-up as one class, however these very high- Retrieved from https://pdfs.semanticscholar.org/f241/ resolution images enable to group further with other 2b450567d7abdec72932ce5dec2b744a6800.pdf classes like residential, commercial and industrial. It is Cushman, S. A., McGarigal, K., & Neel, M. C. (2008). Parsimony in landscape metrics: Strength, universality, also possible to classify the vegetation cover in different and consistency. Ecological Indicators, 8(5), 691–703. types and this may again strengthen the interpretation Davies, S. A., Hernebring, C., Svensson, G., & Gustafsson, L. G. more exhaustively and to measure the class-level (2008). Theimpacts of climatechange and urbanisation on metrics of each LULC classes specifically. In addition, drainage in Helsingborg, Sweden: Combined sewer system. this study mainly analyzes the spatial configuration of Journal of Hydrology, 350, 100–113. Dobbs, C., Escobedo, F. J., & Zipperer, W. C. (2011). the landscape structure by taking a given time. While A framework for developing urban forest ecosystem ser- the temporal dynamics of this landscape is another vices and goods indicators. Landscape and Urban weak-spot of the study, essentially, urban landscape is Planning, 99, 196–206. rapidly altered through time, thus it is necessary to deal Farina, A. (2000). Landscape ecology in action (1st ed ed.). on this aspect. Venice, Italy: Kluwer Academic Publishers, Springer Science+Business Media Dordrecht. Fikirte, D., Kumelachew, Y., Mengistie, K., & Thomas, S. (2017). Land use/land cover changes and their causes in libokemkem Acknowledgments district of South Gonder, Ethiopia. Remote Sensing Applications: Society and Environment, 8(April), 224–230. We gratefully acknowledge the financial support from Addis Griffith, J. A., Martinko, E. A., & Price, K. P. (2000). Ababa University Office of the Director of Research under Landscape structure analysis of Kansas at three scales. the thematic research title “Resource efficiency, environ- Landscape and Urban Planning, 52,45–61. mental quality and sustainability of urban areas in Habtamu, T., Nyssen, J., Zenebe, A., Haregeweyn, N., Ethiopia”, Grant no. TR/11/2013. Kindu, M., Lemenih, M., & Haile, M. (2013). Ecological succession and land use changes in a lake retreat area (Main Ethiopian Rift Valley). Journal of Arid Disclosure statement Environments, 91,53–60. Hakan, A., & Nil, Ç. (2014). Mapping spatial variations of No potential conflict of interest was reported by the authors. land cover in a coastal landscape using pattern metrics. Procedia - Social and Behavioral Sciences, 120,23–30. Herold, M., Scepan, J., & Clarke, K. C. (2002). The use of ORCID remote sensing and landscape metrics to describe struc- tures and changes in urban land uses. Environment and Asfaw Mohamed http://orcid.org/0000-0003-3283-1776 Planning A, 34(8), 1443–1458. Herzog, F.,Lausch,A.,Müller,E.,Thulke,-H.-H., Steinhardt, U., & Lehmann, S. (2001). Landscape metrics References for assessment of landscape destruction and rehabilitation. Environmental Management, 27(1), 91–107. Adamczyk, J., & Tiede, D. (2017). ZonalMetrics – A python Jones, J. A., Swanson, F. J., Wemple, B. C., & toolbox for zonal landscape structure analysis. Computers Snyder, K. A. I. U. (2000). Effects of roads on hydrology, and Geosciences, 99,91–99. geomorphology, and disturbance patches in stream Birch, C. P. D., Oom, S. P., & Beecham, J. A. (2007). networks. Conservation Biology, 14(1), 76–85. Rectangular and hexagonal grids used for observation, 172 A. MOHAMED ET AL. Kevin, F. J., Robert, C. M., & Tait, M. (1986). The application of services. Arboriculture & Urban Forestry, 34(6), exploratory factor analysis in applied psychology: A critical 347–358. Retrieved from https://www.itreetools.org/eco/ review and analysis. Personnel Psychology, 39(2), 291–314. resources/08UFORE.pdf Kim, K. H., & Stephan, P. (2005). Landscape metrics to Olivier, P. I., Rolo, V., & van Aarde, R. J. (2017). Pattern or assess the ecological conditions of city regions: process? Evaluating the peninsula effect as a determinant Application to Kwangju City, South Korea. of species richness in coastal dune forests. PloS One, 12 International Journal of Sustainable Development and (4), e0173694. Retrieved from http://journals.plos.org/plo World Ecology, 12(3), 227–244. sone/article?id=10.1371/journal.pone.0173694 Kirstein, W., & Netzband, M. (2001). Landscape Metrics as a Rempel, R. S., Kaukinen, D., & Carr, A. P. (2012). Patch Tool for the Comparison of Different Urban analyst and patch grid. Ontario Ministry of Natural Regions. Remote Sensing of Urban Areas/ Resources. Centre for Northern Forest Ecosystem Fernerkundung in urbanen Räumen. Regensburger Research, Thunder Bay, Ontario. Geographische Schriften, (35). Schindler, S., Poirazidis, K., & Wrbka, T. (2008). Towards Lausch, A., & Herzog, F. (2002). Applicability of landscape a core set of landscape metrics for biodiversity assess- metrics for the monitoring of landscape change: Issues of ments: A case study from Dadia National Park, Greece. scale, resolution and interpretability. Ecological Ecological Indicators, 8(5), 502–514. Indicators, 2(1–2), 3–15. Schindler, S., von Wehrden, H., Poirazidis, K., Wrbka, T., & Li,X.,Ling,L.,Cheng,G.,&Xiao,H.(2001). Quantifying landscape Kati, V. (2013). Multiscale performance of landscape structure of the heihe River Basin, North-West China using metrics as indicators of species richness of plants, insects FRAGSTATS. Journal of Arid Environments, 48(4), 521–535. and vertebrates. Ecological Indicators, 31,41–48. Livesley, S. J., McPherson, G. M., & Calfapietra, C. (2016). Seto,K.C.,&Fragkias,M.(2005). Quantifying spatiotemporal The urban forest and ecosystem services: Impacts on patterns of urban land-use change in four cities of China with urban water, heat, and pollution cycles at the tree, street, time series landscape metrics. Landscape ecology, 20(7), 871-888. and city scale. Journal of Environment Quality, 45(1), 119. Shi, Y., Xiao, J., & Shen, Y. (2008). Landscape pattern Retrieved from https://dl.sciencesocieties.org/publica change and associated environmental implications in tions/jeq/abstracts/45/1/119 the Haihe River Basin, China. ISPRS J. Photogramm. Lu, D., & Qihao, W. (2004). Spectral mixture analysis of the Remote Sens, 37, 569-574. urban landscape in indianapolis with Landsat ETM+ Simson, A. J. (2017). Landscape structure planning and the imagery. Photogrammetric Engineering & Remote urban forest in polycentric city regions. IOP Conference Sensing, 70(9), 1053–1062. Retrieved from http://open Series: Earth and Environmental Science, 61(1), 012025. url.ingenta.com/content/xref?genre=article&issn=0099- Simonian, L., Hernández Munguía, J. G., Arana Noguera, G. 1112&volume=70&issue=9&spage=1053 A., Thomas, L., Middleton, J., & Phillips, A. (2010). Malinowska, E., & Szumacher, I. (2013). Application of Global forest resources assessment 2010: main report landscape metrics in the evaluation of geodiversity. (No. FAO FP-163). FAO, Rome, Italy, 340. Miscellanea Geographica – Regional Studies On Solomon Mulugeta. (1985). Meeting the Housing Shortage Development, 17(4), 28–33. in Addis Ababa: The Case of Housing McGarigal, K., Cushman, S. A., Neel, M. C., & Ene, E. Cooperatives (Doctoral dissertation, MA Thesis. Addis (2002). FRAGSTATS: spatial pattern analysis program Ababa, Addis Ababa University). for categorical maps. Computer software program pro- Tadesse, A. (2010). Review of the urban environment in Ethiopia in duced by the authors at the University of Massachusetts, 2008. ed. Edwards. Addis Ababa: Forum for Environment. Amherst. Available at the following web site: www. Temesgen,D.,Cruz,F.,Kindu,M.,Turrión,M.B.,&Gonzalo,J. umass. edu/landeco/research/fragstats/fragstats. html, 6. (2014). Land-Use/Land-Cover (LULC) change and socioeco- Mengistie, K., Schneider, T., Döllerer, M., Teketay, D., & nomic conditions of local community in the central highlands Knoke, T. (2018). Scenario modelling of land use/land of Ethiopia. International Journal of Sustainable Development & cover changes in Munessa-Shashemene landscape of the World Ecology, 21(5), 406–413. Ethiopian Highlands. Science of the Total Environment, Temesgen, G., Tulu, T., Argaw, M., Worqlul, A. W., Tolessa, T., & 622–623, 534–546. Kindu, M. (2018). Estimating the impacts of land use/land Mengistie, K., Schneider, T., Teketay, D., & Knoke, T. (2013). Land cover changes on ecosystem service values : The case of the use/land cover change analysis using object-based classification Andassa watershed in the Upper Blue Nile Basin of Ethiopia”. approach in Munessa-Shashemene landscape of. Remote Ecosystem Services, 31, 219–228. Sensing, 5, 2411–2435. Terefe, T., Feyera, S., & Moges, K. (2016). Landscape com- Mesfin, T. B. (2009). Spatial metrics and Landsat data for urban position and configuration in the central highlands of landuse change detection in Addis Ababa (Ethiopia Spatial Ethiopia. Ecology and Evolution, 6(20), 7409–7421. Metrics And Landsat Data For Urban Landuse Change Walz,U.(2011). Landscape structure, landscape metrics and bio- Detection In Addis Ababa). Ethiopia: Universitat Jaume I. diversity. Living reviews in landscape research. 5 (3), 1–35. Mpofu, T. P. Z. (2013). Environmental challenges of urba- Wolde, M. (2013). Changes in regulating ecosystem services fol- nization: A case study for open green space management. lowing establishing exclosures on communal grazing lands in Research Journal of Agricultural and Environmental Ethiopia: A synthesis. Journal of Ecosystems, 2013,1–12. Management, 2(4), 105–110. Retreieved from http:// Wu, J., Jelinski, D. E., Luck, M., & Tueller, P. T. (2000). www.apexjournal.org Multiscale analysis of landscape heterogeneity: scale var- MUDHo. 2016.Riversand RiverBuffer Green Infrastructure iance and pattern metrics. Geographic information Design Standard Implementation, Manual. Addis Ababa, sciences, 6(1), 6-19. Ministry of Urban Development and Housing. Yared, G., Heyaw, T., Pauleit, S., & Mengistie, K. (2019). Nowak, D. J., Crane, D. E., Stevens, J. C., Hoehn, R. E., Urban green spaces supply in rapidly urbanizing Walton, J. T., & Bond, J. (2008). A ground-based method countries:Thecaseof. Remote Sensing Applications: of assessing urban forest structure and ecosystem Society and Environment, 13, 138–149. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Geology Ecology and Landscapes Taylor & Francis

Quantification and mapping of the spatial landscape pattern and its planning and management implications a case study in Addis Ababa and the surrounding area, Ethiopia

Download PDF
12 pages

Loading next page...
 
/lp/taylor-francis/quantification-and-mapping-of-the-spatial-landscape-pattern-and-its-h107hdjGpc

References (52)

Publisher
Taylor & Francis
Copyright
© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON).
ISSN
2474-9508
DOI
10.1080/24749508.2019.1701309
Publisher site
See Article on Publisher Site

Abstract

GEOLOGY, ECOLOGY, AND LANDSCAPES 2021, VOL. 5, NO. 3, 161–172 INWASCON https://doi.org/10.1080/24749508.2019.1701309 RESEARCH ARTICLE Quantification and mapping of the spatial landscape pattern and its planning and management implications a case study in Addis Ababa and the surrounding area, Ethiopia a,b a c Asfaw Mohamed , Hailu Worku and Mengistie Kindu Ethiopian Institute of Architecture, Building Construction and City Development, Addis Ababa University, Addis Ababa, Ethiopia; b c Department of Geography and Environmental studies, Hawassa University, Hawassa, Ethiopia; Institute of Forest Management, Department of Ecology and Ecosystem Management, TUM School of Life Sciences Weihenstephan, Technische Universität München, Freising, Germany ABSTRACT ARTICLE HISTORY Received 22 May 2019 Rapid urbanization in Addis Ababa and the surrounding area resulted in the rapid land use/ Accepted 27 November 2019 land cover (LULC) change that affected landscape structures and ecological functions. This study aimed at quantifying and mapping the spatial patterns of landscape structure for KEYWORDS a sustainable city region landscape conservation planning and management. GIS and statistical LULC; Landsat; remote tools were used to compute important landscape metrics. Pearson Correlation and factor sensing; city region; green analysis were also applied to reduce redundant indices and identify underlining factor of the space landscape structure by network of hexagonal area. The analysis depicted four landscape and four class-level underlined metrics. Accordingly, as the region overall landscape was character- ized by patch size and density, shape and texture (interspersion) index, the forest class also attributed by patch size and density, and shape metrics. The result shows that the region landscape planning and management schemes must emphasize on the level of patch frag- mentation and landscape complexity to maintain the natural land cover habitat functioning, the amount of ecological process and extent of human intervention. This research will help scientific base decision-making in conservation planning and management of the tropical highland urban landscape in general, and the study area in particular. 1. Introduction acceptable quality of life for their residents (Simson, 2017). Concurrently, studies were dealing more on the Urban areas, nowadays, are growing at an alarming rate identification and quantification of the ecosystem ser- and becoming more complex. Urban built-up, agricul- vices and goods generated from the city region’svegeta- ture and other anthropogenic uses take over areas tion. For instance, Livesley, McPherson, and Calfapietra which are previously covered by natural vegetation, (2016)review different researches about the urban for- particularly forest and other green spaces (Yared, est impact on the urban heat island effect, human ther- Heyaw, Pauleit, & Mengistie, 2019). Such mal comfort, soil and water pollution, urban catchment a degradation of the city regions’ vegetation and the hydrology and air pollution. Likewise, Nowak et al. rate of their ecosystem functions caused by uncon- (2008) measure urban forest contribution to pollution trolled urban growth and other anthropogenic activities removal and carbon storage and sequestration. Dobbs, have not yet given much emphasis. Researches also have Escobedo, and Zipperer (2011) also indicated soil PH shown the danger of ignoring the city region (metro- and organic matter as the most influential ecosystem politan area) vegetation. For example, Davies, services and goods generated from urban trees. Hernebring, Svensson, and Gustafsson (2008)pre- A mix of remotely sensed data, GIS and the con- sented urbanization and the changing climate as the ceptual basis of landscape ecology is helpful to assess main factor for more frequent and severe flooding, and monitor the large-scale ecology (Herold, Scepan, heat and air pollution. In fact, the notion of the sustain- & Clarke, 2002; Lausch & Herzog, 2002; Mengistie, able city brought the relationship of the social and Schneider, Döllerer, Teketay, & Knoke, 2018). The biophysical environment by highlighting the enormous concept of Landscape ecology is an interdisciplinary benefits of the urban vegetation and the challenge of its subject concerned with the interaction between the governance and management (Brown et al., 2013). The spatial pattern and ecological process (Li, Ling, idea of urban green infrastructure thus appears in rela- Cheng, & Xiao, 2001), and it is also an important tion to urban forestry considering the landscape struc- method to measure the status of the city regions’ ture planning of cities and city regions to provide an LULC configuration and related ecosystem functions. CONTACT Asfaw Mohamed asfethiop@yahoo.com Ethiopian Institute of Architecture, Building Construction and City Development, Addis Ababa University, Po. Box. 518, Addis Ababa, Ethiopia © 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group on behalf of the International Water, Air & Soil Conservation Society(INWASCON). This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 162 A. MOHAMED ET AL. At the same time, the advancement of remotely sensed ecological footprint which could be even more than data and GIS technology gives rise to the concerned their immediate hinterlands. In order to cater to the experts’ ability to measure the LULC pattern and to urban immediate economic need, a huge amount of have better knowledge about its spatial configuration energy and construction material resources was and landscape structure (Mengistie, Schneider, extracted and utilized. This is also responsible for the Teketay, & Knoke, 2013). Currently, various landscape degradation of the natural support system and irrever- metrics that measure both the structural and func- sible damage and loss of critical ecosystem functions tional characteristics of the landscape were recognized (Tadesse, 2010). On the other hand, Mpofu (2013) (Malinowska & Szumacher, 2013). Different showed the sub-Saharan African context, uncontrolled researches depicted a wide range of its application. expansion of cities is followed by the degradation of Walz (2011) utilized the metrics to investigate, evalu- the ecosystem services like sinking or recycling of ate and measure the relationship between landscape polluted substances. This deterioration is mainly due structure and biodiversity. Also, Kim and Stephan to the complexity of managing and regulating the (2005) used it to show the level of species richness, system as it tends to lay across administrative bound- patterns of species dispersal and the extent of ecologi- aries. Similarly, the study area, Addis Ababa and the cal stability of the different landscapes and Jones, surrounding Oromiya Special Zone peri-urban vege- Swanson, Wemple, and Snyder (2000) analyze the tation covers and its landscape functions are suffering ability of landscape metrics to predict nutrient and with such critical problem. sediment yield to streams. Therefore, having a clear Studies were conducted on the landscape patterns knowledge and understanding of the structure of the quantification and mapping, but the urban landscapes city regions landscape pattern is an important require- of the Afro-mountain region are not well discussed. ment for management decisions on its conservation Some planning document in Ethiopia has lately and development. brought up the notion of landscape in relation to the The Ethiopian vegetation covers suffered from green infrastructure development (Ministry of Urban a wide range of deforestation and some small forest Development and Housing) (MUDHo, 2016). Mesfin patches remained in inaccessible and marginal areas (2009) also made other attempts in measuring the (Simonian et al., 2010; Fikirte, Kumelachew, landscape metrics for built-up class of Addis Ababa Mengistie, & Thomas, 2017; Habtamu et al., 2013, city only. This research, however, takes into account Temesgen, Cruz, Kindu, Turrión, & Gonzalo, 2014). the comprehensive city region (metropolitan area), At the same time, the deforestation and degradation and involves the landscape-level and class-level land- contribute high for the depletion of the area vegetal scape metrics of ecologically valuable land covers such cover ecosystem regulating service (Temesgen et al., as forest. Such quantification and mapping of the 2018). Terefe, Feyera, and Moges (2016) have shown urban landscape pattern therefore have enormous that the decline of overall ecosystem services by 68%, importance in the ecosystem conservation process of mainly because of deforestation in the study con- the city region. As the case study of Afro-mountain ducted in the central highlands of Ethiopia. region landscape, Addis Ababa and the surrounding Therefore, the protection of these valuable vegetation area characterized by rapid built-up growth and dete- covers has a significant impact on the extent of eco- riorating ecosystem conditions, thus the landscape system functions provided by the system. Wolde analysis enables to design the framework for sustain- (2013) for example, has shown the wide difference able environmental planning and management. The between a closure and communal grazing land ecosys- main objectives of this study are to (1) analyze the tem functioning in terms of carbon, total soil nitrogen study area landscape and class-level metrics and map stock and the availability of phosphorus stock particu- zonal characteristics of the landscape structure, (2) larly in northern highland areas of Ethiopia. describe the underlined landscape structure showing Specifically, in the study area, Addis Ababa and the Addis Ababa and the surrounding area overall land- surrounding Oromiya region, the problem is more scape and forest class-level structure, and (3) identify conspicuous. The city and the overall city region vege- the landscape planning and management schemes for tation and its ecosystem service’s degradation are sustainable landscape functioning. mainly caused by the huge dependence of urban and rural areas demand of their product particularly for their fuel wood consumption. For example, Solomon 2. Materials and methods (1985) relates the continuation of Addis Ababa as the 2.1. Study area capital of the country with the introduction of Eucalyptus tree after long years of wandering on the The selected study area encompasses the biophysical surrounding hilltops for their fuel wood need. The environment found within the administrative bound- implication of the cities’ physical expansion also aries of Addis Ababa and the surrounding districts of beyond its population size as they have a large Oromiya that covers about 4400 sq.km area. In fact, GEOLOGY, ECOLOGY, AND LANDSCAPES 163 the total area of the regions is 5400 sq.km while the Forest LULC type is essentially forested patches and study was conducted on an area of 4400 sq.km, where mixed woodland comprises shrubs and woody grass- the proposed hexagonal zones fully overlay on. This land. Cultivated land also is a land use with crops and area encompasses the major forest patches and other harvested land. Besides, bare land is an area that ecologically valuable LULC at the center of the coun- includes soils, quarries, and exposed rocks, and built- 0 0 try. It is geographically located between 8 38 0N to up cover mainly represents roads, parking lots and 0 0 0 0 0 0 00 buildings. Accordingly, this research applied 9 30 20N latitude and 38 28 4E to 39 6 0 E longitude a confusion matrix technique to assess the accuracy (Figure 1). The major source and potentials of Addis of the classification. The producer, user, overall accu- Ababa green spaces and other ecological functions are racy, and the Kappa coefficients were calculated to from the forest patches found in the peri-urban part of check the image classification and signature selection the city and the surrounding Oromiya districts. dependability. 2.2. Data used and LULC classification 2.3. Hexagonal area delineation and zonal During the last decade, the study of landscape metrics landscape metrics statistical computation has been used different types of remotely sensed data (Kirstein & Netzband, 2001). A number of landscape The landscape metrics evaluation was undertaken based structure studies were conducted using Landsat on the prepared LULC map. To measure the zonal images (Bruce et al., 2001; Lu & Qihao, 2004; Seto & landscape metrics, the overplaying hexagonal area Michail, 2005). We generated the LULC map of the map was prepared as it was more suitable for analyzing study area through GIS and remote sensing techni- the ecological modeling chosen by various studies ques. The main data source to classify the LULC map (Adamczyk & Tiede, 2017; Birch, Oom, & Beecham, of this study was Landsat-8 OLI, which scanned on 2007; Schindler, Poirazidis, & Wrbka, 2008). This 23rd of December 2015 with a row 168 and path 54. research, therefore, used Arc GIS patch analyst spatial The LULC classification was made by a supervised statistics by region tool that enables users to main class classification technique through a maximum likeli- and landscape-level analysis on a raster layer for indi- hood algorithm. For training samples about 60 sample vidual polygon regions like the overlying hexagonal sites of LULC type were chosen. To sharpen the sig- zones. This function also permits the analysis of the nature selection further, normalized difference vegeta- relationship between the structural characteristics of tion index (NDVI) based unsupervised classification the landscape and some identified functions or values was also used. Then, six end members (LULC types) (Rempel, Kaukinen, & Carr, 2012). Of the total area of were selected; these are water bodies, forest, mixed 54,000 sq.km, 178 full hexagonal zones were taken to woodland land, cultivated land, bare land, and built- support the equal area representation of the selected ups. Water bodies mainly include reservoirs and dams. zones statistical analysis. The network of the hexagon Figure 1. Study area map of Addis Ababa and the surrounding Oromiya. 164 A. MOHAMED ET AL. has 25 sq.km and this size designated as an optimal permit near-natural disturbance regimes (Kim & representation of the necessary details and the viability Stephan, 2005). The variation of the patch size and of the data. Thus, the entire 178 hexagonal zones were density as represented in this study by TLA, MPS, applied for landscape-level analysis and 66 of them used PSSD and PSCov, measuring the level of fragmenta- for class-level forest landscape metrics. In fact, as tion and fractal of a certain LULC type in each zone of Adamczyk and Tiede (2017) have shown, the entire the categorical map (Shi et al., 2008; Wu et al., 2000). landscape metrics may not calculate well in such zonal For instance, mean patch size (MPS) is a function of level evaluation since the specific conditions for calcu- the total area of the landscape and the patch number lating landscape metrics within statistical zones that and it is an indicator of grain. Smaller values the index considered as a small subset of the landscape. showed a higher fragmentation of the landscape Therefore, out of the 42 landscape metrics, 31 of them (Kirstein & Netzband, 2001). applied for forest class and 35 of them applied for The grid shape of the LULC arrangement reflects an landscape-level measurement. anthropogenic intervention. However, the shape land- After the patch analyst landscape metrics computa- scape metrics mainly measure how irregular a certain tion, the output further evaluated statistically. Some of LULC is, and as the habitats shape becomes more a highly correlated metrics having the coefficient of irregular it can also comprise more plant species. greater than or equal to 90%, thus excluded from Walz (2011) thus indicated how the shape affected further analysis to remove redundancy (Griffith, the number of species. Therefore, the shape complex- Martinko, & Price, 2000). As a result, 17 class-level ity used to asses LULC data as an index for species forest landscape metrics and 16 landscape-level richness. MPFD, MSI, and AWMSI as a measure of metrics were depicted for further analysis. The statis- landscape shape showed the level of LULC type irre- tical treatment of the data hase performed the normal- gularity and complexity in this research. For example, ity tested to get information about relationship MSI calculates to mean patch shape complexity and behavior between landscape metrics factor analysis when all patches are square and increase without limit using varimax normalized rotation through Shapiro- as patch shape becomes more irregular. In fact, it is the Wilk normality test. To identify the number of com- simplest and most straightforward measure of overall ponents first the principal component analysis (PCA) shape (McGarigal et al., 2002). However, it appears was evaluated. Though different criteria were sug- more meaningful when it is related to patch density gested, in this case, the number of components deter- and size measures like MPS. Since the patch area and mined by taking the engine-value greater than one as perimeter length were used to produce the shape criteria used by Kevin, Robert, and Tait (1986). Based indices. Concurrently, AWMSI calculates the com- on the calculated number of components factor ana- plexity of the patches in the landscape according to lysis was executed. Factor analysis as a data reduction their size. Circular patches have an AWMSI value technique applied for a long list of landscape metrics close to one and it increases with patch shape irregu- evaluation (Herzog et al., 2001; Lausch & Herzog, larity (Olivier, Rolo, & van Aarde, 2017). Larger 2002; Schindler et al., 2008). Thus, the highest loading patches enter more strongly in the calculation than of each factor was taken as a representative metric smaller ones do, thus greater values index is the indi- (Herzog et al., 2001; Schindler et al., 2008). In general, cator for a higher form variety of the surfaces (Kirstein for the LULC and landscape metrics computation, Arc & Netzband, 2001). GIS tools were applied, and for statistical measure- Diversity interspersion or configuration metrics are ment, R studio programing language was used. another essential measures. It is basically a spatial characteristic that reflects the spatial distribution of a certain LULC type (Farina, 2000). The analysis 2.4. Major landscape metrics categories highly dependent on the resolution of the data and the size of the study area. IJI is an example of Although many landscape metrics have been calcu- a configuration or interspersion measure, calculated lated, some of the main metrics computed for this from the relationship between the length of edge type study area have been highlighted. Accordingly, the and a total edge of the landscape, divided by the patch density and size are the important factors that number of land-use type. As a result, when the value influence a number of basic ecological processes in close to 0, the distribution of adjacencies among patch a landscape. Generally, large patches provide large types becomes increasingly uneven, and when the benefits and small patches have small supplemental value equals to 100 means all patch types are equally benefits (Cook, 2002). In addition, large patches of adjacent to all other patch types (Herzog et al., 2001). natural vegetation are an important measure of the In addition, the edge metrics reflect the level of landscape’s ecological integrity as they protect aquifers human interference or it eventually measures how and interconnected stream networks. They also enable the natural state of the environment is (Chmielewski, to bear viable populations of particular species in Kułak, & Malwina, 2014). Edge metrics again interior habitats, which provide core habitats and GEOLOGY, ECOLOGY, AND LANDSCAPES 165 considered as a representation of landscape configura- The average total core area of landscape-level mea- tion. TE is an absolute measure of the total edge length sures was about 1038 hectare and it appears about 373 of a particular patch type at the class level or of all hectare for forest class analysis. patch types at the landscape level (Hakan & Nil, 2014). Some of these landscape metrics depicted the Since this study computes the metrics based on the redundancy of a certain dimensions, thus the correla- hexagonal region, it introduced an artificial edge of tion matrix helps to remove such repetitive measure- each hexagonal area and the analysis was executed by ment. The analysis correlates the metrics and the cutting the layer with the statistical unit. The main landscape level using 35 metrics with 595 correlation landscape metrics considered in this study are shown pairs and forest class-level measures using 30 metrics in Table 1. with 435 correlation pairs and. The correlation coeffi- cient matrix of the forest class- and landscape-level metrics portrayed by Figure 3 that showed the rela- 3. Results tionship between the computed landscape metrics. To make the data suitable for factor analysis and useable 3.1. The LULC classification and statistical for further interpretation it removes some of highly analysis of landscape metrics correlated metrics that have the correlation coefficient A total of six LULC types were extracted with an greater than or equal to 0. 9. overall accuracy of 89% with a kappa coefficient of In addition, the factor analysis result depicted that 0.87 (Table 2). From the extracted 4400 sq.km area, the level of the relationship of the component factors Figure 2 depicts that 0.35% occupied by water, 7% of with each landscape metrics. The first four compo- the area is forest, 38% of the area is mixed woodland, nents of the overall landscape and forest class metrics 26% of the area is cultivated land, 20% of the area is were identified. The factor loading of each metric bare land, and the remaining 8% of the area covered by corresponded with the components, and the ortho- built-up. gonally rotated component matrix has shown in The 39 landscape metrics summary statistics are Table 4. Regarding the landscape-level analysis, while shown in Table 3, and it shows the mean, standard the first-factor component was highly correlated with division and coefficient of variation for the hexagonal PSSD, MPS, Nump and TE, the second component area. The summary gives the general pictures of the factor has a high correlation with AWMSI and PSCoV. quantitative result of the landscape metrics at forest The third component factor also was correlated with class level and, the overall landscape-level analysis. In MSI and MPFD and the fourth factor was correlated this section, the study tries to see at least some of the highly with IJI. The first four highest factor loading of representative measures of the broader landscape orthogonally rotated landscape-level metrics again metrics categories (Area, Patch Size and Density, depicted in the component matrix and the first axis Edge, Shape, Diversity and Interspersion, and Core depicted PSSD, the second axis identified AWMSI, the Area), although the measures are not applied at the third one picked MPFD and the fourth axis took IJI. landscape level, an average total landscape area of Of these, two of them were from shape landscape forest- and landscape-level metrics is equal to 2500 metrics category (AWMSI and MPFD), one from hectare, as it is considered the equal-sized hexagonal patch size and density (PSSD) and the other one is area. Also, the percentage of landscape area of forest from diversity interspersion or texture category (IJI). was about 19% with average coefficient of variation Accordingly in the forest class-level metrics, the first (90%). The average number of patches of landscape- component axis has a high relationship with MPS, TCA level analysis is 1336 and for forest class-level analysis and MP and the second factor highly correlated with was about 34. The average edge density is relative to an PSSD and AWMSI. Likewise, the third factor has a high area, and the average edge density of the landscape- correlation with only MPFD and the fourth also factor level measures was about 198 meter/hectare and aver- highly correlated with MSI and Nump. Consequently, the age edge density was the forest class was about 22 highest component loading is considered to depict one meter/hectare and. Concerning, shape index the aver- representative metrics, and the analysis helps to select age landscape shape index (LSI) of the landscape level MPS, PSSD, MPFD and MSI representing thefirst, second, was about 25 and it became about 4 for forest class- third and the fourth component axis, respectively. Addis level evaluation. Interspersion Juxtaposition Index Ababa and the surrounding area forest landscape, thus, (IJI), on the other hand, measures how even the are highly characterized by shape and patch size and patch adjacencies is, thus, in the landscape level ana- density class-level metrics. Of these four chosen metrics lysis it became more even, which is about 76, where as two of them are from the shape (MPFD and MSI) and the in the forest class level analysis, it was at the middle other two are from patch size and density (MPS and level evenness and it is about 53. The total size of the PSSD) category of the landscape metrics. computed core patches evaluation showed different Furthermore, the factor analysis data reduction values for landscape and forest class-level analysis. technique helps to identify the underlined landscape 166 A. MOHAMED ET AL. Table 1. The Landscape metrics selected for detailed landscape- and class-level analysis. Metrics Acronyms Descriptions Area Metrics Total Landscape Area TLA Sum of areas of all patches in the landscape. Example: Landscape Area (TLA) = 46,872.719 + 359,047.844 + . . . + 62,423.574, TLA = 184.11 hectares Patch Density & Size Metrics Mean Patch Size MPS Example: Mean Patch Size of Conifer Patches (Class Level), MPS = (359,047.844 + 139,531.484 . . . + 65,819.984)/4, MPS = 17.42 hectares Example: Mean Patch Size of Patches (Landscape Level), MPS = (46,872.719 + 359,047.844 + . . . + 62,432.574)/14, MPS = 13.15 hectares Patch Size Coefficient of PSCoV Example: Coefficient of Variation of Conifer patches (Class Level) Variance PSCoV = PSSD/MPS = (11.05 hectares/17.42 hectares) *100 = 63 Example: Coefficient of Variation of all patches (Landscape Level) PSCoV = (9.51 hectares/13.15 hectares)*100 = 72 Patch Size Standard PSSD Example: Patch Size Standard Deviation of Conifer Patches (Class Level) Deviation PSSD = 11.05 hectares Example: Patch Size Standard Deviation of all patches (Landscape Level) PSSD = 9.51 hectares Edge Metrics Total Edge TE Example: Total Edge Conifer (Class Level) TE = Sum of perimeter of all conifer patches. TE = 10,858.88 meters Units are expressed in native maps units. Example: Total Edge all patches (Landscape Level) TE = Sum of perimeter of all patches TE = 28,607.27 meters Shape Metrics Mean Shape Index MSI MSI is equal to 1 when all patches are circular (for polygons) or square (for rasters (grids)) and it increases with increasing patch shape irregularity. MSI = sum of each patch’s perimeter divided by the square root of patch area (in hectares) for each class (when analyzing by class) or all patches (when analyzing by landscape), and adjusted for circular standard (for polygons), or square standard (for raster (grids)), divided by the number of patches. Area Weighted Mean AWMSI AWMSI is equal to 1 when all patches are circular (for polygons) or square (for rasters (grids)) and it increases with Shape Index increasing patch shape irregularity. AWMSI equals the sum of each patch’s perimeter, divided by the square root of patch Mean Patch Fractal MPFD Mean patch fractal dimension (MPFD) is another measure of shape complexity. Mean fractal dimension Dimension approaches one for shapes with simple perimeters and approaches two when shapes are more complex. Diversity & Interspersion Metrics Mean Proximity Index MPI Mean proximity index is a measure of the degree of isolation and fragmentation of a patch. MPI uses the nearest neighbor statistic. The distance threshold default is 1,000,000. If MPI is required at specific distances, select Set MPI Threshold from the main Patch pull-down menu and enter a threshold distance. Both MNN and MPI use the nearest neighbor statistic of similar polygons in their algorithm. Interspersion IJI Measure of evenness of patch adjacencies, equals 100 for even and approaches 0 for uneven adjacencies Juxtaposition Index Table 2. Accuracy assessment result by study period. the landscape structure and holds majority of other metrics information. LULC Type PA UA Water 100 100 Forest 95 96.6 Mixed Woodland 83.3 84.7 3.2. The pattern of the underlining landscape Cultivated Land 81.7 80.3 metrics Bareland 78.3 82.5 Builtup 91.7 91.7 The result of the factor analysis is helpful to show the Overall accuracy 89 Overall kappa statistics 87 spatial pattern of the representative underlining metrics that configure the comprehensive picture of the area landscape structure. The landscape-level spa- structure of the study area and depicted the surrogate tial structure reflects a mosaic of the overall LULC metrics to give a special emphasis on the landscape types. The most important dimension of the study monitoring and conservation planning. Therefore, the area landscape structure pattern explained by PSSD, landscape-level factor analysis identified PSSD, AWMSI, MSI and IJI. The patch size at the landscape- AWMSI, MPFD and IJI as they contributed the high- level metrics score highly variable at the center of the est loading on the four component axis, and the forest region where an area mainly covered by an urban built class-level computation depicted MPS, PSSD, MPFD up and comprise divers LULC type. PSSD, as and MSI contributing the highest loadings on the four a measure of patch size deviations, is getting higher components. As a result, these eight landscape metrics in this center and somehow in the northwestern part were representing best as the underlined dimension of of the study area. The two shape metrics do not have GEOLOGY, ECOLOGY, AND LANDSCAPES 167 Figure 2. The Land use-land use/land cover map of the study area. the same pattern, and a very high value of AWMSI central, northeast central and northern part of the found in the southwestern part and some dispersed region. spots of high score was seen in the southeast periph- eral pat. However, the simple MSI high value mainly 4. Discussion found in the northeast and northwest part of the study area were relatively less affected natural environment. Although a number of landscape metrics were identi- The low value of MSI and AWMSI, conversely, was fied it is very important to understand the theoretical concentrated at the center of the study region that is and the empirical essence of them. For instance, an area in which high urban built up is dominated. Cushman, McGarigal, and Neel (2008) indicate that A high IJI also observed at the southwest and south- most of these landscape metrics are theoretically central part of the region shows that seams all patch related and many others are empirically associated types are equally adjacent to all other patch types. due to the consistent integration of different aspects Most of the other part appears increasingly uneven of structure in real landscapes. In fact, few representa- (Figure 4). tive metrics could be adequate to capture the land- Figure 5 is the map showing the pattern of the scape pattern. Griffith et al. (2000), for instance, landscape metrics based on the factor loading scores analyze the Kansas area landscape structure by using of the selected forest class-level landscape metrics. The different data reduction and quantification techniques, first two are from patch size and density landscape Schindler et al. (2008), Herzog et al. (2001) and Lausch metrics (MPS and PSSD) that have an almost similar and Herzog (2002) also used factor analysis. The cor- pattern, a high value of the score was dominantly relation analysis in this study showed that the mere visible around the southwestern and north-central relationships of the metrics in describing the Addis part of the study area, and it extends from south- Ababa and the surrounding city region’s landscape at west to north-eastern side continuously at a different class level (forest) and the overall landscape level. The scale. The third and fourth represent the shape cate- correlation also helps to reduce highly associated gory of landscape metrics represented by MPFD and metrics and make the data suitable for factor analysis, MSI. Though these metrics high score area have the which enables to choose the underlining landscape same pattern and extends southwest to northeast metrics for monitoring landscape situation in terms direction, and a very high score value of these metrics of land-use pattern and structure. When compared are dispersed different part of the region with with studies conducted in other ecosystems a varying level of gray tone spots. However, this high (Schindler et al., 2008) it has a similar amount of the gray ton spots are mainly visible in the west central, overall variance explained by the first four factors of 168 A. MOHAMED ET AL. Table 3. Summary statistics of 39 computed metrics for hex- variance. Sufficient research works were not made in agonal regions at landscape and class levels. quantifying the landscape structure and patterns in Landscape Level Summary Forest Class Level summary Ethiopia and most of the African tropical landscape. statistics statistics Nonetheless, various assessment methods and results Metrics Mean SD CoV Mean SD CoV produced in other parts of the world can help signifi- MPI 520.4 235.4 45.2 1800.7 2395.9 133.1 cantly to refer particularly for a warm and cool tem- MNN 44.8 10.1 22.6 60.9 50.9 83.7 IJI 76 14.9 19.6 52.8 21.1 40 perate highland landscape of the tropics like the study TLA 2492.2 13.3 0.5 2498.3 6.2 0.2 area Addis Ababa and the surrounding. NumP 1336.2 358.5 26.8 34.1 44.7 131 MPS 2 0.8 37.6 29.4 41.6 141.4 The landscape-level landscape structure analysis of PSCoV 1600.8 575 35.9 325.2 193.4 59.5 the study area mainly characterized by patch density, PSSD 33.1 17.8 53.8 88.2 104.2 118.1 TE 493,516.2 123,886.7 25.1 55,856.4 45,066.5 80.7 shape and interspersion categories of landscape ED 198.1 50.2 25.3 22.4 18 80.6 metrics which are represented by PSSD, AWMSI, MSI 1.4 0.1 5.3 1.4 0.2 13.2 AWMSI 9.5 3.7 39.3 3.8 2 52.4 MSI, and IJI. The size and patch density metrics help MPFD 1.1 0 0.7 1.1 0 1.8 to evaluate the various attributes of patches within the AWMPFD 1.3 0 2.4 1.2 0.1 5.3 ecosystem. As Cook, Yao, Foster, Holt, and Patrick SDI 1.1 0.2 19.4 SEI 0.8 0.2 18.6 (2005) indicated that large patches have a large benefit TCA 1037.8 343.7 33.1 373.5 381.1 102 and small patches have small supplementary benefit. CAD 19.4 7.2 36.9 0.9 0.9 103.8 MCA 3.2 4.3 133.7 44.4 98.5 221.9 The study area landscape mainly characterized by CASD 39.2 37.5 95.7 81.6 109 133.5 shape indices of AWMSI and MSI. Patch size and CACoV 4000.3 1683.4 42.1 283.5 188.6 66.5 TCAI 41.6 13.7 32.9 63.9 25.6 40.2 shape influence a number of basic ecological processes CACV1 2267.6 747.8 33 353.4 210.7 59.6 in a landscape in a particular decrease with increasing CASD1 21.4 16.7 78.1 74.7 95.8 128.3 LPI 41.9 20.4 48.7 15.6 17.1 109.4 human land use (Kim & Stephan, 2005). The AWMSI LSI 24.7 6.2 25.2 3.8 2.1 55.2 calculates the complexity of the patches in the land- MCAI 1.9 0.6 31.5 12.5 11 87.9 scape according to their size and it is a good predictor MCA1 0.9 0.7 79.6 24.7 38.9 157.5 NCA 482.9 177.9 36.8 22 22.9 103.9 of birds and species richness pattern (Schindler, von MSIDI 1 0.3 26.7 Wehrden, Poirazidis, Wrbka, & Kati, 2013). IJI also PR 3.9 0.8 19.8 PRD 0.2 0 19.8 measures the distribution of adjacencies among differ- SHEI 0.8 0.2 18.6 ent patch types or configuration. Shape with the SIEI 0.8 0.2 18.5 MSIEI 0.7 0.2 27.1 arrangement of patches of different land-use and CA 480.5 436 90.8 cover types is used together with areal statistics-to %LAND 19.2 17.4 90.7 C%LAND 14.9 15.2 102 quantify the landscape structure and composition. DLFD 1.3 0.1 9.4 The forest class-level landscape structure of Addis Ababa and the surrounding city region is highly char- acterized by patch size and density and shape land- landscape-level analysis. However, other studies scape metrics. The analysis depicted MPS and PSSD as (Griffith et al., 2000) that contain greater than four- the main representative of patch size and density cate- factor component had explained the very little gory. Shi et al. (2008) describe the MPS as an indicator Figure 3. Correlation coefficient of all computed landscape metrics (a) landscape-level correlation matrix and (b) forest class-level correlation matrix. GEOLOGY, ECOLOGY, AND LANDSCAPES 169 Table 4. Factor varimax rotated matrix for the selected class- and landscape-level metrics. Landscape level metrics Forest class level metrics F1 F2 F3 F4 F1 F2 F3 F4 MPI 0.846 0.361 MPI 0.325 −0.425 0.475 MNN MNN 0.772 −0.537 IJI IJI 0.915 TLA TLA 0.331 0.542 NumP 0.31 0.919 NumP −0.855 −0.367 MPS 0.896 −0.342 MPS 0.929 PSCoV 0.731 0.341 PSCoV 0.36 −0.862 PSSD 0.997 PSSD 0.937 TE 0.769 0.533 TE −0.839 0.385 MSI 0.979 MSI 0.341 0.908 AWMSI 0.865 AWMSI −0.911 MPFD 0.881 MPFD −0.314 0.842 TCA 0.868 0.364 SDI −0.672 0.564 MCA 0.683 −0.425 SEI −0.715 0.337 0.462 TCAI 0.628 MCAI 0.799 MCAI −0.351 0.704 PR 0.31 −0.417 −0.757 DLFD 0.596 F = factor loadings. Factor loadings smaller than 0.3 have been removed. Figure 4. Landscape-level landscape metrics pattern. Figure 5. Forest class-level landscape metrics pattern. 170 A. MOHAMED ET AL. of the level of landscape fragmentation or grain, as it 5. Conclusions gets lower with increased disturbance and human Landsat and other similar remotely sensed satellite ima- intrusion. Lausch and Herzog (2002) also discussed geries offer valuable information helps to map the land- MPS as an important element of forest patches that scape configurations that raised the interests of determine habitat functioning, particularly if it is com- environmental planners and managers. Alongside, the bined with core area and neighborhood indices. digital mapping and statistical quantification packages Likewise, the shape metrics help to examine the most brought the LULC maps with a wide range of landscape important dimension of the spatial structure of the metrics that gives-rise to a new conception and techni- landscape in many previous studies (Griffith et al., ques of processing in the field of landscape ecology and 2000; Schindler et al., 2008). This study identified the remote sensing. The rapid development in these tech- shape metrics of MSI and MPFD that determined the nologies, thus, enable to measure the sophisticated landscape shape complexity and the patch fractal levels of landscape structure, ecological functioning dimension. Therefore, it is a vital dimension to predict and attempts were also made to monitor or manage birds and plant species richness pattern (Olivier et al., a very large, multi-jurisdictional landscapes more often 2017). The landscape pattern map of the area again now. The landscape pattern mapping and quantifica- helps to prioritize the specific zone or locality and tion, currently, appears one of the important urban monitor the functions for proper conservation plan- landscape planning and management instrument that ning and management that helps to maintain the helps to measure the extent of fragmentation, network- stated benefits of the landscape analysis. ing, species richness and other functions of the given Generally, landscape metrics are a useful tool for ecosystem or LULC types. The potential of landscape incorporating ecological knowledge in planning and metrics yields information on landscape pattern to management. One of the most important concepts is develop the environmental LULC indicators for differ- that landscape pattern strongly influences the ecological ent parts of the earth. process and characteristics. Moreover, it describes how Ethiopian highlands were one of the important the spatial structure influences the most fundamental examples to study Afro-mountain urban landscape ecological process, and how landscape planning and structures. This study thus has a significant role in the management intern influences the landscape structure. Addis Ababa and the surrounding area’slandscape As discussed above, four class levels and four land- ecology to provide comprehensive knowledge for the scape-level metrics were distinguished pertaining to city region decision-making organ. The analysis mainly the area landscape structure based on their factor used Arc GIS patch analyst, statistical tools to compute loading. The metrics were the best representative of 39 metrics for 178 hexagonal zones of landscape-level a large information generated by other metrics and and 66 hexagonal zones of forest class metrics. Out of form a core set of the structural characteristics of which the correlation matrix computes 35 landscape- landscape monitoring, and they are then helpful to level and 30 forest class-level metrics scoring 0.9, and address the main requirements to apply to the land- the above coefficient of variation was omitted to remove scape structure analysis. repetition. The multivariate factor analysis measured 17 The conservation planning, management schemes class- and 16 landscape-level metrics and the analysis, thus focusing on the forest LULC and the overall then identified 4 forest class- and 4 landscape-level landscape mosaic. As this city region hosts some of metric that has a higher component loading. the indispensable mountain forest and other green Accordingly, it depicted PSSD, AWMSI, MPFD and spaces that support the main landscape services and IJI for landscape-level, and MPS, PSSD, MPFD and goods needs, particularly through nutrient up taking MSI for forest class-level analysis that have to be given and biodiversity maintenance entails the special special emphasis for any landscape monitoring scheme. emphasis in this research. Therefore, any changes in Therefore, the landscape conservation planning and composition and configuration of the landscape monitoring endeavors should focus on the level of demand analysis of a different evaluation technique. patch fragmentation that determines the habitat func- At the same time, the knowledge generated in this tioning level on natural land covers like forest, the analysis also is useful to confront the challenge growth of dispersed built-up and sprawling in urban through different intervention technique. The moun- areas. In addition, it also concerned on maintaining the tain forest and other green spaces in the city region natural landscape complexity that determines the num- have to be maintained, and efforts of land-use man- ber of ecological processes and the extent of human agement should be supported by the landscape struc- intervention. The maps showing the pattern of the ture analysis, and this analysis also helps to measure underlining landscape metrics also portrayed the level the improvements due to the attempts of conservation of priority of space that has to be given for a set of not only in this study area but also the highland area of conservation mechanism and monitoring scheme. the tropical city region’s landscape. GEOLOGY, ECOLOGY, AND LANDSCAPES 171 On the whole, the Addis Ababa and the surrounding experiment and simulation in ecology. Ecological Modelling, 206(3–4), 347–359. city-region landscape structure analysis and mapping of Brown, D., Band, L. E., Green, K. O., Irwin, E. G., Jain, A., its spatial pattern are helpful to maintain the region’s Lambin, E. F., ... & Verburg, P. H. (2013). Advancing landscape ecology as they have a valuable role in biodi- Land Change Modeling: Opportunities and Research versity conservation and maintaining species richness, Requirements. Washington, D.C.: The National increasing nutrient up taking capacity and ecosystem Academies Press. Bruce, J. K., Neale, A. C., Nash, M. S., Van Remortel, R. D., connectivity, etc. In a nutshell, the spatial landscape Wickham, J. D., Riitters, K. H., & O’Neill, R. V. (2001). structure metrics have a vital role to analyze, monitor Predicting nutrient and sediment loadings to streams and providing a comprehensive understanding of from landscape metrics: A multiple watershed study a given landscape pattern. This study also explored the from the United States Mid-Atlantic region. Landscape underlined landscape structure metrics and their impli- Ecology, 16(4), 301–312. cation on the sustainable ecological conservation plan- Chmielewski, T. J., Kułak, A. A., & Malwina, M.-Ś.(2014). How to evaluate and forecast changes in landscape image: ning and management of the tropical highland urban The case of a Small River Valley in Poland. Landscape landscape in general and the study area in particular. Research, 40(4), 466–475. Some of the important concerns of future studies in Cook, E. A. (2002). Landscape structure indices for assessing the area could be stated as follows. Using a very-high urban ecological networks. Landscape and Urban spatial resolution satellite imageries less than the Planning, 58, 269–280. Cook, W. M., Yao, J., Foster, B. L., Holt, R. D., & 30 m helps to identify a comprehensive urban land Patrick, L. B. (2005). Secondary succession in an experi- uses. Urban areas typically exhibit spatially heteroga- mentally fragmented landscape : Community patterns mous LULC. This study considers the whole urban across space and time. Ecology, 86(5), 1267–1279. built-up as one class, however these very high- Retrieved from https://pdfs.semanticscholar.org/f241/ resolution images enable to group further with other 2b450567d7abdec72932ce5dec2b744a6800.pdf classes like residential, commercial and industrial. It is Cushman, S. A., McGarigal, K., & Neel, M. C. (2008). Parsimony in landscape metrics: Strength, universality, also possible to classify the vegetation cover in different and consistency. Ecological Indicators, 8(5), 691–703. types and this may again strengthen the interpretation Davies, S. A., Hernebring, C., Svensson, G., & Gustafsson, L. G. more exhaustively and to measure the class-level (2008). Theimpacts of climatechange and urbanisation on metrics of each LULC classes specifically. In addition, drainage in Helsingborg, Sweden: Combined sewer system. this study mainly analyzes the spatial configuration of Journal of Hydrology, 350, 100–113. Dobbs, C., Escobedo, F. J., & Zipperer, W. C. (2011). the landscape structure by taking a given time. While A framework for developing urban forest ecosystem ser- the temporal dynamics of this landscape is another vices and goods indicators. Landscape and Urban weak-spot of the study, essentially, urban landscape is Planning, 99, 196–206. rapidly altered through time, thus it is necessary to deal Farina, A. (2000). Landscape ecology in action (1st ed ed.). on this aspect. Venice, Italy: Kluwer Academic Publishers, Springer Science+Business Media Dordrecht. Fikirte, D., Kumelachew, Y., Mengistie, K., & Thomas, S. (2017). Land use/land cover changes and their causes in libokemkem Acknowledgments district of South Gonder, Ethiopia. Remote Sensing Applications: Society and Environment, 8(April), 224–230. We gratefully acknowledge the financial support from Addis Griffith, J. A., Martinko, E. A., & Price, K. P. (2000). Ababa University Office of the Director of Research under Landscape structure analysis of Kansas at three scales. the thematic research title “Resource efficiency, environ- Landscape and Urban Planning, 52,45–61. mental quality and sustainability of urban areas in Habtamu, T., Nyssen, J., Zenebe, A., Haregeweyn, N., Ethiopia”, Grant no. TR/11/2013. Kindu, M., Lemenih, M., & Haile, M. (2013). Ecological succession and land use changes in a lake retreat area (Main Ethiopian Rift Valley). Journal of Arid Disclosure statement Environments, 91,53–60. Hakan, A., & Nil, Ç. (2014). Mapping spatial variations of No potential conflict of interest was reported by the authors. land cover in a coastal landscape using pattern metrics. Procedia - Social and Behavioral Sciences, 120,23–30. Herold, M., Scepan, J., & Clarke, K. C. (2002). The use of ORCID remote sensing and landscape metrics to describe struc- tures and changes in urban land uses. Environment and Asfaw Mohamed http://orcid.org/0000-0003-3283-1776 Planning A, 34(8), 1443–1458. Herzog, F.,Lausch,A.,Müller,E.,Thulke,-H.-H., Steinhardt, U., & Lehmann, S. (2001). Landscape metrics References for assessment of landscape destruction and rehabilitation. Environmental Management, 27(1), 91–107. Adamczyk, J., & Tiede, D. (2017). ZonalMetrics – A python Jones, J. A., Swanson, F. J., Wemple, B. C., & toolbox for zonal landscape structure analysis. Computers Snyder, K. A. I. U. (2000). Effects of roads on hydrology, and Geosciences, 99,91–99. geomorphology, and disturbance patches in stream Birch, C. P. D., Oom, S. P., & Beecham, J. A. (2007). networks. Conservation Biology, 14(1), 76–85. Rectangular and hexagonal grids used for observation, 172 A. MOHAMED ET AL. Kevin, F. J., Robert, C. M., & Tait, M. (1986). The application of services. Arboriculture & Urban Forestry, 34(6), exploratory factor analysis in applied psychology: A critical 347–358. Retrieved from https://www.itreetools.org/eco/ review and analysis. Personnel Psychology, 39(2), 291–314. resources/08UFORE.pdf Kim, K. H., & Stephan, P. (2005). Landscape metrics to Olivier, P. I., Rolo, V., & van Aarde, R. J. (2017). Pattern or assess the ecological conditions of city regions: process? Evaluating the peninsula effect as a determinant Application to Kwangju City, South Korea. of species richness in coastal dune forests. PloS One, 12 International Journal of Sustainable Development and (4), e0173694. Retrieved from http://journals.plos.org/plo World Ecology, 12(3), 227–244. sone/article?id=10.1371/journal.pone.0173694 Kirstein, W., & Netzband, M. (2001). Landscape Metrics as a Rempel, R. S., Kaukinen, D., & Carr, A. P. (2012). Patch Tool for the Comparison of Different Urban analyst and patch grid. Ontario Ministry of Natural Regions. Remote Sensing of Urban Areas/ Resources. Centre for Northern Forest Ecosystem Fernerkundung in urbanen Räumen. Regensburger Research, Thunder Bay, Ontario. Geographische Schriften, (35). Schindler, S., Poirazidis, K., & Wrbka, T. (2008). Towards Lausch, A., & Herzog, F. (2002). Applicability of landscape a core set of landscape metrics for biodiversity assess- metrics for the monitoring of landscape change: Issues of ments: A case study from Dadia National Park, Greece. scale, resolution and interpretability. Ecological Ecological Indicators, 8(5), 502–514. Indicators, 2(1–2), 3–15. Schindler, S., von Wehrden, H., Poirazidis, K., Wrbka, T., & Li,X.,Ling,L.,Cheng,G.,&Xiao,H.(2001). Quantifying landscape Kati, V. (2013). Multiscale performance of landscape structure of the heihe River Basin, North-West China using metrics as indicators of species richness of plants, insects FRAGSTATS. Journal of Arid Environments, 48(4), 521–535. and vertebrates. Ecological Indicators, 31,41–48. Livesley, S. J., McPherson, G. M., & Calfapietra, C. (2016). Seto,K.C.,&Fragkias,M.(2005). Quantifying spatiotemporal The urban forest and ecosystem services: Impacts on patterns of urban land-use change in four cities of China with urban water, heat, and pollution cycles at the tree, street, time series landscape metrics. Landscape ecology, 20(7), 871-888. and city scale. Journal of Environment Quality, 45(1), 119. Shi, Y., Xiao, J., & Shen, Y. (2008). Landscape pattern Retrieved from https://dl.sciencesocieties.org/publica change and associated environmental implications in tions/jeq/abstracts/45/1/119 the Haihe River Basin, China. ISPRS J. Photogramm. Lu, D., & Qihao, W. (2004). Spectral mixture analysis of the Remote Sens, 37, 569-574. urban landscape in indianapolis with Landsat ETM+ Simson, A. J. (2017). Landscape structure planning and the imagery. Photogrammetric Engineering & Remote urban forest in polycentric city regions. IOP Conference Sensing, 70(9), 1053–1062. Retrieved from http://open Series: Earth and Environmental Science, 61(1), 012025. url.ingenta.com/content/xref?genre=article&issn=0099- Simonian, L., Hernández Munguía, J. G., Arana Noguera, G. 1112&volume=70&issue=9&spage=1053 A., Thomas, L., Middleton, J., & Phillips, A. (2010). Malinowska, E., & Szumacher, I. (2013). Application of Global forest resources assessment 2010: main report landscape metrics in the evaluation of geodiversity. (No. FAO FP-163). FAO, Rome, Italy, 340. Miscellanea Geographica – Regional Studies On Solomon Mulugeta. (1985). Meeting the Housing Shortage Development, 17(4), 28–33. in Addis Ababa: The Case of Housing McGarigal, K., Cushman, S. A., Neel, M. C., & Ene, E. Cooperatives (Doctoral dissertation, MA Thesis. Addis (2002). FRAGSTATS: spatial pattern analysis program Ababa, Addis Ababa University). for categorical maps. Computer software program pro- Tadesse, A. (2010). Review of the urban environment in Ethiopia in duced by the authors at the University of Massachusetts, 2008. ed. Edwards. Addis Ababa: Forum for Environment. Amherst. Available at the following web site: www. Temesgen,D.,Cruz,F.,Kindu,M.,Turrión,M.B.,&Gonzalo,J. umass. edu/landeco/research/fragstats/fragstats. html, 6. (2014). Land-Use/Land-Cover (LULC) change and socioeco- Mengistie, K., Schneider, T., Döllerer, M., Teketay, D., & nomic conditions of local community in the central highlands Knoke, T. (2018). Scenario modelling of land use/land of Ethiopia. International Journal of Sustainable Development & cover changes in Munessa-Shashemene landscape of the World Ecology, 21(5), 406–413. Ethiopian Highlands. Science of the Total Environment, Temesgen, G., Tulu, T., Argaw, M., Worqlul, A. W., Tolessa, T., & 622–623, 534–546. Kindu, M. (2018). Estimating the impacts of land use/land Mengistie, K., Schneider, T., Teketay, D., & Knoke, T. (2013). Land cover changes on ecosystem service values : The case of the use/land cover change analysis using object-based classification Andassa watershed in the Upper Blue Nile Basin of Ethiopia”. approach in Munessa-Shashemene landscape of. Remote Ecosystem Services, 31, 219–228. Sensing, 5, 2411–2435. Terefe, T., Feyera, S., & Moges, K. (2016). Landscape com- Mesfin, T. B. (2009). Spatial metrics and Landsat data for urban position and configuration in the central highlands of landuse change detection in Addis Ababa (Ethiopia Spatial Ethiopia. Ecology and Evolution, 6(20), 7409–7421. Metrics And Landsat Data For Urban Landuse Change Walz,U.(2011). Landscape structure, landscape metrics and bio- Detection In Addis Ababa). Ethiopia: Universitat Jaume I. diversity. Living reviews in landscape research. 5 (3), 1–35. Mpofu, T. P. Z. (2013). Environmental challenges of urba- Wolde, M. (2013). Changes in regulating ecosystem services fol- nization: A case study for open green space management. lowing establishing exclosures on communal grazing lands in Research Journal of Agricultural and Environmental Ethiopia: A synthesis. Journal of Ecosystems, 2013,1–12. Management, 2(4), 105–110. Retreieved from http:// Wu, J., Jelinski, D. E., Luck, M., & Tueller, P. T. (2000). www.apexjournal.org Multiscale analysis of landscape heterogeneity: scale var- MUDHo. 2016.Riversand RiverBuffer Green Infrastructure iance and pattern metrics. Geographic information Design Standard Implementation, Manual. Addis Ababa, sciences, 6(1), 6-19. Ministry of Urban Development and Housing. Yared, G., Heyaw, T., Pauleit, S., & Mengistie, K. (2019). Nowak, D. J., Crane, D. E., Stevens, J. C., Hoehn, R. E., Urban green spaces supply in rapidly urbanizing Walton, J. T., & Bond, J. (2008). A ground-based method countries:Thecaseof. Remote Sensing Applications: of assessing urban forest structure and ecosystem Society and Environment, 13, 138–149.

Journal

Geology Ecology and Landscapes – Taylor & Francis

Published: Jul 3, 2021

Keywords: LULC; Landsat; remote sensing; city region; green space

There are no references for this article.