ISSN 11222211–11227733, E-ISSN 22006655–33440099 Article no. 226622110055–771144 [617233]

AAnnaalleellee UUnniivveerrssiittăățțiiii ddiinn OOrraaddeeaa,, SSeerriiaa GGeeooggrraaffiiee Year XXXXVVII, no. 22//22001166 (December), pp. 115599–116666
ISSN 11222211–11227733, E-ISSN 22006655–33440099 Article no. 226622110055–771144

http://istgeorelint.uoradea. ro/Reviste/Anale/anale.htm

TTHHEE SSPPAATTIIOO–TTEEMMPPOORRAALL AANNAALLYYSSIISS OOFF UURRBBAANN EEXXPPAANNSSIIOONN OOFF
BBIISSKKRRAA CCIITTYY ((SSOOUUTTHH EEAASSTTEERRNN AALLGGEERRIIAA)) BBYY TTHHEE UUSSEE OOFF
LLAANNDDSSAATT SSAATTEELLLLIITTEE IIMMAAGGEESS

RRaabbaahh BBOOUUHHAATTAA
Université Batna 2, Laboratoire ”risques naturels et aménagement du territoire” Batna, Algérie,
e-mail: [anonimizat]

MMaahhddii KKAALLLLAA
Université Batna 2, Laboratoire ”risques naturels et aménagement du territoire” Batna, Algérie,
e-mail: [anonimizat]

AAiiddaa BBEENNSSEEKKHHRRIIAA
Université Batna 2, Laboratoire ”risques naturels et aménagement du territoire” Batna, Algérie,
e-mail: [anonimizat]

YYaahhyyaaoouuii HHAABBIIBBII
Université Batna 2, Laboratoire ”risques naturels et aménagement du territoire” Batna, Algérie,
e-mail: [anonimizat]

Abstract: Understanding urbanization and better manage ment of the city im plies a historical
perspective of its dynamics. Urban extension whic h today is a major challenge, is materialized
by a rapid urbanization of territories in the marg ins of cities. Its study in a medium sized city
(city of Biskra) is an important issue where it requires monitoring and a detailed analysis of
accurate and geolocalized informations. Our research is part of an approach which is based on
the use of new geomatics techniques for better analyzing the dynamics of urban extensions. In
order to spatialize and evaluate this dynamic, we used a series of Landsat satellite images of
medium resolution with thre e different dates TM 1987, TM 2009 and 2014 for SCEA. The
comparison of results obtained confirm that th e urban area has increased from one year to
another and that the city of Biskra has a medium level of urban sprawl.

Key words: Biskra, dynamic, extension, Landsat, spatiotemporal analysis,

* * * * * *

INTRODUCTION
The urban area of the world has expanded rapidly from 1970 to today. In 2008, half of
the world's population began living in urban areas (Tewolde & Cabral, 2011). This population growth has caused a major expansion at the expense of surrounding farmlands, because the
urban built was not always under control. In Algeria, the rapid growth of urban population
representing 64% of the population in 2008, where it was of 49% in 1987 and 31% in 1966

 Corresponding Author

Rabah BOUHATA, Mahdi KALLA, Aida BENSEKHRIA
160
according to the general census figures of populations and housing (RGPH), this growth caused
a significant uncontrolled extension. The phenomenon of urban sprawl is complex. First of all, it strengthens itself and it is difficult to distinguish its causes from its effects.
The urban sprawl is the advancement of the city on agricultural and forest areas, it is the
transformation of the "natural" zones into built-up areas. Urban sprawl is also a process of social
and functional differentiation of the city, resulting in a heterogeneous distribution and responding
mainly to economic criteria and population activities in the territory. Any form of urban growth is not necessarily synonymous with urban sprawl; the urban sprawl is the propensity of urban areas
grow and develop over much wider areas. While some use this term interchangeably with that of
suburbanization, others distinguish: urban sprawl is an urban extension in continuity with the compact city while suburbanization is an urban extension in discontinuity.
However, urban sprawl, as a concept, is su ffering from difficulties in definition (Barnes
et al., 2001; Roca et al., 2004; Peiser, 2001) believes that the term urban sprawl is used to mean the space of the greedy and inefficient use, the monotonous development without
interruption and discontinuity.
The city of Biskra occupies national ranking of the ninth city of Algeria (in terms of
population) and is ranked at the top of urban stra tum (S.U.S.). The city reached a growth rate
of 2,15% between 1987 and 1998, and a growth ra te of 1,5% between 1998 and 2008; with a
natural increase of 37,08 ‰, reflecting the strong population growth of the city and as its main urban dynamic factor.
The pressure and population growth of the city of Biskra is not only due to natural increase,
other reasons involved in this demographic dynamics; those are the rural exodus, internal and external migration dynamics. On the contrary, the rural population showed a decrease of 68% in 1966 to 33% in
2008 related to the transformation of the relations between town and country (immigration and the
transfer of rural population to urban centers).
Therefore, the extension of the city is faster than population growth and the consumed area per
capita increases. In this framework and to control urban growth and to minimize the caused problems, policymakers should obtain accurate spatial data to detect the growth of this city and develop strategies
for monitoring and environmental management. These strategies must be based on technological
performance (Garouani, 1993; Benmessoud, 2011).
The application of remote sensing has made possible the study of the occupation and land use in
less time, at lower cost and with great accuracy (Kachhwaha, 1985).
Currently, Landsat satellite imagery is an important source of information for observing the
Earth surface because of its digita l nature, repetitive and availability.
The objective of this study is the analysis of spatio-temporal evolution of land use,
especially the extension of the city of Biskra, from the processing of Landsat satellite images of
different dates; 1987, 2009 and 2014.

PRESENTATION OF THE STUDY AREA
The study area is part of the Wilaya of Bisk ra, which is between the Tell and the big Sahara
of Algeria, itoccupies the northern part of the South East region of the country.
The Wilaya of Biskra covers an area of 21, 671.20 km², currently account 12 dairas and
33 municipalities.
Biskra occupies a strategic position that gives it synonymous with desert door of Algeria.
The city of Biskra is part of a fertile oasis to the southwest of the massif of the Aures. It is
considered as an important economic and intell ectual center, even was a trading center for the
nomads of the region.
The cultivation of fruits (dates, olives, apricots and pomegranates) is a dominant
activity, but also its mild climate has made it a popular health resort. The climate of Biskra is
defined by two variable seasons with a short season of comfort; hot and dry in summer, cold and wet in winter.

The Spatio-Temporal Analysis of Urban Expansion…
161

Figure 1. Location of the study area

MATERIALS AND METHODS
We used for this diachronic study three Landsat satellite images of 30×30 m resolution; the
first dating 6 May 1987; the second dated 15 April 2000 and the third dated 16 May 2014. We
opted for a color composite channel 432 for images TM 1987 and 2000, instead of 543 for the image LDCM, 2014, which allows us to cleanly id entify the different units of Land Use compared
to other colored compositions. Based on the visual interpretations and especially the knowledge of
the reality of the field (Bouhata et al., 2015), we defined three classes of land cover, (Buildings, Palmary and bare soils). This choice is made in a way to allow better identification of the city, and
its extension in reference space-time.

Pan-Sharpening
Generally, the Pan-sharpening is defined as a combination of two or more different images
to form using an algorithm a ne w image (Manfred et al ., 2010). The aim of image fusionis to
integrate complementary data in order to obtain more and better information about an object or a
study area than can be derived from single sensor data alone (Sarup & Singhai, 2011).
In our work the pan- sharpening is to merge a panchromatic image (band 8) at high spatial
resolution (15 m) obtained by the LDCM sensor with a multispectral image of medium spatial
resolution (30 m) also obtained by the same sensor and the result produced a multi- spectral image
with the same resolution as the panchromatic image. A supervised classification based on the
method of maximum likelihood, was used on the previous two colored composition images using the
image processing software (ENVI 4.5). This method is considered as a powerful technique for classification. The rule of the decision of this method is based on the probability of a pixel belonging
to a given category (Fojstng, 1999) and (Omar et al, 2014). The average performance of
classifications is 95,2% for the classification of TM 1987, 96,1% for the classification of TM 2000
and 96,3% for the classification of 2014 LDCM.

Rabah BOUHATA, Mahdi KALLA, Aida BENSEKHRIA
162
According to (Rupali & Karbhari, 2015), when 0,81 < Kc <0,99, it means that our results
are statistically perfect (table 1).

Table 1. Validation the results of classifi cation by good values of GP and KC
Matrix of Confusion Global Precision (GP) Kappa Coefficient (KC)
Confusion Matrix 1987 0,952 0,892
Confusion Matrix 2000 0,961 0,932
Confusion Matrix 2014 0,963 0,952

RESULTS AND DISCUSSIONS
The results of supervised classification of Landsat satellite images TM (1987), Landsat TM
(2000) and Landsat LDCM (2014) by the maximum method likelihood will appear as follows:
(figure 2, 3 and 4).
The analysis results to determine the occupati on surfaces for this built class are presented in
the following table (table 2.)

Table 2. Evolution of the area of Biskra city in 1987, 2000 and 2014
Year Area (ha)
1987 1080,03
2000 1311,31
2014 1655,97

Figure 2. Thematic map of land occupancy in Biskra city in 1987

The Spatio-Temporal Analysis of Urban Expansion…
163

Figure 3. Thematic map of land occupanc y in Biskra city in 2000

Figure 4. Thematic map of land occupancy in Biskra city in 2014

Rabah BOUHATA, Mahdi KALLA, Aida BENSEKHRIA
164
The area of the city of Biskra increased from 1080,03 ha in 1987 to 1311,31 ha in 2000 and
over 1655,97 ha in 2014; that makes it of 38,71% of rate. This means that there is an increase in
land demand each year, the Biskra city is in need of 24,5 ha each year for its extension. This
expansion and urban growth is a logical response of a population growth that is marked in both
periods (from 129,611 inhabitants in 1987, 172 905 inhabitants in 1998 to 223 492 inhabitants in
2011). It was accompanied by a national po licy mainly based on the principle of "planned
economy" and "industrialization" . Security conditions that Algeria had known in the 90s led to
record 60% of migratory entries came from the nort hern provinces of Algeri a, 28% from internal
(municipalities and towns of the Wilaya), and 15 % entries from Sahara, which implies a
considerable increase of the urban system of the city. The change detection approach helps to better assess the socio-spatial changes that occur over a territory. To do this, vectorization and
layering Multi dates builts (figure 5) are an exce llent source of information for determining the
size and trend of urban dynamics of Biskra city.
The most interesting extensions are made by building ZHUN (new urban habitat area) to
the east of the city and ZHUN to the west of the city. In the mid-1980, the city has seen an
expansion in the planned tissue of which the most important equipments are: parks area at the ZHUN East, the industrial area at the ZHUN West and the equipment areas to the North of the
city. The important event of this phase is undoub tedly the appearance of the planning land. In 1981
there's been the creation of the National Agency for Planning (ANAT), which deals with the
development of the national plan of land use plan in particular, and in 1990 there was an
enactment of 03 important laws: the Law 90/25 of 11/18/1990 carrying the land orientation, the
law 90/29 of 01/12/1990 relating to planning and development, and the 90/30 law of 01/12/1990 relating estate law. The constitution of the munici pal property portfolio must be in accordance with
urban planning instruments and determining the scope of reserve requirements as a first condition.

Figure 5. Map synthesis of urban sprawl in Biskra City between 1987, 2000 and 2014

The Spatio-Temporal Analysis of Urban Expansion…
165

Figure 6. Highlighting the increased surfaces built in the north ern part of the city of Biskra, in 2013 and 2016

CONCLUSION
The main objective of this study is to test th e ability of this type of multi-date images from
the Landsat satellite to identify the contours of th e urban fabric, which led us to evaluate the
spatial and temporal spread of the city.
The spatio-temporal analysis of urban expansion of the city of Biskra for the period from
1987 to 2014, using space imagery and GIS reveals a major spa tial growth. This extension and the
new urbanization of Biskra urban group is manife sted by the excessive extensions of peripheral
settlements in one hand and in the other hand by the proliferation of informal habitats.
This study illustrates concretely the operationa l advantages offered by the processing of
satellite images, combined with the use of GIS in the approach and analysis of urban areas. Thus,
these techniques are very useful in the absence of updated mapping or rare or unreliable statistics, especially in many developing countries.

REFERENCES

Barnes K.B., Morgan J.M., Roberge M.C., Lowe S. (2001), Sprawl development: its patte rns, consequences and
measurement , Geospatial Research and Education Laboratory, Towson University.
Benmessaoud H. (2011), Use of images alsat1 for the implementation of maps land in semi-arid area – case of the
Aurés Region (Algeria) , Analele Universit ății din Oradea, Seria Geografie, Editura Universit ății din Oradea,
Oradea, pp. 16-22.
Bouhata R ., Kalla M., Bensekhria A. (2015), Contribution of remote sensing in the analysis of the dynamics of Chotts and
Sebkhas: case of the endoreic area of Gadain (Northeast of Algeria) , Analele Universit ății din Oradea, Seria
Geografie, Editura Universit ății din Oradea, Oradea, pp 234-241.
Fojstng E. (1999), Contribution de la Morphologie mathématique à la cartographie de l’occupation du sol a partir
d’image SPOT (Région de l’extrême Nord-Cameroum) , Rapport de doctoral en télédét ection, Réseau télédétection,
AUPELF – UREF, p. 91.
Garouani A. (1993), Cartographie de l’occupation du sol et des zones humi des par télédétection dans la basse vallée de la
Medjerda (Tunisie) , Cinquième journées scientifiques du réseau de la télédétection d’AUPELF – UREF, Tunisie,
21-24 Septembre 1993, p.163-170.
Kachhwala T.S. (1985), Temporal monitoring of forest land for change detection and fore st cover mapping through satellite
remote sensing , Proceedings of the 6th As ian Conference on Remote Sens ing, Hyderabad, pp 77–83.
Khalid Omar M., Shakil A., Romshoo (2014), Determining the Suitability and Accuracy of Various Statistical Algorithms
for Satellite Data Classification , International Journal of Geomatics a nd Geosciences, Vol. 4, n° 4, pp 585-599.
Manfred E., Sascha K., Pä r Johan Å., Pablo R. (2010), Multi-sensor image fusion for p ansharpening in remote sensing ,
International Journal of Image and Data Fusion, 1:1, 25-45, DOI: 10.1080/19479830903561985.
Peiser R. (2001), Decomposing urban sprawl, Town Planning Review, 72, in Xiao jun, urban remote sensing, 167.
Roca J., Burnsa M.C., Carreras J.M. (2004), Monitoring urban sprawl around Barcel ona’s metropolitan area with the aid
of satellite imagery , Proceedings of geo-imagery bridging co ntinents, XXth ISPRS congress, 12–23 July,
Istanbul,Turkey.http://www.ispr s.org/publications/related/semana_geomat ica05/front/abstracts/ Dimecres9/T07.pdf
Rupali R.S., Karbhari V.K. (2015), Performance evaluation of Support Vector Ma chine and Maximum Likelhood classifier for

Rabah BOUHATA, Mahdi KALLA, Aida BENSEKHRIA
166
multiple crop classification , International Journal of Remote Sensing & Geosciences (IJRSG), 4, 1, p. 67.
Sarup J., Singhai A. (2011), Image fusion techniques for accurate classification of Remote Sensing data , International
Journal of Geomatics and Geosciences 2(2):602-612.
Tewolde M., Cabral P. (2011), Urban Sprawl Analysis and Modeling in Asmara , Eritrea, Remote Sensing. 2011, 3, pp.
2148-2165; Doi: 10.3390/rs3102148.

Submitted: Revised: Accepted and published online
April 17, 2016 July 20, 2016 September 21, 2016