ASSESSMENT OF VEGETATION COVER STATUS IN DRY LANDS OF THE SUDAN USING SOCIAL AND TERRESTRIAL DATA

The current study was conducted in 2015 in Bara Locality, North Kordofan, Sudan. The study area has experienced recurrent drought spells since 1970s of the past century. The main objective of this study was to assess and map the vegetation cover in the area using social, terrestrial and remotely sensed data. To accomplish the above mentioned objective, the study was based on qualitative and quantitative data. In qualitative data, household survey was conducted in which 100 respondents were randomly interviewed. Quantitative data was collected using terrestrial inventory and satellite imageries. In terrestrial inventory, 22 ground control points (GCPs) were randomly registered using GPS in order to get general overview of the land cover of the study area. In each GCP, tree species by number was inventoried within an area of 1 ha. Remote sensing data, covering the target study area, were acquainted using LANDSAT5 imageries (2014) with spatial resolution of 30×30 m. Results of the household survey revealed that only 13 shrub/tree species mentioned by 45% of the respondents, while only 9 woody species were identified, belonging to 8 families from terrestrial inventory. The results of the household survey, 45% of the respondents, indicated that vegetation cover was very good 20 years ago. The study categorized the present land cover as woody vegetation (19%), Acacia senegal stands (5%), shrubs i.e. Leptadenia pyrotechnica and Acacia nubica (18%), small scale farms and grasses (19%) and sandy soil and dunes (39%). The results of the land cover distribution indicated that vegetation cover decreased by 24% while sand/sand dunes was increased by 21% from 1985 to 2015. The study concluded that the study area is under threat of land degradation that may lead to depletion of vegetation cover and decline land productivity.


INTRODUCTION
Land degradation has been recognized as a global problem associated with desertification and loss of biological diversity, particularly in drylands (Kellner et al., 2011).Drylands occupy 41% of Earth's land area in which 10-20% of them are already degraded and 1-6% of the dryland people live in desertified areas (Millennium Ecosystem Assessment, 2005).
vegetation cover assessment was applied by many authors (e.g.Chalmers and Fabricius, 2007;Ibrahim, 2013;Khiry, 2007;Khiry et al. 2014). Huntington (2000) noted that the traditional ecological knowledge (TEK) or social survey provides insight knowledge, from resourced users, to understand and predict environmental events upon which the livelihood or even survival of the individual depends.

Study Area
The current study was carried out in Gireigikh Administrative at Bara Locality which is situated at the north eastern part of North Kordofan State between latitudes 13º 34´ to 14 º 47´ N and longitudes 30 º 5´ to 31 º 47´ E covering an area of about 11850 km 2 (Fig. 1).The climate of Bara Locality as described by Elasha et al. (2008)

Data Collection and Analysis
The study was based on qualitative and quantitative data to accomplish the above mentioned objective.In qualitative data, household survey was conducted in which 100 respondents were randomly interviewed on issues related to vegetation cover past and current status in the area.The house field survey was supported by key informant questionnaire in order to provide insight data and confidential information that may not be obtained by household field survey (Kumar, 1989).
Quantitative data was collected using terrestrial inventory and satellite imageries.In terrestrial inventory, 22 ground control points (GCPs) were randomly recorded using GPS in order to get general overview of the land cover of the study area.In each GCP, tree species by number was inventoried within an area of 1 ha.The suitable period for satellite data collection to interpret forested areas in Sudan is the dry season with cloud free conditions from October -May (Adam, 2011;Wallace and Furby, 1994).Additionally, ERDAS Imagine 9.1 was used for imagery pre-processing, processing and classification.Parameters calculated for the vegetation cover included species frequency (F%) and abundance using the following formulas:

Vegetation Cover Status and Trend Based on Respondents' Viewpoint
Respondents, as resource users, are aware with the status of vegetation cover in the study area whether it is developing or deteriorating.More than twenty percent of the respondents declared that vegetation cover in the study area is developing while 58% of the respondents were indicated that vegetation cover is deteriorating (Fig. 2).In this context, respondents could identify six factors that lead to deterioration of vegetation cover.Among them are erratic low rainfall, overgrazing and overcutting.However, four reasons e.g.protection, projects intervention, seed broadcasting and grazing system were acknowledged by respondents to have considerable contribution in improving vegetation cover.The results of the household survey, 45% of the respondents, indicated that vegetation cover was very good before 20 years.On average, more than 70% of the respondents see the current situation is better than before and the landcover is considered good (Table 1).The respondents' judgment about the

Species Diversity and Abundance
Results of the household survey revealed that only 13 shrub/tree species mentioned by 45% of the respondents, while only 10 and 12 of woody and grass species, respectively, were identified during terrestrial inventory with average abundance of 47 stems (  GEF (1994).The presence of Hyphaene thebaica could be attributed to the new plantation for providing saaf (leaves of Hyphaene thebaica) for women to produce handcrafts.The ten tree/shrub species encountered in the study was not different than 11 species enumerated by GEF (1994) in the same area.However, the average abundance (only 47 stems) was less than 280 stems during the  3).The increase in woody vegetation by 3.34 percent could be attributed to the community awareness to the importance of tree cover in their livelihood.
However, the decreasing trend of shrubs and Acacia senegal stands may be refer to erratic and low quantity of rainfall in the study area.Nevertheless, Figure 4 shows increasing trend of rainfall quantity through 1985 to 2015.This trend may encourage small scale farming, the main livelihood activities (Francis, 2010), to increase which has negative impact on vegetation cover.The increasing rate of sand encroachment at the expense of other land cover Condition and type of vegetation cover are important aspects of land cover change for sustainable natural resources management.To do so, vegetation cover assessment has to be done through one or combination of various methods to quantify degradation process.The main objective of vegetation cover assessment is to detect changes over time due to land use patterns.Several methods have been used to assess vegetation cover.Based on the purpose of vegetation or biodiversity assessment, Lee et al. (2005) recognized five methods of monitoring, these are; management inventory/ monitoring, general habitat inventory, status and trend monitoring, surveillance monitoring and research monitoring.Recently, remote sensing technique has been used widely to support other methods in assessing vegetation cover as it considered an efficient tool to investigate human induced changes in land cover and land use (Kumpula et al., 2012).Land-use and land-cover charge associated with agricultural expansion and intensification is the most visible indicator of the human footprint on the biosphere (Gardner et al. 2013).Remote sensing technique is considered a suitable tool for observing vegetation cover change as it provides cost-effective information that plays a useful role in understanding the nature of changes in land cover and at the same time essential for natural resource planners and decision makers (Squires, 2010).
According to Sudan National Adaptation for Action,NAPA (2007), the western parts of Sudan and Kordofan in particular, have experienced recurrent drought spells since 1970s of the last century.As noted byElasha (2006) two types of drought, with varying severity, have affected Sudan.The first, widespread drought is caused by below normal rainfall across the country; the second is localized drought that affects only some parts of the country like Kordofan and Darfur.She further noted that recurrent series of dry years have become normal phenomena in the Sudano-Sahel region, especially in North Kordofan.Consequently, desertification and desert encroachment have threatened thousand hectares which once upon a time considered as productive areas as far as crops, trees, rangelands and biodiversity aspects are concern.The main objective of this study was to assess the vegetation cover status in the area using social survey and temporal data acquired from LANDSAT5 imageries in 1985 and 2015.
is semi arid, average annual rainfall is quite low 250-300 mm with high seasonal and interannual variability.Undulating surface of poor sandy soils and sand dunes with some pockets of clay are the prominent feature of the locality.Vegetation cover can be describes as poor range lands and scatter woody plants dominated by Acacia tortilis, Leptadena pyrotechnica, and Acacia senegal.

Figure 1 .
Figure 1.Location of the study area, Geraigikh Administrative in Bara Locality, North Kordofan-Sudan.
Accordingly, imageries for the current study were acquired representing two dates from TM landsat5(dated 12.01.1985)and TM landsat 8 (dated 20.03.2015) with spatial resolution of 30 × 30 m and covering two scenes (174/50 and 174/51).The mosaic imageries were geo-referenced to the WGS84 datum and Universal Transverse Mercator (UTM) projection.A subset from each mosaic image covering an area of 29978.96ha was created.Visual interpretation was conducted to satellite imageries and comparing the features with their in-situ characteristics in the study area.The GCPs were used for creating training fields to support supervised classification using maximum likelihood method based on Mather and Brandt (2009).Representative training data for each predefined classes was used in the supervised classification.The obtained data was subjected to analysis using Statistical Package for Social Science (SPSS) and Microsoft Excel.
of a species in all plots Total number of plots in which the species ocurred observation in a particular area without real assessment which may consider a subjective tool for assessment.Results obtained from household survey indicate that respondents prefer to plant a certain tree species such as Acacia senegal and Ziziphus spina-christi in their farm lands while Azadirachta indica in their homelands.The main reason behind planting such (multi-purposes tree species) is to provide non-timber forest products (NTFPs) such as gum and fruits as well as fire-wood and building poles.Other reasons were that those tree species require low-input technologies and less labor compared to agricultural crops(Glover, 2005).

CONCLUSIONS
In general, the trend in land cover change at Geraigikh area, Bara Locality will continue since the basic driving forces of land cover change are continue.Such factors are for example expansion of farming activities.Results about vegetation quality from social survey are slightly differ from the terrestrial and remote sensing images, that cover the whole study area, because social survey depends on the individual respondent view point about specific area which might be subjective.Nevertheless, the findings of this study indicate the possibility of integrating qualitative data (social survey) and terrestrial one supported by remote sensing techniques in vegetation cover assessment of the dry lands of the Sudan.

Table 1 .
Trends of vegetation cover based on respondents' (n = 100) viewpoints in Geraigikh Administrative in Bara Locality, North Kordofan-Sudan.

Table 2 .
Occurrence and abundance of some tree/shrubs and frequency of understory species in Geraigikh Administrative in Bara Locality, North Kordofan-Sudan.
classes necessitates certain measures since it threats agricultural farms and livelihood of local population.Figure 3. Classified TM imageries showing land use and land cover classes at Geraigikh Administrative in Bara Locality, North Kordofan-Sudan, during 1985-2015.* = Leptadinia pyrotechnica and Acacia nubica