The Effect of Land Use Changes on Desertification Hazard (Case Study: Maraveh Tappeh Region of Golestan Province, Iran)

Land degradation and desertification caused by land use change is mainly due to human activities in arid and semi-arid and affect on the sustainable use of lands. The aim of this study was to evaluate the effects of land use changes on the desertification hazard in Maraveh Tappeh region. In this research, land use maps of 1986, 2000 and 2014 is provided using images of MSS, ETM and OLI sensors of Landsat satellite and land use map of 2024 is predicted using Markov chain model. According to the results, dense forest area is decreased during study period and with passing time, the area of agricultural land has increased. The results for the time interval of 2014-2028, showed it is possible that will be decreased semi-dense forest and dense rangelands and will be increased other land-use areas according to results of model predictions. In the study years, desertification maps were prepared using ESAs method and with the assumption of fixed all factors other than land use factor. The results showed that there was a lack of desertification as a class in 1986, but this class has been removed in other study years and has been added the area of this class into other classes. The compare of desertification hazard classes and theirs percentage of area in studied time periods showed that critical (C3) and fragile (F3) classes had increasing trend, clearly. The difference between classes since 1986 to 2000 is clearer and more specific and destruction seem clearer.


Introduction
Land use is defined as use and management of land cover under human use and can be defined human intervention in the style of a certain type of land cover to produce, change or preservation of land (Orekan, 2007).Common patterns of land use are changed often as a result of human activity in different temporal and spatial scales.Desertification in arid, semi-arid, and dry sub-humid regions, is a global environmental problem.(Yang et al., 2005).Desertification sensitivity can be defined, in this context, as the response of the environment or part of it to a change in one or more external factors (Smelser & Baltes, 2001).The UN (United Nation) Food and Agriculture Organization (FAO), has announced the agricultural sector causes one-third of global warming and climate change which the result of improper management and land use change (Lal, 2005).Sustainable use of land is under threat due to land degradation in arid and semi-arid that is mostly caused by human activity (Dumanski & Pieri, 2000).Nassauer et al. (1999) and Bartel (2000)  Geometric correction of satellite images were performed by ground control points.Average error for geometric correction was achieved 0.735 pixel that is acceptable.In this study, histogram equalization of spectral values method was used for image enhancement.In order to identify the land uses, the supervised classification was used.Visual interpretation of existing data based on self-recognition and using Google Earth software also used in order to producing proper land use maps.Land use map of 2024 is predicted using Markov chain model that performed by Markov chain command in Idrisi Selva.Map classes of 2024 were carried out based on achieving land use maps from fuzzy classification in 2000 and 2014.In Markov model, land use map of 2000 and 2014 were introduced as the old and new maps, respectively.

Assessment and Mapping of Desertification Hazard
In this study, ESAs method was used for evaluating desertification hazard in the studied period.It is proper model for mapping desertification with regards to purpose of the study and sensitivity of model to land use change.Model parameters were initially revised according to the condition of study area as well as literature review.Then by relying on Geographic Information System, the evaluation and mapping of desertification was performed in studied years.All required data in the model were collected using previous studies or sampling and ultimately entered the Geographic Information System and its algorithms for calculating indices.Finally, sensitivity map of the area to desertification was prepared using these indicators.For mapping desertification hazard of Maraveh Tappeh region in 2014, was used data and information derived from field visits and laboratory analysis and for mapping desertification in other studied years, data of desertification map 2014 (representative of the current desertification hazard) was used with the assumption of fixed all parameters other than land use parameter.Each index is calculated from the geometric mean of its layers according to the following formula (Kosmas et al., 1999) Index The management quality index (MQI) is assessed as the product of land use intensity and the enforcement of policy for environmental protection using the following algorithm.MQI = (land use type * policy enforcement* population pressure) 1/3   Each layer in relation to their impact on desertification received weighing between 1 and 2 that weight 1 and 2 was defined as best quality or no impact on desertification and worst quality or most effective in desertification, respectively.Desertification status in the final map of desertification is divided into 8 subtypes according to Table 2.

Quantification and Classification of Land Use Changes
The overall accuracy and Kappa index for used images is presented in Table 3 that has been achieved more than 80 for all cases and all studied years.These results are indicating high accuracy (above 60%) and reliability of prepared maps.

Analyz
The result hazard of criterion T    Dewan and Yamaguchi (2009) in the study of land use changes  in Dhaka of Bangladesh reported that many constructions were carried out in the study period, as well as a significant decrease in the amount of water, arable land and vegetation cover due to increased urbanization, population growth and economic development.In this study also, forest and rangeland degradation reduced, change and conversion of these land uses area into other land uses are evident that is consistent with the results of Parabaharan et al. (2010) and Zhang et al. (2015).The results showed that vegetation cover is dominant criterion in desertification of Maraveh Tappeh.Vegetation cover affects on status and severity of wind and water erosion directly, this means that with creating roughness on the soil surface, prevent from direct collision of wind and rain into soil surface.In this study vegetation cover is the dominant criterion that is consistent with the results of Sepehr et al. (2007).Giordano et al. (2003) with MEDALUS model illustrated that climate is the most important indicator in desertification of Cicily region of Italy.According to obtained results, weighted average was achieved 1.40 for current state of desertification intensity based on studied criteria that by comparing this amount to the ESAs table, desertification hazard class is calculated critical (C1).This class is calculated F3 for 1986 and C1 fro other studied years.With the passage of time, desertification intensity class increased from middle and weak classes to the severe and very severe classes.In 1986, non-affected is existed as a class but is removed during other years and the area of this class is added to other classes.With moving from 1986 to 2028, the area of theF1, C1 and C2 are decreased and the area of F3 and C3 classes are increased and land degradation is more evident over time.
According to the results, increase of the desertification intensity and trend is due to land use changes and is affected by human factors that is consistent with the results of Roder and Syrbe (2000), Huang and Siegert (2002) and Symeonakis et al. (2007).In confirmation of these findings, Li et al. (2009) indicated that Lake QINGHAI watershed is sensitive to climate change and human challenge and land degradation process is rising in the study period (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004) and illustrated that effective factors on desertification are human activities such as population growth, overgrazing of livestock and wrong policy making in the exploitation of resources.Incorrect policy of resource management such as intensive grazing and over-exploitation of land and water resources and deforestation are the main factors in the destruction of resources and the development of desertification process (Abahussain et al., 2002).The results of Salvati et al. (2013) in Basilicata, Italy showed that region with stable pattern of land-use are lower at risk of desertification than region with land use changes.
Desertification map of studied periods are indicating the change of hazard classes and increase of hazard intensity with the passing of time and these changes are representative the effects of land use changes on desertification hazard.These increase the intensity and class of hazard from 1986 to 2028 represents non-normative and out of the principles of sustainable development management and lack of attention to natural resources capacity.The results of this study can be used in future management for achieving sustainable development, so that the valuable ecosystems and resources in arid and semi-arid regions will be protected.The results of this study indicated wide degradation of natural resources especially in mountain area of this region has been occurred due to intensity of land use changes which indicates the need for paying more attention to these changes by administrative organizations and experts.Also by doing land use planning actions can be controlled land use changes and desertification hazard.

Table 1 .
indicated that landscape pattern and function Properties of used satellite images in this study

Table 2 .
The frequency distribution of intensity classes of desertification status (ESAs)

Table 3 .
Overall accuracy and Kappa index for used imagesLand use maps for 1986, 2000 and 2014 were prepared using satellite images and field observations that are presented in figures 1 to 3. The area of land-use classes for the different studied years has been shown in Table4.In all years, low-density rangeland has largest proportion.With approaching the end of the study period, percent of area in agricultural land increased Table4.

Table 4 .
The Area of Land Use Classes in Studied Years mas.ccsenet t.org

Table 7
has increased during 1976 to 2002 in Ardakan of Iran that is consistent with these results.