Land Use Land Cover Dynamics and Farmland Intensity Analysis at Ouahigouya Municipality of Burkina Faso, West Africa

Sahel zone has been reported as one of the most vulnerable regions to climate change, so serious attention must be paid to this zone by researchers and development actors who are interested in environmental-human dynamics and interactions. The aim of this study was to bring more insight into the impact of actions aiming at reducing land degradation, regreening the Sahel, stopping population migration and reducing the pressure on land in the Sahelian zone. The study focused on farmland dynamic in Ouahigouya municipality based on remote sensing data from 1986 to 2016 using intensity analysis. The annual time interval change was 0.77% and 2.46% for 1986-2001 and 2001-2016, respectively. Farmlands gained from mixt vegetation, water bodies and from bar lands. Mixed vegetation and water bodies were both active during both intervals while the other land use such as woodland and bar land were dormant. Combining land use land cover analysis and intensity analysis was found to be effective for assessing the differentiated impact of the various land restoration actions.


Introduction
In sub-Saharan Africa, where agriculture is rain-fed and characterized by smallholder farming, farmers developed several strategies to cope with low and variable productivity [1] [2] [3]. Burkina Faso is expected to face significant consequences from climate change, particularly in the Sahelian zone already characterized by scarce water resources and highly vulnerable rural populations. The impact of climate variability and change is felt so severely because livelihood and production systems are so tightly linked to the availability of rainwater particularly in the northern region where Ouahigouya municipality is located [4] [5] [6]. This municipality is known for its high level of land degradation and water scarcity leading to population migration from this zone to the southwest where land fertility and amount of rainwater are far better. The negative impact of climate variability, leading to population migration has retained the attention of decision-makers, development agents, NGO's and researchers. Since the severe drought of 1970, climate-smart agriculture and some practices such as zai, half-moon and irrigation initiatives, have been developed particularly in the Sahelian zone [7] [8] [9].
Many studies have been done on the dynamics of protected forest areas in the country [5] [10] and on land degradation pattern in those areas, which is not the case for farmlands dynamics [11], particularly in the municipality of Ouahigouya where several land restoration actions have been implemented. A better understanding of farmlands dynamics is necessary for policymakers to forecast the impact of climate change on the livelihood of smallholder farmers in the near future. Such understanding will enable the policymakers to anticipate on appropriate adaptation and mitigation actions or strategies [11] [12]. That will contribute to avoiding population migration in search of suitable land. There is, therefore, the need to assess in this area land use land cover dynamics and farmland intensity in order to show inductive impact of various projects and update the national strategy on sustainable land management and restoration. It has been argued that, to better understand and mitigate the possible negative impacts caused by land change, it is essential to detect the trend of land change to better grasp the processes of land change [13]. Reference [14] revealed that research studies concerning land change pattern should be at least focused on why, where and when the change occurred.
This study aimed at responding to the following questions: 1) During which time intervals annual change area is relatively slow or fast? 2) Which land use categories are relatively dormant versus active ones during a given time interval? 3) Which transitions are targeted versus avoided during a given time interval in this area? This study will contribute to bringing more insight into the impact of actions aiming at reducing land degradation in the Sahelian zone.

Study Area
The study was carried out in Ouahigouya (13˚35'00''N, 2˚25'00''W) municipality ( Figure 1). The climate of Ouahigouya is considered to be a semi-arid climate. The average annual temperature is 28.7˚C and the average annual amount of precipitation is 599 mm. The vegetation is characterized by local steppe and agricultural landscape dominated by protected species such as Vitellaria paradoxa, Parkia biglobosa, Tamarindus indica, Adansonia digitata. Breastplates and iron shells are the main soil types in this climatic zone [15].

Data Collection and Analysis
Three Landsat datasets (

Land Use Land Cover Change Dynamics
Land-cover changes analysis map ( Figure 2) and statistics (

Net Change Analysis of Land Use Land Cover
Land cover land use net change analysis showed a significant loss of bar land at 19.10%; 26.51% and 7.41% in 1986; 2001 and 2016, respectively. During the same period, however, the other four land use land cover types: farmland, mixt vegetation, water body and woodland showed both gains and losses (Figure 3).

Time Interval Intensity Analysis of Land-Cover/Land Use Change
The comparison of the land area changes per year with the uniform rate of the change shows that in the time interval between 2001 and 2016, the rate of land cover change was faster (2.46%) than between 1986 and 2001 (0.77%) (

Farmland Transmission Intensity Analysis/Agroforestry Parkland Dynamic Analysis
Intensity analysis showed that mixt vegetation and water body were both active during both intervals while the other land uses such as woodland and bar land were both dormant ( Figure 4).
Between 1986 and 2001, only mixt vegetation was active and was significantly converted to farmland at 0.23% (Figure 4(a)). While from 2001 to 2016 farmland was gaining from four land uses, it was only water body and mixt vegetation which were more active at 0.35% and 0.25%, respectively (Figure 4(b)).

Farmland Dynamic
For both time intervals farmland was gaining more than it was losing and the net rate of gain in the time interval between 2001 and 2016 was higher (0.45%) than between 1986 and 2001 (0.02%) ( Figure 5).

Discussion
Land use land cover dynamics revealed considerable changes over time between 1986 and 2016, although these changes were not time linear. Indeed, land use land cover change rate was higher from 2001 to 2016 compared to the period of 1986 to 2001 (Table 3). This difference in land use land cover change rate can be explained by several factors such as changes in agricultural practices, productivity instability, climate change and most particularly population growth [17].
Looking at specific land use type, net change analysis showed a loss of bar land over the three reference years, but intensity analysis revealed that the change ob-  increase in irrigated areas in the last decade to promote vegetable and rice production in the Sahel zone. Some authors argued that the combination of political, social and economic events encouraged changes in farming systems and land use [19] [20] [21]. Indeed, in the study area lowland planning and conversion into irrigated farmland to promote vegetable and crop production during both raining and dry seasons have been actively supported by the government as well as some projects and NGOs. The loss of bar land to farmland as well as the conversion of lowlands to farmland increase biomass production in the Sahelian zone. This finding corroborates the results of previous works which reported a re-greening of the Sahel [10] [22].

Conclusion
The study revealed that farmland and woodland proportion increased over the three points in time: 1986, 2001 and 2016 at 2.02% to 4.27% for farmland and 7.25% to 12.88% for woodland. Annual land use land cover change during 2001-2016 was faster than that during 1986-2001. The net change analysis showed a significant loss of bar land at 19.10%; 26.51% and 7.41% in 1986; 2001 and 2016, respectively Indeed, the study showed that mixt vegetation and water body were both active during the period of study at 0.35% and 0.25%, respectively while the other land uses such as woodland and bar land were dormant. For time intervals used, farmland was gaining more than it was losing and the net rate of gain in the time interval between 2001 and 2016 was higher (0.45%) than that between 1986 and 2001 (0.02%).