FOREST COVER CHANGE PROCESS AND IMPACT ON CLIMATE CHANGE

Introduction

Changes in LULC can alter the supply of ecosystem services and affect the well-being of humanity The LULC has the potential to influence the biological processes, and alter the provision of ecosystem services. The change in LULC has an impact on hydrological fluxes and greenhouse gas emissions

Forest plays a vital role in regulating climate change through sequestering atmospheric carbon dioxide and mitigates global climate change. Land cover change, especially the conversion of forested areas into other uses has been identified as a contributing factor to climate change, accounting to 33% of the increase in atmospheric CO2 (de Sherbinin,2002). the most significant historical image in land cover has been identified as the expansion of agricultural land at the expense of forest land.

Problem statement

In a local scope, a land cover change problem can be seen in South Nandi forest which could be a very potential carbon store. Unfortunately, the area has been experiencing a high deforestation rate in the recent years thus contributing to Kenya’s Greenhouse gas emissions through land use change and forestry.

As a conservation forest, SNF belongs to the spatial use for protection purposes such as biodiversity conservation, research, ecotourism, recreation and wildlife habitat and therefore the development for cultivation purposes is not allowed at all. However, SNF has encountered a very serious problem and tends to decrease gradually because of human activities.

One of the main causes of the above condition is due to the lack of integrated and detailed spatial information that can support policy making in the rescue of SNF. Important information about the forest cover change spots that lead into deforestation takes place most, the extent, how the process evolves over time, including the drivers and if there is a direct relation with the demand for agricultural crop are not available yet hence it creates difficulties for south Nandi forest management and protection.

Objectives

Main Objective 

To evaluate the GIS capability in assessing the forest cover change in south Nandi forest

Specific Objectives

i. To quantify land cover changes and describe the rates of deforestation for the years 1990-2000, 2000-2010 and 2010-2020.

ii. To assess the major causes of forest degradation in south Nandi

iii. To locate suitable sites for tree plantations for the purpose of reforestation.

Significance of the study

The research analyses forest cover change process, how it leads to deforestation, the rates and locations of deforestation, analysis of the drivers as well as the establishment of suitable tree plantation locations for the purposes of reforestation. It is hoped that this research will assist the government through the ministry of forestry and the management of SNF to develop appropriate planning and strategies and define priorities and actions for conserving

Study area

South Nandi forest is located in Nandi County, Kenya and is shared by three constituencies which are Aldai, Emgwen and Chesumai. It lies within the coordinates of 0° 14’ 3’’ N and 35° 6’ 0’’ E and covers approximately 14000Ha of land. The areas topography ranges from gentle to steep slope with an average elevation of 1800m above the sea level and two rivers kimondu and sirua which merge to form (river Yala) flowing into lake victoria

Data AnalysisLand cover change analysis

A time series analysis was conducted to determine deforestation rate using four different images from 1990, 2000, 2010 and 2020 using ArcGIS software. The selection of periods was based on management changes established in 2010 and to forecast the amount of forest cover for the year 2030 as expected by the National tree strategy policy in Kenya.

In performing the change analysis, 3 main steps were done which were image processing, image classification(maximum likelihood), and change detection

Proximity analysis

Deforestation is highly related to proximity to road and urban areas (Greenberg et al.,2005). the existence of road in the study area creates accessibility to forest and pressure to forest also increase. The deforestation occurrence related to distance from roads was done through spatial analysis in ArcGIS 10.5 through Euclidean distance tool. The distance classes to road was classified into five classes i.e. 0-1km,1-2km,2-3km, 3-4km, and 4-5km.

Demographic factor

The relationship between population increase and forest cover change was analyzed statistically generally, more population lead to more demand for food and land which in turn leads to more forest land being converted. The population data was obtained from Kenya National Bureau of Statistics portal for the years 1999, 2009 and 2019 census

Suitability analysis

Location Selection Criteria for Tree Plantations Establishment

Land cover- Areas that are literally non-forest areas should be targeted for plantation establishment. Thus, crop land was ranked as the first priority and bare/cleared land were considered as second priority for plantation establishment. Other land cover classes such as dense forest, other land (settlements, roads, and infrastructure) and water were ranked as unsuitable areas for plantation establishment.

Slope- the slope was generated from the downloaded 12.5m resolution DEM. Slopes from 0 to 15% were regarded as suitable for artificial regeneration. Slopes from 15 to 30% were ranked second and slopes exceeding 30% were not considered.

Rainfall- rainfall affects the condition of water for plants and soil. Upon addition of the updated Kenya rainfall data, the range of between 1600-200 and 2000-2400mm were obtained hence the amount of 2000-2400 was given a higher value in the weighted overlay

Proximity to roads - Because accessibility influences the cost and success of long-term plantation management, proximity to roads was ranked as the second most important criterion. 0-800 m was the most appropriate condition to have regular access to the plantation site. The range of > 800m - 1600 m was set up as the second-best condition. Beyond that point (1,600 m), it was ranked as unsuitable. To categorize the distance classes from the road, the Euclidean distance (Spatial Analyst) tool available in ArcMap was used.

Weighted Overlay

the last step of the analysis was performing analysis through a “suitability map model.” This model combined the reclassified output of slopes, rainfall, land covers, and proximity to roads. Considering that some criteria have more importance in the suitability model, a “Weighted Overlay” was applied in the Spatial Analyst Tools. The weighted overlay combines several raster’s using a common measurement scale and weighted them on the basis of its importance. After ranking each criterion, the characteristics were reclassified so that each characteristic has a value. In this standpoint, the characteristic that has a higher rank would have a higher value. For the spatial criteria (proximity to roads), after applying Euclidean distance calculation, they were directly reclassified so that each segment of buffer has a value. It is assumed that the closer a land to the roads the better it is for a tree plantation.

FINDINGS AND DISCUSSION

Multi-spectral images from Landsat TM, of 1990, 2000,2010 and 2020 were used to evaluate forest cover changes in the study area. Images were classified into four LULCC. These included; forest, water, agricultural land, and settlements. Land cover/use over the study period within SNF was dynamic by land cover conversion sequences like conversion to agricultural land, and settlements

Agricultural land and settlement were increased while there was a reduction in forest cover. Similar to agricultural expansion, forest logging, firewood harvesting, and increase built up area (settlement) are depleting the forest resources within the SNF priority area. It was asserted that through time series analysis there has been a significant LULCC especially the conversion of agricultural land and settlement at the expense of other LULC classes. decreasing forest cover over time is seen and on the other hand increasing farm/crop land and settlement land while water was more or less constant over the entire period.

land cover changes

Deforestation and deforestation rate

The conversion from forest to non-forest (other land) for each year was classified as deforestation as indicated in the table below with the shaded cells showing the size of deforested land in Ha

Deforestation and proximity to roads

Suitability analysis

In a weighted overlay system, ESRI has illustrated how to value the characteristics in a slope criterion for analyzing a new site. In my study, a value of 1 represents slopes of 0 to 15 degrees, a value of 2 characterizes slopes of 15 to 30 degrees, and a value of 3 symbolizes slopes of 30 to 100 degrees. The values were then reversed such that 0-15 degrees has a higher value of 3 to mean higher importance

In determining the criterion weight in this study, a “Point Allocation method” was applied. in the Point Allocation method, the decision-maker has a budget of points to allocate among the attributes in a way that reflects their relative importance. Clearly, in this method, it is not necessary to normalize the weights since the sum of 100 is already prescribed. The format of the Point Allocation method is presented in the table below

The maps of reclassified input raster’s (criterions) including land cover, slope, rainfall and distance to roads are shown below;

A final suitability map for tree plantation locations for reforestation. 0.23% of the forest land was established to be unsuitable and is covered by water while 57.1% is also covered by dense forest hence unsuitable for reforestation. 35.81% is moderately suitable and the most suitable locations taking up only 6.85% of the forest land.

CONCLUSION AND RECOMMENDATIONConclusion

Information about forest cover change over time and the analyzes of the factors causing this change is important for management and planning of the remaining natural resources in the study area. Based on the observation during fieldwork and data analysis, the research is able to meet the objectives and provide answers and conclusion to the following research questions;

How does forest cover change occur in SNF?

During the period 1990 to 2000, forest cover decreased by approximately 1457.33Ha and was converted to crop (1265.89 Ha) cleared/bare land (291.12Ha) and settlement(76.81Ha)

In the period 2000 to 2010 forest reduced to 12990.79Ha. Total conversion to non-forest was 1397.61Ha while the Conversion to crop was 468.62 Ha and 436.92Ha converted to built up area

From 2010 to the current year (2020), the unchanged was approximately 12434.24Ha and the change between non forest was greater(2963Ha) the total deforestation (Forest to non-forest) for that period was 965.86Ha

Therefore, the change detection shows a gradual decrease over time and increase of crop and built up area while water remained more or less constant.

What are the rates of deforestation for each period?

The rate of deforestation in south Nandi forest in the period 1990 to 2000 is 9.63% deforestation area(1457.33Ha). in the period 2000 to 2010 is 9.71%, deforestation area (1397.61). during the period 2010 to 2020, the deforestation rate is 7.20%, deforestation area being (965.86) the average rate of deforestation for the entire period is (8.84%) with total area of (3820.8Ha). in land cover change 1990-2020, most deforestation occurred in the north western part of the SNF

What are the main drivers of deforestation in SNF?

The main drivers of deforestation in the study area were expansion for agricultural expansion, illegal logging, charcoal burning and firewood collection (direct causes.) the underlying/indirect causes are increasing population, demand for construction materials and weak law enforcement practices.

Which are the most suitable reforestation areas

According to GIS modeling, this study identified sites suitable for tree plantations, ranging from most suitable to unsuitable. Four suitability classes were most suitable taking up 6.65% of the total size of the study area, moderately suitable taking 35.81%, unsuitable (dense forest) taking up 57.1% and unsuitable (water) taking the remaining 0.23% as shown in the suitability map. In other words, this GIS modeling recommended 6.65% of the study area for tree plantations.

Recommendations

Law enforcement must be taken to towards illegal activities in SNF. For the tea and other crop plantations inside the forest area, it is important to propose a win-win solution or land compensation among stakeholder involved

It is necessary to disseminate forest information and implement reforestation programs and REDD (reducing emission from deforestation and Degradation) programs and involve the community in the management and planning phase

Decision-makers in tree plantation planning should be concerned about the desires of the local communities because they know specific details of the environmental conditions around them.