By: Mario Sandoval

UCLA, Geography

msandoval90@ucla.edu

Introduction:                                                                                                               

The acquisition of arable land for agricultural development has increased dramatically in recent years and has taken on the form of neo-colonialism. These acquisitions or "land grabs" have become increasingly popular for foreign investors and foreign governments looking for large tracts of land for the development of large-scale agribusiness projects (Woodhouse and Ganho 2011). The areas that have been targeted for land acquisition are typically resource rich, yet financially destitute and desperate for foreign investment to help ease the crushing poverty and help the local governments with development projects. The areas that have been most popular for these acquisitions include Sub-Saharan Africa and South America. This project will focus on land acquisition in Sub-Saharan Africa and more specifically in the country of Mali. This recent influx in land grabs has been sparked by the global food crisis of 2008 and has created a frenzy of investment in the hopes of securing the food supply of the investing nations. In the case of Mali, many countries including China, Saudi Arabia, Libya, the UK, USA, France and the Netherlands have all invested in either leasing or purchasing large tracts of land for agricultural development in Mali (Woodhouse and Ganho 2011). All of these investing countries combined have purchased or leased close to 100,000 hectares of arable land for the production of rice, vegetables and Jatropha (Woodhouse and Ganho 2011). Within the context of these land grabs comes a whole host of other environmental and social issues including the appropriation of water for agricultural use, the effects on soil from large-scale mechanization through agribusiness projects, and the displacement of local farmers and whether or not they will gain access to water diversions. For the purposes of this project I am going to focus on the detection in the change in greenness prior to 2008 to the most recent imagery collected and visually demonstrate how these land acquisitions are transforming the local landscape. I will create a time series of images dating from 2004-2012 to show the expansion of agricultural development and the increased use of local resources. Through this analysis of the changing landscape in Mali, I hope to expand the understanding of the impacts that land acquisitions can have on the host nations.

Methods:                                                                                                                    

Study Area: Mali, Sub-Saharan Africa

A landlocked country in northwest Africa that is approximately 480,000 square miles with a population of 14.5 million.

It's primary sources of water are the Niger and Senegal rivers.

 (Baxter 2011)

Our study focuses on the spatial and temporal distribution of agricultural development in Mali and it's water resources. To show the change from 2004 to present we used the USGS Earth Resource Observation and Science Center website to download LANDSAT 7 Enhanced Thematic Mapper Plus (ETM+) satellite images from 2004-2012.

We will use the ENVI software for the purpose of comparing greenness by uploading a series of LANDSAT images from each year from 2004 to 2012 all from the month of October to avoid any seasonal variation that may disrupt the observation of agricultural development. Only bands 1-4 were needed for this study and to cut down on the amount of data being managed. The bands were arranged in a 3-2-1 sequence for the visible spectrum and in a 4-3-4 sequence (the red and infrared bands) to better view the photosynthetic activity. This was done for every image collected for each year. The images displayed some flaws, they had black horizontal lines running across each image. This problem was corrected by replacing the "bad values" and setting all the units to "0" and the image is fixed with just slight variations making the image appear wavy when zoomed closer. We will be using the Normalized Difference Vegetation Index to determine the difference in greenness over the period of time. Two new images had to be created using a 3-4-3 sequence for both 2004 and 2012 because this sequence is more favorable when generating an NDVI. NDVI indexes were created using the function Transform --> NDVI. The red band was set to "1" because it was the first in the sequence, and the infrared band was set to "2" because it was the second in the sequence. From the NVDI images, a change detection image of the greenness index was created. The necessary algorithm to create this image was Basic Tools --> Change Detection --> Compute Difference Map. The 2004 image was chosen first, then the 2012 image, to accurately detect changes in greenness.

 Results:                                                                                                               
The resulting maps shown below demonstrate that land acquisitions can be accurately seen and analyzed from space. Although we are not able to see everything clearly due to inadequate resolution and it is difficult to quantify, it is possible to analyze the change in land cover type, photosynthetic activity, and greenness of areas dominated by the new leaseholders or investing landowners over a period of time by comparing these factors from images of the same site for nine different years.

Google Earth Image: Mali

 Google Earth Image: Area of Development

Visual Spectrum Image: 2004

 Visual Spectrum Image: 2005

 Visual Spectrum Image: 2006

 Visual Spectrum Image: 2007

 Visual Spectrum Image: 2008

 Visual Spectrum Image: 2009

 Visual Spectrum Image: 2010

 Visual Spectrum Image: 2011

 Visual Spectrum Image: 2012

4-3-4 Sequence Image: 2004

 4-3-4 Sequence Image: 2005

 4-3-4 Sequence Image: 2006

 4-3-4 Sequence Image: 2007

 4-3-4 Sequence Image: 2008

 4-3-4 Sequence Image: 2009

 4-3-4 Sequence Image: 2010

 4-3-4 Sequence Image: 2011

 4-3-4 Sequence Image: 2012

Google Earth Image: Focus Area

 3-4-3 Sequence Image: 2004

 3-4-3 Sequence Image: 2012

 NDVI Image: 2004

NDVI Image: 2012

 Change Detection Map: From 2004-2012

Change Detection Map: Focus Area

Discussion:                                                                                 
        The images downloaded from LANDSAT do not show agricultural development on a parcel by parcel basis due to the limits of resolution and the large area of study, but the images uploaded from Google Earth show the same areas and prove that large agricultural development is currently taking place across the entire area of study on a massive scale. The image that shows the "Difference in Photosynthetic Activity" clearly shows how much the development in this area has drastically expanded over the course of nine years. In the Change Detection Map the red coloration symbolizes an increase in greenness, and the blue coloration signifies a decrease in greenness. It appears that there was a certain degree of decrease in greenness, shown by the blue coloration, but the image is more heavily dominated by red, signifying an increase in greenness over the past nine years. Given that the Tomota Group has purchased over 100,000 hectares alone in the area of study, not to mention other agribusiness groups and private investors from foreign countries, it is obvious to make the correlation between agricultural developers purchasing land and the vast expansion of photosynthetic activity (Baxter 2011).
        When I first proposed this project I was curious to see if the difference in agricultural development was visible measurable from space and if it would appear significant just by looking at it. Reading the different reports discussing the land acquisitions occurring across Africa, and particularly in Mali, made me realize that this was a growing trend and I wanted to know how much of an impact these acquisitions would have on the local landscapes and the local people. It is difficult to quantify how people are being affected using remote sensing as a tool, however these images reinforce what the reports have been saying and clearly illustrate the extent and gravity of this situation that is sweeping across much of the developing world. The reports throw around many large numbers, usually 100,000 hectares at a time, but it is difficult to fathom that scale without actually seeing it firsthand and I wanted to produce something that can actually contribute to the research. The reports claim that it is very tricky trying to track who is purchasing or leasing the land and for how much and for how long. They are only aware that it is occurring in these general areas and the documentation has been well concealed by the local governments and by the World Bank both financing and aiding many of the acquisitions (Gurara and Birhanu 2012). All that is known is that this is definitely happening and that the local people once farmed this land that was publicly owned for their subsistence and at the very most for limited gains at market (Baxter 2011). What has happened to the locals is unclear and whether they have been relocated to browner pastures or hired as labor for agribusiness development is not known. Future research needs to be conducted on large scale and must also focus on the socioeconomic implications of these acquisitions. It is difficult to compare the work I have conducted to that of anyone else because it has either not yet been published or is just completely unavailable. In the future, given the large expansion of agricultural development, research should be conducted for the availability of water and the amount of diversions that have been constructed to support such a large expansion. During a period of time outside of the growing season an analysis of the soil quality should be conducted on an annual basis to see how rapidly the soil is being degraded as well as how much fertilization and how many pesticides are being used and washed into the water supply. Due to the fact that I was working alone and the limited time my schedule has permitted, I was unable to delve any further and hope this project can somehow raise interest for future exploration and research.

References:                                                                  

 Baxter, Joan. Understanding Land Investment Deals in Africa Country Report: Mali. Rep. The Oakland Institute, 2011. Web. 18 Nov. 2012. <http://allafrica.com/download/resource/main/main/idatcs/00021029:0d0f31641207deae38bb314ff8a1bccd.pdf>.

 Gurara, Daniel Z., and Dawit Birhanu. Large-scale Land Acquisitions in Africa. Rep. African Development Bank, May 2012. Web. 17 Nov. 2012. <http://www.afdb.org/fileadmin/uploads/afdb/Documents/Publications/AEB%20VOL%203%20Issue%205%20Mai%202012_FINAL.pdf>.

Woodhouse, P., and A. S. Ganho. "Is Water the Hidden Agenda of Agricultural Land Acquisition in Sub-Saharan Africa?" Land Deal Politics Initiative (2011): 30-48.Web. 18 Nov. 2012.<http://www.iss.nl/fileadmin/ASSETS/iss/Documents/Conference_papers/LDPI/12_P_Woodhouse_and_A_S_Ganho.pdf>.

USGS/EROS. LANDSAT (ETM+). 18 Nov. 2012. Raw data. Global Visualization Viewer.<http://glovis.usgs.gov/>.