This post originally appeared on Sense & Sustainability.
By Patrick Bell, PhD candidate in Environmental Science at Ohio State University and 2014 Next Generation Delegate
Part two of a two-part series on the landscape approach to agricultural development.
As mentioned in a previous post, Landscaping food security, the landscape approach can be beneficial at addressing food security, environmental, and development concerns at a large scale. This post will focus on how a particular landscape approach, the Land Degradation Surveillance Framework (LDSF), can be used for assessing landscape health.
Because the landscape approach differs from traditional agricultural research, new methods of studying at such a large scale are needed. The LDSF is one such method that has been developed and applied throughout many developing countries. However, before discussing the LDSF in more detail, it is important to restate that using a landscape approach means measuring effects and impacts of various land management practices at much larger scales than is used for more traditional agricultural research at the single plot or single farm scale.
The LDSF was designed in such a way that projects can conduct baseline surveys of an area and then subsequently resample and analyze how interventions change landscape health overtime. Therefore, the first step in a LDSF is conducting a baseline survey.
Conducting a LDSF is highly intensive and can take anywhere between 2-4 weeks. A Typical LDSF team would be between 5-10 people and usually includes working closely with local farmers when sampling. The LDSF has been used in many projects throughout the world to assess the impact of various agricultural and conservation development interventions at the landscape scale. While the process involved in conducting a LDSF is very intensive, understanding the basics are not.
LDSF sites cover an area of 100 km2. For development projects, the site is selected in an area of proposed agricultural and conservation interventions. Alternately, the LDSF site can be randomly selected within a specific watershed or region. This 100 km2 area is then separated into 16 tiles that are 2.5km x 2.4km. Inside each of these tiles, a central point is randomly selected and 10 plots of 0.1 ha in size around the central point are sampled. Additionally, each plot has 4 sub plots that are 0.01 ha in size.
Example layout of randomized sampling for LDSF within a 100 km2 landscape. Image Credit: LDSF Field Guide
While in practice it is very important to understand the details of this sampling scheme, for this discussion it is only provided to give an idea of the complexity of working at the landscape scale. It is also important to point out that while this is an extensive sampling protocol, the actual area sampled versus the area within the LDSF (100 km2), is only a tiny fraction.
At the many different spatial scales (plot, sub-plot), samples are taken for subsequent analysis or measured in the field. The measurements taken include soil samples, landform, and land cover classification.
After the baseline survey is complete, the data collected is already immediately useful. The measurements taken give an idea of vegetative cover, soil conditions, prevalence of soil erosion, and topographical information not readily available for most remote locations. However, because of the sampling scheme for the LDSF, the data collected can be used for many other purposes. Soil maps of the area can be created delineating areas where soil nutrients are below critical levels for crop production or where the soil is too dense to allow proper infiltration of water into the soil. These maps based on the LDSF can also be used to identify areas within the landscape where potential interventions would be most needed.
This information can be used as inputs into decision support simulation models, which further evaluate the effect of interventions. Alternatively, the area can be resampled according to the baseline LDSF plan and the data analyzed to determine the real-time effect that these agricultural and conservation interventions have had on landscape health.
A large project is being carried out by the African Soil Information Service (AfSIS), in which they are conducting LDSF surveys at 60 sites throughout Sub-Saharan Africa (SSA). These sites are being conducted in areas (known as sentinel sites) that are representative of the vegetation, climate, and topography of SSA. Ultimately, these sites will be used to assess land degradation throughout SSA as well as to evaluate potential interventions. Because information on soils and their conditions are limited through SSA, the AfSIS effort has allowed the collection and quantification of soil types and conditions throughout SSA. This is of great use to governments, NGO’s, and farmers looking to best manage their soil resources. It is also providing a baseline from which to monitor changes in ecosystem health and soil conditions at a very large scale.
Another project lead by Consultative Group on Agriculture Research’s (CGIAR) program on Forests, Trees, and Agroforestry is using the LDSF within their own sentinel sites throughout the world. In the Nicaragua Sentinel Site, one of the aims of the project is to evaluate what conditions underlie the restoration of vegetative cover by trees. The baseline LDSF survey and subsequent surveys in the future will provide the framework and data necessary to address these key questions.
While there are many different “landscape” approaches, each has to be evaluated against the specific objectives for the landscapes in which they are used. It is easy to see that the LDSF method for quantifying ecosystem health and evaluating changes in ecosystem health with interventions is a very viable option for this. However, it is also very apparent that the “push” by development organizations, governments, and funders for landscape scale projects must also be matched with financial, logistical, and resource support necessary to carry out effective approaches such as the LDSF.
Patrick Bell
Patrick is a PhD student in Environmental Science at Ohio State University and currently serves as a US Borlaug Fellow in Global Food Security. His research focuses on adapting smallholder agroecosystems to future climate change and evaluating agricultural development impacts on ecosystem services at the landscape level.