From the Field to the Landscape: Designing a System
     for Assessing the Impact of Farming Systems on
     Biodiversity using Landscape Metrics - Abstract

      Kalliopi Avanidou1, Dimitris Kavroudakis2, Thomas Alexandridis3, Thanasis
                                      Kizos2
           1
            University of Aegean, Lesvos, Greece; e-mail: kellyavanidou@aegean.gr
                             2
                              University of Aegean, Lesvos, Greece
                          3
                           Aristotle University of Thessaloniki, Greece



Summary

   This paper deals with the effect of agricultural land-use practices on biodiversity.
The case study is an extensive arable farming system on Limnos, a Greek island in
Norther Aegean. The particular farming system combines arable crops grown for food
or feed with grazing lands and sheep husbandry. We present the design of a system
that (a) records farming practices of this system; (b) associates practices with
biodiversity impacts on the bases of extended field work with farmers and
measurements for annual plants and animals; (c) links the field and the landscape level
with the use of landscape-metrics in a G.I.S. environment used to facilitate the
investigation of the relation between landscape structure and biodiversity. The reason
for this is that although farmers apply practices at the individual field and the farm
level, many of the impacts are manifested at the landscape level.
   The system of indicators and landscape-metrics will be implemented in a multi scale
approach: -patch-classes-landscape-. At the field level all single fields of farmers will
be considered as patches. Similar land-cover classes will form the secondary level of
analysis. The landscape level will depend on the geomorphology and the land-cover at
a scale of some square Km. Landscape metrics at the patch level will include diversity,
shape, edge, patch density and size metrics. At the class level, metrics will include the
frequency and size of patches, largest patch index, edge density, shape index mean,
among others. At the landscape level, landscape structure and quantifying changes of
this structure will be calculated with the use of indexes such as: Grain Index, Relative
richness, Menhinich richness, McIntosh diversity, Shannon diversity and Simpson
dominance.
   The system of indicators will be built on (i) an application that farmers will use to
record their practices at the field and farm levels; (ii) a GeoDatabase where spatial and
non-spatial data from the application and the rest of the data will be stored, and
standardized. This GeoDatabase will be accompanied by a Web-GIS system for
geographical visualization and analysis.
   During the application phase of the system, farmers will provide practices’
information for their fields via the application for two cultivation periods in a number
of case study landscapes across the island. Information will be standardized and




                                           26
processed and the selected metrics will be calculated. With the use of these metrics the
impact of these practices and the system as a whole will be estimated and then
presented to the farmers to help them improve their biodiversity performance while
assisting the monitoring inputs/outputs of the use of the application. Finally, this work
contributes towards informed-decision about farming in multiple scales.

       Keywords: Landscape metrics; land -use practices; biodiversity.

       JEL Codes: Q26, M35; L12.




                                            27