GIS: The Essentials - Sampling Spatial Units for Agricultural Surveys - page 61

environment. The installation of the R /GRASS interface is very easy. Within the
R session, you must install the library spgrass6. Once you have installed R and
GRASS, you can directly launch R from the GRASS prompt. Then, to use the
R/GRASS interface, you must load the library spgrass6.
Following these simple instructions it is possible to directly use R in GRASS. In
this way, we can transfer GRASS data to R and run statistical functions on the
imported data as R objects in memory. It is also possible to transfer the results back
to GRASS. The current interface supports raster, vector and site data.
GRASS 6 mainly depends on three other packages: sp, maptools, and
rgdal. In addition to the base package of R , it is also useful to install the following
contributed extensions: akima, fields, geoR, grid, lattice, spa-
tial, spatstat, mcspatial, and spdep (available from the R web site).
Using these packages, it is possible to calculate the spatial statistics measures that
were described in Sect. 1.4 .
Conclusions
This chapter was devoted to the description of the main characteristics of
GISs. GISs have resulted in new approaches for handling and using spatial
data for assessment, planning, and monitoring.
GISs have an increasing role in agriculture production, helping farmers to
increase production, reduce costs, and manage their land more efficiently.
The abilities of GISs for analyzing and visualizing agricultural environments
have resulted in many benefits to those involved in the farming industry.
A GIS can also be used to create more useful data for rural development
policy makers. Small area estimations combined with geo-coded data and
GIS can provide a different view of the poverty distribution at a sub-region
level. Statistical precision can be improved by including agricultural or other
environmental characteristics into the prediction models. Additionally, the
visual nature of maps may highlight hidden relationships that are very
important in a standard regression analysis.
Furthermore, recent crop surveys are based on GIS and spatial sampling
methods (see Chap. 2 ). In fact, the sampling units in crop surveys are based on
an area frame obtained using geographical areas such as villages, cities, and
regions. Census data, survey data, and satellite images are all integrated into a
GIS. Finally, there is a great effort to improve the accuracy of crop area
estimates by incorporating the effect of spatial dependencies through an
integrated application of remote sensing technologies and GISs.
From these simple considerations, it is evident that GISs have a central
role in the definition of statistical surveys for agriculture.