User Tools

Site Tools


sitedesign:site_fieldsample

From field samples to Maps

To monitor agricultural productivity and soil parameters or map our site for planning new landscape improvements we can use low cost technology. Soil sampling for precision farming and permaculture site planning can be carried out on a regular grid sampling pattern. We are looking at collecting representative samples of the heterogeneity of our site. As a general rule, more spatially heterogeneous is the pattern we are interested in mapping, the more points we would need to acquire. On the other side, if this high heterogeneity is strongly correlate with a soil parameter which we have accurate map, we would need few key samples to precisely map the target feature.
As an example, if we have a high correlation between a long history of agricultural practices (or soil solope) on our site and the soil carbon content of soils, we would need few points describing the correlation between soil carbon content and agricultural land cover type (or soil slope) to precisely describe the heterogeneity of soil carbon in our field.

For our study case we are going to map two key ecological patterns in our farm site:

  • Top soil depth by measuring the depth of the horizon A.
  • Flood prone areas - detecting paddles after rain events.

Field Sampling

Top soil organic depth:
Our field sampling design is organized on a regular grid of core samples extracted north to south every 80m distance and east west distanced of 45m.
Over this grid, we geolocate sampling location geographic coordinates using smarth phones GPS dig the soil and look at coulor / soil texture changes from dark black to yellow clay and also looking at the absence of rooth systems in the clay horizon B.

Flood prone areas
While collecting data from the top soil depth, if any water accumulation paddles are detected, they are geolocated using smarth phones GPS, and we measure the rough x-y-z dimension of paddles.

The csv database we acquire looks as follow:

idsoil depthlatlongsite infogroup
1839.861975-94.376532sparse veg low – Asteraceae1
............


Mapping

From a text table to a vector file

First we need to import our text file database (comma separated value .csv file format is good) into Qgis using the Add delimiter Text layer option. As parameters we need to select:

  • Comma separated file type
  • First record has field names (we have an file header)
  • Geometry definition is point
  • define X-field as longitude
  • Define Y-field as latitude

Output vector maps can be coded according to the soil depth or the group of people sampling in the field.

Once we have loaded the text data as a vector file in Qgis, we use the Select feature tool in Qgis to select only features sampled in the farm site and save those features into a new .shp vector file. Use

  • Select the soil vector file loaded from the text csv file
  • Click on Layer
  • Select Save as
  • Seve as: soilsample.shp/home/user/ost4sem/siteplan/data/vector/soilsample.shp
  • Check the Save only selected features option
  • Define Spatial Reference System as EPSG:26915
  • Click on OK

From point to surface maps

To convert point to surface map we need some approximation algorithms to provide a surface map from sparse point location samples. The interpolation of sparse point to continuous surface can be done using different techniques. GRASS gis and QGIS have different modules to perform this task. Here a good review of GRASS modules for point surface interpolation / resampling methods, and here a guide to the interpolation module in Qgis.

Between the different options we are going to use an Inverse Distance Weighted algorithm for the point surface interpolation. In grass we can process this map typing three lines in a terminal:

  1. Open grass in command line mode
  2. Import the previously created vector map from .shp file to grass vector format
  3. Process IDW algorithm and produce the output map

More info on the options available for the GRASS IDW module.

grass -text ost4sem/grassdb/Missouri/PERMANENT/
v.in.ogr -r dsn=/home/user/ost4sem/siteplan/data/vector/soilsample.shp output=soilsamples
v.surf.idw input=soilsamples  output=soilmapIDW column=soil_depth --overwrite

Topsoil depth map from soil samples interpolated using an Inverse Distance Weighted algorithm.

sitedesign/site_fieldsample.txt · Last modified: 2021/01/20 20:36 (external edit)