Theil-Sen estimated median change in rain normalised soil moisture 2001-2016, Indonesia

Map: Theil-Sen estimated median change in rain normalised soil moisture 2001-2016, Indonesia

Soil analysis

Introduction

The field spectral and biophysical sampling using xSpectre extended spectrometer can be used for both model formulation, calibration and validation. Laboratory analysis is more reliable and can be used for either refined model formulation or for quality checking the results of extension probing.

This post summarises methods used for agricultural soil analysis is Sweden: Jordbruksverket, 2015, Nationell jordartskartering. Matjordens egenskaper i åkermarken. Rapport 2015:19.

Soil sampling and analysis

Soil sampling and analysis can be done in many different ways. It is thus important to save metadata about the sampling and analysis. This manual describes a simply and generic methods for soil sampling that is also the recommended method for applying xSpectre’s spectrometer and app. Apart from xSpectres spectrometer you need a normal owen and a set of muffin or cupcake forms (or wrappers). Preferably made of paper but you can also use silicon forms. Metal forms should s be avoided due to chemcial reactions.

Detailed scale (1g or better)

Field sampling

Soil sampling can be done at different depth and the sample extracted using different methods. The soil sampling for determining soil moisture, pH and nutrient status usually focuses on the topsoil and this manual only deals with the topsoil. xSpectre’s recommended standard sampling depth for agricultural topsoils is 10 to 15 cm (the sample should represent the soil at this depth). The best option is to use an ordinary spade and dig a hole with a vertical wall and extract a known volume of soil from between 10 and 15 cm below the soil surface. If you have access to a soil cylinder that is the bes option. But you can also use a small metall can (any conserved grocery can). Hard soils might require a proper soil sampling cylinder.

When you have extracted the sample (with or without a known volume) store it in plastic bag, preferably one with a zip-lock that also has surfaces for writing. Note the date, place, depth and any other relevant information direcly on the plastic bag. Use permanent ink or an old time pecil. Make sure the platic bag is properly sealed.

The time between digging the hole and extracted the sampel should be short, especially if you want to analyse soil moisture.

xSpectre probing

If you have an xSpectre extended spectromtere with soil probes (e.g. for temperature, moisture, electric conductiviy, dielelectric properties, pH, or any aother ion specific electrod) you should take you readings at the same depth, and next to, where you extract the soil for the laboratory analysis. As you can not use the probes without disturbing the soil you sould not take the field readings from the exact same soil volume that you collect for the lab analysis.

Laboratory analysis

When you return to your “laboratory” (or kitchen etc), the generla sequence of analysis should be:

  1. weigh the total (volumetrically known) sample,
  2. weight the type of contaoiner (platci bag) that the sample came in,
  3. mix each individual sample while crushing large (> 2 mm) aggregates (clumps of small particles glued together),
  4. Sieve the soil and only retain particles with a diameter of 2 mm or smaller,
  5. Extract one or more small amounts of sample (typically 10 to 15 g, usually equal to a table spoon) and put each in a well marked cupcake form with a known empty weight,
  6. immediately weigh the cupcake form with the soil content,
  7. let the cupcake form with the soil sample air dry until the weight does not decrease,
  8. put the cupcake form with the airdried soil sample in the owen for at least one (1) hour at 100 degrees celsius,
  9. Let the cup cake sample cool off to 50 degrees celcius,
  10. Weight the air dried and owen baked soil sample,
  11. Further analysis should be done using the dried samples with known weight, as otlined below for Soil Organic Carbon (SOC) and pH.

Moisture content

The soil moisture content can be estiamted both as the mass of free water (g) over the mass of the bulk soil (g). This equals the net difference of the wet and dried cup cake dried soil. If the original soil sample was collected as known volume, the volumetric content of soil moisture (vol/vol) can be inferred.

Soil Organic Carbon (SOC)

SOC is a very important soil property. SOC is strongly correlated to water holding acapacity and water infiltration capacity, and also to gerneal soil feritiliy and health. Globally accumualted changes in SOC laos has a significant influence on the global balance of Green House GAses like carbon dioxide (CO2) and methance (Ch4). To determine SOC you need to heat up the dried soil sample in a cermaic holder for at least a couple of hours.

The recommended xSpectre method is to take a dried soil sample (approxiamtely 10 g) and heat it up to 500 degrees celsius for 2 hours. As 500 degrees can not be acheived by ordinary household ovens, a abckup method is to heat the sample up for 8 hours at 275 degrees. Heating the sample to hihger temperatures can idnuce chemical reactions that convert mineralized carborn (for example calcium carbonates) to CO2, causing biased results.