.. _example-akva-02:

02 - Irrigation Suitability
===========================

Is this water fit for irrigation? The same 905 analyses as
:ref:`example-akva-01`, asked a question with an established answer: the US
Salinity Laboratory's classification, which judges a water on two hazards at
once and names the result.

Nothing here is computed by a chemical model. Both hazards follow from the
major-ion analysis, which is why this is an analysis of the samples rather
than a simulation of them.


The two hazards
---------------

**Salinity** is read from the electrical conductivity, in four bands:

.. list-table::
   :header-rows: 1
   :widths: 12 30 58

   * - Class
     - EC (µS/cm)
     - Meaning
   * - C1
     - below 250
     - Low. Usable on most crops and soils.
   * - C2
     - 250 to 750
     - Medium. Needs some leaching.
   * - C3
     - 750 to 2250
     - High. Needs drainage and salt-tolerant crops.
   * - C4
     - above 2250
     - Very high. Not suitable under ordinary conditions.

**Sodium** is read from the sodium adsorption ratio, in meq/l:

.. math::

   \mathrm{SAR} = \frac{\mathrm{Na}}{\sqrt{(\mathrm{Ca} + \mathrm{Mg})/2}}

Sodium matters not because it harms the crop directly but because it
displaces calcium and magnesium on the soil's exchange sites, and a soil whose
exchanger is dominated by sodium loses its structure and stops draining. The
ratio is the right form for that: what counts is sodium *relative to* the
divalent cations competing with it, not its concentration. Its four bands,
S1 to S4, slope with EC, because a more saline water keeps its calcium in
solution and does less damage at the same ratio.

A sample's USSL class is the pair -- ``C3-S1``, ``C4-S1`` -- and the *Cation/
Anion Analysis* computes it along with the other irrigation indices: sodium
percentage, residual sodium carbonate, Kelly ratio, magnesium hazard,
permeability index and hardness. They are reported together because no single
one of them decides a case on its own.


Picking out the worst
---------------------

A filter view on the analysis lists just the samples classed ``C4-S1``, the
ones not suitable under ordinary conditions. It filters on the ``ussl_class``
column the analysis computed, which is the useful consequence of the class
being a value in the results rather than only a position on a diagram: it can
be filtered, counted and joined like anything else.


The diagram
-----------

.. figure:: WilcoxCategoryPlot.svg
   :alt: USSL diagram of the samples, coloured by source, all in the C3 and C4 salinity bands
   :align: center

   The USSL diagram, coloured by source. Every sample falls in the high (C3)
   or very high (C4) salinity band, and all but three in the low sodium band
   (S1) -- the points form a flat cloud along the bottom of the diagram rather
   than spreading up it.


What it concludes
-----------------

Salinity, not sodium, limits these waters. 593 samples are C3-S1, high
salinity with low sodium: usable with moderate leaching and salt-tolerant
crops. 309 are C4-S1, too saline for ordinary use. Only 3 reach S2. Source C
is the most saline of the three, with 15 of its 18 samples in C4.

The practical reading is that these waters would be managed for salt --
drainage, leaching fraction, crop choice -- and that the sodium indices, while
computed, are not what decides anything here.


Source
------

* Richards, L. A., ed. (1954). *Diagnosis and improvement of saline and alkali
  soils.* Agriculture Handbook No. 60, United States Department of
  Agriculture, Washington D.C. The salinity and sodium bands, and the diagram
  itself, are taken from this handbook.
* Wilcox, L. V. (1955). *Classification and use of irrigation waters.*
  Circular No. 969, United States Department of Agriculture, Washington D.C.

The data set is synthetic.