| dc.description.abstract | Conducting a 1 :5 soil:water extract to measure electrical conductivity (EC) is an approach
to assess salinity and is the preferred method used in Australia. However, the influence of
salinity on plant growth is predominantly based on saturated paste extract electrical
conductivity (ECe) and ECe is recommended as a general method for estimating soil
salinity internationally, so it is necessary to convert EC1:s to ECe, The objectives of this
research were to 1) compare methods of agitation (shaking plus centrifuging
(shaking/centrifuging), shaking, and stirring) for determining EC1: 5; 2) determine optimal
times for equilibration for each method across a range of salinity levels determined from
saturated paste extracts (ECe) (objectives 1 and 2 are for paper 1); and 3) develop
predictive models to convert ECu data to ECe based on four different 1 :5 extraction
methods listed above and a USDA-NRCS equilibration technique ( objective 3 is for paper
2). The soils evaluated for the two studies were from north central North Dakota, USA,
where 20 soil samples having ECe values ranging from 0.96 to 21 dS m-1were used for the
first study (objectives 1 and 2), and 100 samples having ECe values ranging from 0.30 to
17.9 dS m-1were used in the second study (objective 3). In the first study, for each method,
nine equilibrium times were used up to 48 hrs. In the second study, a uniform agitation
time (8 hrs) was applied to the first three agitation methods, and 1 hr was also used for the
USDA-NRCS method. For the first study, significant relationships (p < 0.05) existed
between values ofEC1:s and agitation time across the three methods. Agitation methods
were significantly different (p S 0.05) from each other for 65% of the soils and shaking/centrifuging was significantly different (p < 0.05) from stirring for all soils. In
addition, for 75% of the soils, shaking/centrifuging was significantly different (p :S 0.05)
from shaking. Based on these results, methods were analyzed separately for optimal
equilibration times. The agitation times required for the three methods to reach 95 and 98%
of equilibration were a function of the level of soil salinity. For soils with ECe values less
than 4 dS m·1, over 24 hrs was needed to obtain both 95 and 98% of equilibration for the
three methods. However, less than 3 and 8 hrs were needed to reach 95 and 98%
equilibration, respectively, across methods for soils having ECe values greater than 4 dS
m·1. These results indicate that establishing a standard method is necessary to help reduce
variation across EC1:s measurements. In the second study, the value ofECe was highly
correlated with EC1:s (p < 0.0001) across four agitation methods in non-transformed, log10-
transformed, and dilution ratio models through regression analysis. The values of
coefficient of determination (r2
) were greatly improved and average about 0.87 using log10-
transformation compared to other two models (r2 values of about 0.68 for the nontransformed
models and 0.69 for the dilution ratio models). Since agitation methods were
determined to be highly correlated with each other, any regression model determined under
the four agitation methods were applicable for the estimation of ECe from another method.
The results from this research indicate that comparing data across studies should be done
with caution because both agitation method and time can influence results. Also, estimation
ofECe from EC1:5 can be done with confidence, but models may not be transferrable across
different soil orders or across various salt types. | en_US |
