How would you address discrepancies in mineral composition between field samples and lab analysis reports?

Um processo de amostragem, por exemplo, para fins geotécnicos, deve seguir as seguintes etapas, mandatoriamente: 1 - Planejamento e especifica??o técnica; 2 - Sele??o e identifica??o do local; 3 - Defini??o do método de amostragem, se em superfície ou em profundidade; 4 - Coleta da amostra em campo; 5 - Identifica??o e armazenamento da amostra em campo; 6 - Transporte campo x laboratório; 7 - Armazenamento em laboratório; 8 - Extrus?o do corpo de prova em laboratório; 9 - Ensaio, resultado e análises... e continua até o relatório e defini??es a partir. Os pontos mais sensíveis no processo s?o as etapas 6 e 7, face à integridade da amostra. De nada adianta uma boa amostragem sem o devido zelo na armazenagem e transporte das amostras.

Before making any conclusion, we must insure that the sample used for lab analysis in the lab is homogeneous as there are times when sample preparation does not follow all the step necessary to have effective representative sample for lab analysis.

At one time, I had a series of lab tests where the small sample used in instrument analysis did not necessarily behave in the same manner that a bulk mass of the same material would behave. Especially in time temperature kinetics of bulk materials, it is common to assume that small masses and large masses exhibit the same trends. This gives fallacious conclusions after instrumental analysis. This is especially relevant to metallurgy applications and could potentially apply to field samples of minerals but not in all cases.

Sometimes, the result from dry analysis (handheld analysers) in the field would differ from those of wet analyses (done in the lab). Obviously from the simple fact that wet analysis is able to dissociate all the elements concerned into individual components and hence able to give a more detailed data. Again the expertise of the one carrying out the field analysis and the lab analysis plays a role as well.

There are many assumptions about the conditions of bulk geology samples over time. Even with an extensive knowledge of geology, you still do not fully know that your data analysis is giving accurate information. More knowledge just gives a better guessing framework.

Replication of samples is crucial to making accurate scientific conclusions. Too few samples sometimes skews results by a major percentage. Especially in mineral applications, it's easy to find one outlier in the samples that throws off the entire data set. And it's easy to make an error in test method too.

Establishing a repeatable procedure that easily reduces human error in testing is very important for good quality control. It is especially difficult to take a test designed by knowledgeable experts and simplify the instruction set to something a technician can repeatable and accurately accomplish.

There are also times when it is beneficial to consult people who are not experts in a field for fresh perspective. Experts sometimes create a bit of an echo chamber where they validate thoughts that are not as clearly tested and established as they ought to be, because the experts have more of a reputation and ego complex to preserve. A fresh person looking at an idea does not have all the knowledge, but they look at information more honestly. Experts are less likely at times to question things that are old knowledge.