Hi Xiao Ming,

it is difficult to say, because I don't know how your land looks like, and your situation is. So I would asked to myself the following questions, in order to get a better understanding of the situation.

Depending on the answers one could say more.

Kind regards

Olaf

Dear Xiao Ming, what is the reason to measure the deformation of so small territory? What is the object?

Hello Xiao Ming,

Is there any trouble with erosion in the area due to groundwater? How long are you expect to measure and monitoring the land deformation and are there any facility around that is going to be hit by this deformation? I fully agree with the other posts too.

Best Regards,

Francisco

Hello Xiao Ming,

In your area place the control points with the required density for your research. Checkpoints measure using an electronic level and / or tachymeter.

Kind Regards,

Michal

It is very interesting problem of study a permafrost behaviour.

I am not sure that classical leveling is appropriate to the study because

it is not easy to place control marks on it.

And guaranty a stability of that marks in relation of the fast changed

permafrost surface.

In my opinion it is better to have the base of two geodetic centers

(points) deep installed in the rock and distanced about 300-500 m each

other. It will be the reference frame for the next surveys. You need to

determine its relative 3D position. It will better to use modern precise

taheometer for the repeated surveys of the so local permafrost slab. You

will receive the number of digital maps distributed in time. It will help

to you for the study.

I do not know how will you measure the soil moisture. I think you have got

some necessary sensors.

Maybe it is important to measure the humidity in the same survey points as for

the topography survey.

So you will collect the necessary measurement data for the research.

InSAR technique maybe will help you too.

But I am not deep evolved in this space survey. And it has some limitations.

The thaheometer approach is ease if you can reach the place without

difficulties.

Hello Xiao Ming,

I think that the main question is the size-scale of the expected shifts/displacements/deformations?

Depending on the expected shifts size you have to choose the method for surveying/determination of the shifts of your area. You have to choose the method which has the accuracy of 10% of the expected shifts. I.e. if the expected shifts are about 10 cm you can choose trigonometric levelling method using total station with accuracy characteristics of angle measurement 1", and distance measurements 1 mm + 1 ppm. If the expected shifts are about 1 cm you have to choose geometric levelling method using analog or digital level with accuracy characteristics of 0.3-0.7 mm/ 1 km (double run, ISO 17123-2, dependent on staff-invar?, technique, parallel-plate micrometer)…

Then you have to make the reference and control network, points outside the area, and points on the observed area. The point stabilization depends on large number of factors, as well as all other things mentioned above, and in other post.

Finally you have to carry out deformation analysis using one of the well-known models (Hannover or Karlsruhe or …) so you can determine if the networks are stable or unstable, if the measurements between the series are homogeneous… Then you will finally be able to determine which points on your area are unstable and what is the amount of their shifts.

Best regards

Hello Rinaldo,

you mentioned the "... well-known models (Hannover or Karlsruhe)". Can you please provide some paper or links to it? Honestly, I have never heard that. To me as a German it is a surprise, I only know, that these are cities in Germany.

Kind regards

Olaf

Hello Olaf,

of course I can. I wrote one paper myself with my colleagues about application of this two deformation analysis models. You can find it in my list of publications and I attached to this post.

King regards,

Rinaldo.

Conference Paper Deformation Analysis of the Holcim Ltd. Cement Factory Objects

Thank you Rinaldo!

I am aware of the Hannover model, but didn't know, that it is called Hannover. One never stops learning.

Best

Olaf

 Dear Rinaldo, were have you receive the critical theoretic values of F criterion? Is it standard Fisher distribution? If yes, is it suits to check the null hypothesis  for not normally distributed values such as deformations? Thank you for the interesting article. Regards, Vladimir

Dear Xiao Ming,

if you are interested in monitoring continuously the area probably the best solution would be GPS (GNSS) positioning. This will give you 3D deformation with accuracy in the order of 1mm for daily solutions (you can also increase the observation rate but that would degrade a little the solution accuracy).

Depending on the availability of a permanent GNSS station close to the study area (

Dear Vladimir,

the critical theoretic values of F criterion I found at this URL: http://www.itl.nist.gov/div898/handbook/eda/section3/eda3673.htm.

Best regards, Rinaldo.

Dear Rinaldo, thank you for the excellent web-resource. But  F-distribution is derived from the Normal (Gauss) one. For this case we need to have both used dispersions of normally distributed values for correct F criterion usage. But we can not expect of Normal distribution for the displacement values. My question is how correct our applying of F-criterion for the not Normally distributed values? I use it too, but do not understand the correctness.

Dear Xia Ming

I should point that that the most important factor in deformation analysis in choosing the method and instruments is accuracy which is not mentioned unfortunately. I should say a leveling accuracy by GPS with an accuracy of 1 mm for a horizontal displacement detection is really optimistic and never ever can expect this accuracy for the elevation element!