### Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

### Messages - Karthikeyan L

Pages: 1 ... 3 4 [5]
61
##### Models / Re: Curious case of Flood Frequency Analysis
« on: September 14, 2014, 06:38:46 PM »
First one has to understand the concept of return period. It is simply the inverse of exceedance probability (or relative frequency like you said). A 2 year flood event has a return period of 1/2 i.e., 0.5. It means that there is 0.5 % chance that the event exceeds this value in any one year. And this does not mean that a 2 year flood event occur regularly every 2 years or once in every 2 years. So, the concept of time does not exist at this point. We have to know that return period is only a statistic which we compute from observed data we have. It gives a ball park estimate that this is what is observed once in 2 years. Its the data that speaks through statistic. It may or may not be right.

Say we are analysing for 100 yr event, the data we have may not event contain event pertaining to such a calamity. So, statistical analysis is way to get an answer.

Referring to your first point, before concluding that FFA doesn't consider length of record, one should always keep questioning themselves, what is the need for FFA?. When it comes to floods, FFA is necessary to carry out dam break analysis and associated with that would be reservoir operations (I cant think of anything else!). So, in such a scenario, with current statistical analysis setup, the length of record may not matter (to obtain 100 yr quantile) and I want to go ahead and say having historical dates also may not be necessary.

There are works that conducted FFA at nonstationary setup. Most consider moments to be time varying and determine quantiles as a function of time. Following is a review by Khaliq et al. (2006).
Khaliq, M. N., Ouarda, T. B. M. J., Ondo, J. C., Gachon, P., & BobĂ©e, B. (2006). Frequency analysis of a sequence of dependent and/or non-stationary hydro-meteorological observations: A review. Journal of hydrology, 329(3), 534-552.
It will be interesting to know how much effect inclusion of nonstationarity is bringing to final output. If it is more or less insensitive, I will go with quicker stationary FFA, add sufficient factor of safety and be done with it (remember the question, what is need for FFA?)

62
##### Data / Re: SMOS Data
« on: September 09, 2014, 08:20:11 PM »
Sorry unable to upload them.
Okay I will use those commands and extract necessary region from .dbl files.

63
##### Data / Re: SMOS Data
« on: September 09, 2014, 08:11:38 PM »
Yes 10 simultaneous downloads is doing better job. Thank you.
These are actually zip files which contain two files .DBL and .HDR file. But the link you have sent has commands that deal with .nc files as input (syntax is indicating that). So, will these work on the above file formats I have mentioned?
I am attaching the files for reference.

64
##### Data / Re: SMOS Data
« on: September 09, 2014, 01:22:02 PM »
I am using FileZilla

65
##### Data / SMOS Data
« on: September 09, 2014, 12:52:49 PM »
Hi all,
I am trying to download SMOS L2 soil moisture data for India through two sources.
1) Their FTP server
2) Eoli-sa software
Both have issues, through FTP, the download speeds are very slow and also I have no option but to download for entire globe which is a constraint for me storage wise.
Through Eoli-sa (though I can download of India), I can download only 100 swath records at a time which will end up like 4000 5000 times for the total time period I need.

Can anyone pl let me know a better way of getting this data?

Thank you.

66
##### Post your question/information / Re: Soil Moisture Retrievals
« on: August 29, 2014, 02:40:21 PM »
Yes it took time for me to find information i need from ATBD but i think one can be sure on what is implemented exactly.
I will look for papers.

67
##### Post your question/information / Re: Soil Moisture Retrievals
« on: August 29, 2014, 01:02:29 PM »
I have found documentation (following link) pertaining to potential models, satellite data, ancillary data etc that will be used for SMAP mission.
https://smap.jpl.nasa.gov/science/dataproducts/ATBD/
I will look into the details on vegetation effects and downscaling.
Thank you for the post

68
##### Study material / Books related to mathematical modelling in MATLAB
« on: August 28, 2014, 06:43:00 PM »
I have found a bunch of books that explain mathematical concepts via MATLAB. Some of these include basics for MATLAB.
They may be helpful for wider audience.

Following link directs you to download rar file.

https://drive.google.com/file/d/0BxEtxgjULPpGa3dDQ2ZIaEhTVkk/edit?usp=sharing

69
##### Study material / Re: Radiative Transfer Theory
« on: August 28, 2014, 05:12:22 PM »
Okay I will read SMOS article. Thanks for your help.

70
##### Post your question/information / Soil Moisture Retrievals
« on: August 28, 2014, 04:58:55 PM »
Hi,
Can anyone comment on the scope of research in the area of satellite based soil moisture retrieval studies?

Thank you.

71
##### Study material / Re: Radiative Transfer Theory
« on: August 28, 2014, 04:53:53 PM »
Thank you.
I have read the radiative transfer model section of this article before.
What I felt is that there are slight inconsistencies in the models they are presenting in the articles.
Due to this, I was not able to zero in on what to understand.
This is the reason why I was asking if there are any books available on this topic.

72
##### Study material / Re: Radiative Transfer Theory
« on: August 28, 2014, 04:31:59 PM »
Thank you Sat Kumar.
I was actually searching for models related to AMSR-E data till now.
This article will be helpful.

73
##### Study material / Radiative Transfer Theory
« on: August 28, 2014, 03:59:55 PM »
Hi,
Can anyone suggest some books/articles concerning radiative transfer models for satellite based soil moisture retrieval studies?

Thank you,
Karthikeyan.

74
##### Models / Re: ANN - number of nodes and population size
« on: April 17, 2014, 07:52:20 PM »
Hi,
1) Number of data points:
I think in your case of sine waves, number of data points doesn't matter. During calibration, the network architecture should be trained in such a way that it would have understood the entire range of input data you are providing to it. When you are providing a sine wave, a series that covers entire range values of sine would be sufficient to train the network. Its like a simple regression model only. More you cover the range in calibration and make sure that your validation dataset (you can pick randomly) falls in this rage, you would end up getting decent results from ANN.
2) Number of nodes:
People are still working in this area to find out a method to decide optimum number of hidden neurons. You can take help from works by Prof. Sudheer from IIT Madras. To the extent of my knowledge, its all trial and error. But, you can take range of this number approximately between (number of inputs) to (2 times the number of inputs).
Pl note that along with number of nodes, network architecture and search algorithm influences the output significantly.
Hope I addressed your issue.

Regards,
Karthik.

Pages: 1 ... 3 4 [5]