What is the unit you want to end with, e.g. m³/s or m³/day?

I have the time series of discharge from zero hours to 23 hours and so on ....

In comec..

Now I would like to compare 2 different time series one is observed daily and one is observed hourly...

In this case I guess it's convenient to convert hourly time series to daily and then compare with daily time series ...

I hope now the question is much clear ..

Yes, to compare the existing daily discharge with your hourly data it makes sense to calculate the mean daily discharge from the hourly data. What you basically have to do is to calculate the mean value of your hourly discharge for one day. That is, take the mean of all hourly data values from 00:00 to 23:00 of all the days. In this case you however assume, that your existing daily values correspond to this time period (00:00-23:00). In some cases daily values are valid for a different time period, e.g. 07:00 - 06:00.

If you have 24 hourly discharge values, i.e. Q1, Q2, ..., Q24, you can simply add them together to get daily discharge. Then you repeat this for each day (if multiple days exist).

You should be caerful about the unit though. For example, if Q1=2 m^3/s it means that in every second of the first hour you had 2 m^3 of discharge. So, the discharge in the first hour was actually Q1=2*60*60 = 7200 m^3/hour. Then after you translated all m^3/s to m^3/day, you can say Qday1=Q1+Q2+...+Q24 and that would be the amount of discharge (m^3) you had in a day (therefore the unit for Qday1 is m^3/day).

@Majed Khadem sir and @Mathew Herrengger sir

Both the answers are totally different one says add the time series and one says the take the average.... It's just the confusing for me

Please can anybody clarify...

Thanks for your valuable comments...

@Praveen:

Can you please provide us a short example of how your hourly discharge values look like? In particular, I am intrested to see what unit is used for your hourly values.

We are actually both correct, it is a question of the units. I was assuming that your hourly and daily values have the unit m³/s or l/s. I agree with Majed - some example including units would be helpful.

@Majed Khadem sir and @Mathew Herrengger sir

here is the sample data (Attached below)

its in Cumec (hourly) i need to convert it in Daily Cumec

Praveen,

you say, that your data is in cumec - this should be the unit m³/s (https://en.wikipedia.org/wiki/Cubic_metre_per_second). Your hourly data tells us, how many m³ of water flow through your cross-section per second in a specific hour. Normally, this high-temporal resolution data are generated by doing many measurements per hour (e.g. every 5 minutes) and then calculating the mean.

In this case, to calculate the daily data, you have to take the mean of the hourly data. The unit stays the same - m³/s, but the time series then tells us, how much water flows through your cross-section per second, in a specific day. This is a mean daily value and has the effect that you dampen peaks observed in the 60-min data.

Attached you can find your modified xlsx-file, in which I calculated the mean daily discharge. Note, around 01.02.2005 you have unrealistic high hourly discharge values, which seem unrealistic. Therefore, I excluded them from the calculation.

One more very important thing: Check the plausibility of your time series - I don´t know anything about the river, where the data comes from - but are the values realistic, if you assume the unit m³/s?

The discharge data thats measured hourly in m3/s can be averaged over period of 24hrs ( average of 24 values) to report mean daily discharge in m3/s.

Thank you so much @Majed Khadem sir and @Mathew Herrengger sir, @ Kshitij Parajuli and all others for your valuable reply.

Praveen,

As you want to compare different timely registered data, depending on the size of the watershed and on the relative location of the stream gages, you may consider comparing the data also on hourly basis. For that you should know how many readings a day and at what time they are done and how the daily data is calculated. During intense rainfall events, if the watershed is small, having quick response, the daily data calculated with only two readings (eg. 7h and 17h as it is done at most gages in Brazil) will result very different from the average of hourly data registered along 24 hours.

with using the daily weighted average, but there are several statistical tools will do the trick

Well,

to convert hourly stream flow discharge to daily discharge, suppose hourly stream flow discharge(Q)= 1m3/hr , then daily discharge is 24 m3/day , conversion factor is multiply by 24.

regards

Interesting question.

Good questions,Informative answers.

If your 2nd time series data is daily mean then get mean for all 24 flows values for each day from your first time series. If 2nd Times series data is daily max then take the max of 24 flow values for each day in first time series.

Softwares like Matlab and R has aggregate functions to get the stats, in excel you can do it by pivot table.

Cheers!

Hi everybody

I think this is a simple question to answer..

When you have an hourly time series (for example 24 hr) you should calculate the total volume of water in that time period (e.g, 24hr) and then divide that volume to desired time period. For example, if the total volume is 24000 m^3 and the total time period is 24hr the average daily discharge will be 1000 m^3/hr or 24000 m^3/day or 0.27777778 m^3/sec..

I hope that works

The best

Azizian

there is one small online tools. just google your statement.

Use the R Package "dynatopmodel"

Quote from the dynatopmodel document

"Time series of observation data are often of different temporal resolutions, however the input to most hydrological models, as is the case with the Dynamic TOPMODEL, requires those data at the same interval. This provides a method to resample a collection of such data to a single interval."

Please share me the best answer might you get...