I guess the answer is: not! For trend analisys you need very long time series, i.e. Demetris Kuotsoyanis recomends time series of hundreds years. Nevertheless papers with short time series (40 years) also are published (see attached).

For your very short time series I can recommend to find sincronized, with yours, larger time series and perform the trend analisys with these series and then try to transfer your conclusions to the short ones.

I guess the answer is: not! For trend analisys you need very long time series, i.e. Demetris Kuotsoyanis recomends time series of hundreds years. Nevertheless papers with short time series (40 years) also are published (see attached).

For your very short time series I can recommend to find sincronized, with yours, larger time series and perform the trend analisys with these series and then try to transfer your conclusions to the short ones.

Two years is too short

With two years data, you can only see the seasonal variability of those years, that may not also reflect the reality. As much as you have data, you may expect good trend, if any. If you don't have data, then start from monthly and seasonal variability with your two years dataset. Best of luck. 

Thanks for your comment and I accept that it is quite not possible to go for trend analysis with shorter duration. It is still debatable that how much data (daily, monthly, yearly) is needed to do the trend analysis irrespective of its application. 

Can you please advise me, what another thing can be doable from the continuous time series to infer about the hydrology and its phenomena. I appreciate if you could help me by providing suitable papers also.

As mentioned in the abstract of this document:

"This paper provides formulae for estimating the number of years

necessary to detect trends"

http://onlinelibrary.wiley.com/doi/10.1029/98JD00995/pdf

 you can use Spearman, Kendall, or Pearson methods to detect the trend and get an output, but the output of these methods are not reliable because the outputs may change by increasing the number of data. 

Two years is too short. However, if there are watersheds of similar characteristics and size nearby with longer duration records, you may be able to use the longer term data and prorate to your watershed area.

There are various software models available. Some use Mann-Kendall or Kendall trend type analyses.

A couple of references are:

https://www.sciencebase.gov/catalog/item/57c83ef4e4b0f2f0cec02393

Daily streamflow datasets used to analyze trends in streamflow at sites also analyzed for trends in water quality and ecological condition in the Nation's rivers and streams

From the above, you can link to the report:

https://pubs.er.usgs.gov/publication/sir20175006

Water-quality trends in the nation’s rivers and streams, 1972–2012—Data preparation, statistical methods, and trend results. Scientific Investigations Report 2017-5006.

By: Gretchen P. Oelsner, Lori A. Sprague, Jennifer C. Murphy, Robert E. Zuellig, Henry M. Johnson, Karen R. Ryberg, James A. Falcone, Edward G. Stets, Aldo V. Vecchia, Melissa L. Riskin, Laura A. De Cicco, Taylor J. Mills, and William H. Farmer

https://pubs.usgs.gov/wri/wri00-4130/pdf/wrir00-4130.pdf

Trends in Precipitation and Streamflow and Changes in Stream Morphology in the Fountain Creek Watershed, Colorado, 1939–99. By Robert W. Stogner, Sr. U.S. GEOLOGICAL SURVEY Water-Resources Investigations Report 00–4130

https://pubs.usgs.gov/sir/2014/5053/pdf/sir2014-5053.pdf

Trends in Annual, Seasonal, and Monthly Streamflow Characteristics at 227 Stream gages in the Missouri River Watershed, Water Years 1960–2011.. USGS Scientific Investigations Report 2014–5053.

http://iahs.info/uploads/dms/13709.89-490-495-03-308-Dixon.pdf

The effect of record length on the analysis of river flow trends in Wales and central England, HARRY DIXON, DAMIAN M. LAWLER, ASAAD Y. SHAMSELDIN, PAUL WEBSTER. Climate Variability and Change—Hydrological Impacts (Proceedings of the Fifth FRIEND World Conference held at Havana, Cuba, November 2006), IAHS Publ. 308, 2006.

https://pubs.usgs.gov/twri/twri4a3/pdf/twri4a3-new.pdf

Statistical Methods in Water Resources by D.R. Helsel and R.M. Hirsch. Techniques of Water-Resources Investigations of the United States Geological Survey. Book 4, Hydrologic Analysis and Interpretation, Chapter A3. 2002.

Hi...

Please see this good study issued by USGS in 2000.

https://pubs.usgs.gov/wri/wri00-4130/pdf/wrir00-4130.pdf 

Regards. --- Hilmi S. Salem(Prof. Dr.)

Hai,

Hydrological processes is that the relationship between variation in climate, quantity, intensity and frequency of precipitation which will have prominent effect on flow. The flow duration curve can be developed based on the hydro logical and geological characteristics of the terrain/drainage area. Thus, long term data of minimum 30 years are required for broad assessment of Hydro logical processes. The flow duration curve with data of more than 30 years will be an indicative towards the nature of hydro logical processes that, a flat slope results from stream flow that are well fed by surface water releases or groundwater discharges or gaining stream. Steep slope indicates that the stream are fed directly by the run-off. The slope of the lower end of the duration curve viz of low flow characteristics, shows the behavior of the perennial storage in the drainage basin; a flat slope at the lower end indicates a large amount of storage and a steep slope indicates a negligible amount.

With Best Wishes,

Murali.K.

Dear Dr. Murali

I do all agree with your comment, but question is that, Is the 30 year data is monthly average or daily or annually?

What is rational for using 30 years and more for only trend analysis. 30 Years and more is required for climate change if I am correct. And as far as FDC is the concern it can be prepared by using daily data. 

Please check the papers: 

http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.509.5903&rep=rep1&type=pdf

http://www.tandfonline.com/doi/pdf/10.1623/hysj.53.4.706

http://attiliocastellarin.people.ing.unibo.it/papers/2007_ADWR_Castellarin_et_al_ADWR1080.pdf

To obtain trends with only two years data are not enough. The longer the time-series is, the robustness you will obtain. However, if want to start to understand the hydrogical processes, you can perform an hydrological characterisation of the flow, for example by applying the Richter indexes, characterising the hydrological regime, among others.

In addition, if you have monthly time-series you can assume that the daily distributing is similar to the daily observed distribution and then, you can study a synthetic daily series with the length of the monthly series.

You should check if in your study area you can find world data series (e.g. CRU)

An alternative is to apply a hydrological model. Starting from Precipitation and PET data you can calibrate the model with the two observed years and then extent the flow series by simulating P and PET.

Good luck

I guess it is difficult to find the trend with 2 years data, in principle. One alternative is to identify another hydrologically similar gauging site (with larger time series data) from where the conclusions can be mapped to this site. 

Its not feasible to generalize trend based on only 2 years data. One of the Pitfalls in trend analysis is conclusions from short record lengths. In case if you have other climatic data it may be better to make rainfall-runoff simulation using suitable/calibrated hydrologic model then analyse the trend based on simulated stream flow.

Yes i agree with Alemayehu. You can try with SWMM model.

F

Estimating trend on such a short time period is not reasonable/ recommended and should priovide mislead information.