During our 16 year drought currently in California, the rainfall does not matter as much as do the daily stream flow data.  

When drought occur, that lowers the overall values of any rainfall events, so that a meter of annual rainfall after a prolong drought does not translate into the same stream flows, if the same amount of rainfall came during a normal precipitation year.

 And frequent very small rainfall events are usually much more important than the same amount of rain falling in a few big batches.  

And for the California stream flow records that have daily means calculated over many decades time, like you can see at https://waterdata.usgs.gov/ca/nwis/dvstat/?site_no=11164500&por_11164500_8850=2208259,00060,8850 (the creek a few hundred meters from me).  

Today that creek is dry, when normally if should be flowing at a pretty good rate.  And if you see the mean flow for tomorrow, it is predicting a good rainfall for tomorrow, and it happens that the weather for tomorrow is supposed to have rain, as it has for the last 76 years.

So I would focus on daily stream flows as the gold standard to see the ultimate interaction between the rainfall and the local ecosystems and watersheds. 

Lives and flood damage from erosion, sediment, landslides, road system, etc. definately get attention and much effort. But as you suggest, an important consideration, but not necessarily the only thing important to hydrologists and hydrologic modeling or analysis.  

Hydrologists, ecologists and fish biologists recognize the importance of water to life, such as plants, or maintaining flow for aquatic organisms and fisheries between the severe events.  There are various functions going on, perhaps ecosystem synergy in response to rainfall, and the conditions that produce water retention, storage or runoff. The more frequent small events in some of the more humid and wetter climates help to maintain perennial flow or intermittent flow for longer durations than if these storms were not so frequent.  In dry climates with loosing streams, these small storms may show little storm runoff response or continuing flow.  So, yes, I agree the daily or sometimes the instantaneous streamflow record can have detail, importance and function.  

There may be many other uses of the fine detail of stream records, such as indicators of when to irrigate, when to allow logging equipment on plastic soils, when to fish, level in nearby water wells, when to conserve water, or be confident that one might boat or raft certain waters.  In my researchgate record about 2011, I uploaded a technical report in which I was using the detailed flow record to help define the frequency of potential boatable days on a Wild and Scenic River using the long term flow record to avoid days of expanded rish with flows too low or high.  In areas with dry or xeric climates, the detail and response or lack of it associated with the hydrograph may give indicators of just how dry conditions are to adjust activities, conservation measures or storage options, or periods of increased well or water supply monitoring.  

Flow record can have importance to help address erosion, sediment, landslides and storm damage with stabilization or response plans.  Access to storm detail may help determine when to apply a practice or why a practice that normally works failed, suggesting evaluation or adjustment.  The flow record with respect to aquatic populations can help make connections to abnormalties as to species shifts, are they flow related or some other disturbance.

My earliest works evaluated some conditions in Missouri relative to the flow connections with water quality, and how this information could help in evaluating nutrient flux.  Throughout my career for different projects, I have even used the flow records to help determine conditions during a specific period of time, such as when LiDAR flights were capturing data.  The same approach may be used for those who want to assess or refine their knowledge of site conditions when satellite images are taken.  When you are in the field evaluating conditions of flow, it may be valuable to know the present conditions on the long term flow duration curve to better define the conditions of observation.  There are many uses of flow data, and that is why I suggest that the USGS has it right, collect quality flow and other data over time and make it available for public uses, frequency forcasting, etc.

I agree with your point! Technically, at hydrologic modeler's standpoint, small events determine the initial condition (soil water content) of large runoff events. Thus, accurate soil water accounting with consideration of the contributions of small events is critical. Regarding water cycle, rain water came with a small event is likely to be evaporated into the air, cooling down the soil surface, and it may come back to the ground in other places. Small rainfall events also can help crop growth by relieving plant drought stress. I believe it would be much more difficult to accurately predict runoff of a small or mid-size rainfall event that is much more sensitive to soil water content. I saw many large runoff hydrographs closely following the observed at the outlet of a large watershed, but it was rare to see similar accuracy in runoff hydrographs simulated for small and mid-size events in a small watershed. In addition, I know soil moisture condition is not an interest of people working on flood and hydraulic structure design, but we know it controls all other hydrologic processes.

I really like your question!

Young

Craig Carlton Dremann 

 I am really delighted by your sound response on my question with very much useful links you provided. your views are perfact and enriched mine crude thought on this question. I still consider that instead for daily magnitude or daily time step, such small events needs to be visualize or evaluated at further micro time gaps say hourly or even minutes. I know it is vast and volumnous kind of thing but atleast for certain representative events the data (both rainfall, runoff) , its variability patterns and relationships under small events is ememrging as need of hour. may be only a few numbers but we must recognise the importance of such micro data to establish the due importance of those cluster of small events on any big or small catchment.

Thaks again Criag.

may be I get some more help from you or other learned friends here to have access to some of these data where rainfall runoff at minute time intervals is available for those small events.

Regards

 William,

as usual you have wonderfull thoughts and knowledge elements and I am very much thankful to you to go through you enriched points on small events and its due importance . flow records are the true and only signatures of any natural catchment to reveal its real performance under varied inputs. One must take its full advantage from various sources, various experienced people like you and thus give a small twist to think over small events with equal care and importance. They looks to be a sensitive ingredients in any hydrologic system over most of the catchments. but we do have vast scarcity of real ground data in that formate. I am sure you have sailed across tones of such data in your long career. I shall be gratified if you could extend support or share some of these here so  people like me could get enriched and may be it facilitate others specially our young generations to incline towards this school of thought which is very thinly attempted . we need to come down from daily to minute time steps for such small events where active duration of storm remains often less than a hour.

hope to have your further comments and guidance on this. 

Thanks again for a nice answer on my question.

Regards

In the case of the Balearic Islands, where what we have are small catchments with seasonal runoff, after raining events of some importance, what we have are small events that last for 24-72 hours.

In that case, all events are important and must be studied but the lack of data is great. The best approach is to try to work in the field soon after the event is finished or near finished. To gather field data is a key part of the research and then the focus is to get enough rainfall information, using all the possible sources like radar, official data and so on.

I think that, depending of the size of the area where you are doing your research, you should focus in a couple of catchments and study all the available data while you try to get a real-time event to study and then compare with the research data you got.

Regards

I would encourage others to seek out existing rainfall and flow records, especially those with some continuing recordings of data, and nearby to location or in similar weather and physiographic conditions.  The tipping bucket raingauges and water level transducers allow for substantial data collection with infrequent site visits. The use of constructed flumes and weirs can increase accuracy of stage discharge measures on catchment scale, and easier if measures need to be done by a single individual or researcher.  Collecting these records may be costly in time invested, but definately can be a benefit to many in helping to understand more and apply information to our lives, resources and ecosystems.  The fine detail of frequent measures can bring much, and so much more easier to analyze now, than 40 years ago.  

Agreed with your comments but William tipping bucket is more true for rainfall part , but we often need simultaneous pair of rainfall and runoff part at same location at same time. For runoff flume, weirs are not a big constraint as that of real stage discharge relationships and plethora of errors ( during measurements/conversion/analysis). properly calibrated digital means ( may be tipping type, sensor type dataloggers) for runoff part is still in dilution stage of applications when we see from the point of view of small catchments or small events. it is the gap rather a big gap to be given massive thought...what i feel...in regards to your beautiful comment ..   

A watershed system is driven by small and frequent events while also exposed to high or low variates (both flood peaks and low flows) that could be infrequent. In a systems perspective, both the types of the events are important in how the watershed system emerges. I believe the focus on peak or low events in hydrology might be because of the intensity of influence these events have on a watershed. Nice way to look at this question is to look at how we model the watershed dynamics. The continuous time series of precipitation and temperature influences, create a series of dynamic influences that influence the watershed system. My answer is both these events are important and research focus on any study depends on the problem that is being evaluated. 

Yes Randhir  you are very much correct in your valued expressions. Mine question was originally focused towards 'level of importance we hydrologists often offer for looking into small events, which are truely a base or a foundation like stones  or bricks , to conceptualize any kind of modeling or model may an most advanced towering kind of model. To be honest if we look into this aspect, I find a lot of food for thought , which may be  lower in quantitative aspects or dimensions of hydro logic predictions but certainly qualifies for its due importance and equal attention , before jumping upon the so called  models or modeling efforts of current days or time....

In relation to the rains:

Extreme rains are like a person's fingerprint, this represents us, but a person is much more. The rains are much more than their extreme rains. I agree with you.

There is no correlation between 24-hour rains and their intensities of small durations (less than two or three hours) are statistically independent.

The areal distribution of a rain event can not be obtained from one or two rain gauges due to the complex nature of the phenomenon. I have worked with simultaneous rainfall-runoff information in micro-basin of 7 km2 in Quito-Ecuador (8 rain stations and 2 flow stations), the amount of rain is irregular, in an event recorded between 50 mm and 27 mm, in another from 50 mm to 30 mm, in another from 38 mm to 3 mm etc. All this, is independent of the hydrologic abstractions; evaporation, transpiration, interception, depression storage or infiltration.

Expressing an event of rain in hydrology, as a hyetograph is a simplification that some time will be accepted as typical of the twentieth century. The new modeling should be expressed by differential equations in time and space, in a storm scale. That is the first barrier to be faced in the precipitation-runoff models of event to reduce time scale, from day to hours or minutes, within the rain space.

Franklinyou

have very nicely and i guess most appropriately touched the nerve what actually i wanted to express in my question. excellent example that finger prints are not all complete signature or representation of our body. there are enormous parts and all are equally important.  

Just wish to share, as i am excited to see you from Quito Ecuador, I personally was suppose to be there for presenting one of mine lead paper in recently held global conference on watershed  but could not make it possible , though it was accepted for presentation. I wanted to raise such kind of issues there in. 

I will discuss in depth and details with you . please share your contact details for it if you can and also the vital observations records on such small events.

[email protected] is my email.

thanks n regards

Dr Gaur

I agree with you, however the reason for ignoring those events is to simplify our analysis which is already quite complicated. To find if this is really crucial we have to balance between the impact of those events and the outcome to our analysis.

In other-words imagine you have 1 million cubic meter of runoff, would another 1000 cubic meter change the scale of your output? 

In addition small events usually occur in two cases one accompanying large events (after or before) or after a prolonged dry period. On both cases it won't make a huge difference in relation to runoff generation.

To make it short, the objective of your study should govern such argument. In my case I am looking into NPS pollution which require large flow events during certain periods, then It won't make a huge difference to include it.

If you are looking at soil moisture content in relation to runoff generation, then absolutely its a good idea to include it.  

While I was writing, it came to my mind an idea of creating a checklist to govern my decision. 

1- How many events of small size compared to large size ones? What are the precipitation/runoff quantities?

2- Pre event soil conditions 

3- soil characteristics

4- Landuse

- any factor related

Sorry to make it more complicated rather than simplifying! 

Nabil

Nabil

you are very much right in your broad hypothesis and it is the real one too.

second , reaching to a complex situation is the last step to achieve a simplest accurate end product , so there is nothing to express sorry for complex. you have narrated wonderful basic points.

What I meant with my question is basically focused to wards AMC , the antecedent moisture condition  , which still be take in empirical mode and use as direct input for most of the models ..may be it is a simple SCS model or a complicated physical based model...

we must give due importance and due consideration to observe and analysis small events. Who knows it may give big answers for big questions of uncertainties at lower cost .. 

 Mr Gaur,

The query carries much significance and the answer is straight forward. I think every researcher has attempted to solve this type of problem as per the requirement that he/she is interested in. So it has been always resolved being biased towards the requirement.  But the explanation given by Mr. Younggu Her is more convincing to me. I agree with him that both small and large rainfall events are equally important. The small events that occur prior to a large event is more important than the succeeding event as the former becomes a launching pad for an extreme hydrologic condition in case of flood analysis. Take the case of urban hydrology. A small rainfall event when succeeds a larger event, the danger associated is always greater than that of  preceding one as under such situation drainage culverts are likely to fail. 

So the choice for the event lies with the problem you are trying to solve. It is always purpose specific. In my opinion, there is no threshold to define an event, whether small or large with respect to their duration ; whole catchment is contributing or a part; flood or drought problem; effluent flushing or ecological studies, basin scale analysis or an individual hydrologic process  analysis.

 Hydrologic processes are always complex and inseparable. So choice is yours.

Hope for your comment please.