Monday, October 24, 2011

The presentation I gave last friday in Montpellier

I was guest of Roger Moussa, and in the committee of the graduation of  Dennis Hallema. My talk summarize my efforts in modeling except the recent Boussinesq equation related work. The title was
The main ideas is that we need different types of models for different scopes, and that this models can be implemented with sound informatics, and without having to redo it again from the scratch. The talk, in a sense, complements the post I made on my future research activities, by specifying the "methods" with which I will envision to do them.  Many live for the motto: getting the right answer for the right reason. I support the idea that for getting the right answer you need sound models. The link under the picture will bring you to the presentation. 

Tuesday, October 18, 2011

A collection of Classical and Historical Paper on Geophysical Fluid Dynamics

were posted on the web by G.K Vallis of Princeton University and can be found here.  The collection includes PDFs about

History of Circulation or Dynamics

This section contains papers that give a historical account or interpretation of something, rather than being original research papers in themselves.

Atmospheric Dynamics

This section contains papers on atmospheric dynamic and circulation. The current emphasis is on large-scale phenomena and the general circulation.

Oceanic Dynamics

This section contains papers on oceanic dynamics and circulation. The current emphasis is on meso- and large-scale phenomena and the general circulation.

Geophysical Fluid Dynamics

This section contains papers on dynamics in general, including geophysical fluid dynamics, turbulence theory, instabilities, etc., that is of interest to both oceanographers and atmospheric scientists. It includes papers by Coriolis, Kelvin, Taylor, Ertel, Rossby, Eady, Charney, Phillips, Welander and others.

Climate, Radiation etc

Contains just a very few (at the moment) papers on climate, radiation etc., that are not particularly dynamical. 

It could constitute a nice reading for many.

Marco Borga's talk

Marco concentrated mainly on the interplay between rainfall and the catchment structure, as derived from the data of his Hydrate EU project. Interestingly he introduced (after, he said, initial suggestions from Jim Smith) some spatial moments (i.e. the moments of the spatial distribution of the rainfall with respect to the catchment) to quantify the storm movements during flood events. His results shows that usually storm can be considered almost stationary (on average) over the catchments, but some events are more strongly characterized than other, and especially, concentrating close to the mountain ridge. This cannot be obviously a general result, since around the World the relative position of the relative location of storm to the catchment can be different. What is of general importance is that the space-time evolution of storms can be of some relevance in producing flash floods, and that now we have a statistical tool for quantifying these storms.

 Here it is Marco's abstract on his seminar: Spatial moments of catchment rainfall: rainfall spatial organisation, basin morphology and flood response

 "In this talk I will introduce a general analytical framework for assessing the dependence existing between spatial rainfall organisation, basin morphology and runoff response. The analytical framework builds upon a set of spatial rainfall statistics (termed ‘spatial moments of catchment rainfall’) which describe the spatial rainfall organisation in terms of concentration and dispersion statistics as a function of the distance measured along the flow routing coordinate. The introduction of these statistics permits derivation of a simple relationship for the quantification of storm velocity at the catchment scale. The talk illustrates the development of the analytical framework and explains the conceptual meaning of the statistics by means of application to five extreme flash floods occurred in various European regions in the period 2002-2007. High resolution radar rainfall fields and a distributed hydrologic model are employed to examine how effective are these statistics in describing the degree of spatial rainfall organisation which is important for runoff modelling. This is obtained by quantifying the effects of neglecting the spatial rainfall variability on flood modelling, with a focus on runoff timing. The size of the study catchments ranges between 36 to 982 km2. The analysis reported here shows that the spatial moments of catchment rainfall can be effectively employed to isolate and describe the features of rainfall spatial organization which have significant impact on runoff simulation. These statistics provide essential information on what space–time scales rainfall has to be monitored, given certain catchment and flood characteristics, and what are the effects of space–time aggregation on flood response modeling."

Meeting with Stuart Lane

Stuart was concerned about the connectivity of the river network and to understand how the upstream connectivity affects the downstream hydrology. His message is that restoration or actions taken in a certain position of a river network can easily cause an apparently random result if the information deriving from the netwok connectivity is not accounted for. But, obviously, his arguments have the main focus on the ecology of rivers which is differently affected by the connectivity. Conceptually a step forward.

 Here it is Stuart's abstract on his seminar: Catchment organisation and the flux of water and material 

 "In this seminar I will think about the importance of understanding catchment organization in terms of the functioning of watersheds. Tracer studies have told us, for many years, that catchments organize themselves - the signals that go in to a basin (e.g. rainfall, eroded soil) look very different to those that come out (e.g. river discharge, suspended sediment concentration). But, we have made less progress in understanding the spatial structure that leads to this organization, something that is crucial to prioritizing what to do where in river catchments. I will begin by presenting new ways of modelling this structure, and then show what this might mean: (1) for flood generation; (2) for the effects of land management upon diffuse pollution; and (3) for the watershed organization of salmonid populations. I will conclude by noting that critical to progress in this area is new forms of hydrological conceptualization and the development of innovative experiments to test such ideas."

Stuart's Papers are almost innumerable and can references be found here. At least two are of interest of the discussion we had in Trento:

Lane, S. N., S. M. Reaney, and A. L. Heathwaite (2009), Representation of landscape hydrological connectivity using a topographically driven surface flow index, Water Resour. Res., 45, W08423, doi:10.1029/2008WR007336.

and the invited commentary on Hydrological processes:

Lane, S.N., What makes a fish (hydrologically) happy? A case for inverse modelling, Hydrol. Process. 22, 4493–4495 (2008). DOI: 10.1002/hyp.7145

In the mood of "thinking different" also this paper can bring some interesting information:

Lane, S.N., N. Odoni, C. Landstrom, S J Whatmore, N Ward, and S Bradley (2010) Doing flood risk science differently: an experiment in radical scientific method: doing flood risk science differently, Trans Br. Geogr.

Monday, October 17, 2011

Meeting Roger Moussa research

which I do not have time yet to comment broadly (post should be considered a stub). In the last two months Roger Moussa, Stuart Lane, and Marco Borga came and gave a seminar in Trento. The topic were all centered about the river catchments but everyone of the lecturer had a different accent.
Here it is a little of summary of Moussa talk.

Roger was mainly concerned with agricultural catchments, and their peculiarities. There were two or three themes to enlighten:
 - agricultural catchment derives from a deep intervention of the natural hydrography, and this obviously affects the hydrological response. In one of his paper he does some virtual experiments in which he compares the hydrologic response of the real catchment with the supposed natural one.
 - the variation of hydraulic properties of soil after tillage, and in general, after the agricultural practice. - the role of vegetation, epitomized by a banano cultivation, in preferential collecting the rainfall.

 Here it is the abstract of Roger's seminar: Model calibration and analysis of model performance :Case of distributed hydrological modelling of flood events from the plot to the catchment scale.

 "During the last decades, flood events which occurred in the Mediterranean zone are a major threat to human life and infrastructures. This situation handicaps development, necessitating the use of modeling approaches for prediction of sites prone to flooding, planning of damage minimization activities, and for environmental prediction of the impact on runoff, erosion and pollutant transport. Moreover, hydrological processes are largely variable in space due to human impact in agricultural and urban zones, causing hydrological discontinuities such as channels, field limits, drains, and tillage practices. MHYDAS (Modélisation HYdrologique Distribuée des AgroSystèmes / Distributed Hydrological Modelling of AgroSystems), a physically based distributed hydrological model, was especially developed to model flood events taking into account hydrological discontinuities. Application cases are shown on catchments from the plot scale (1000 m²) to large scales (2000 km²) in various agro-hydro-climatic conditions : i) to assess the optimal subdivision into sub-catchments for distributed hydrological modeling applications; ii) to study the spatio-temporal distribution of rainfall and the soil hydrodynamic properties; iii) to define a parameterisation strategy, and to compare various multi-objective functions and analyze the significance of well-known criteria functions."

With Roger we actually discussed a little more about other issues regarding the topic of comparing data and simulations.  Some considerations derived:

- Discharge data ata should always be checked independently from the outcomes of the model.  To verify their correlation structure, and their statistics, i.e. lag times (discharges from rainfall), centroids, time to peak, volumes, runoff coefficients, etc.

- Be conscious of which data you really use, and distinguish model also for their use of data: do they use vegetation information ? Soil information ? What is really distributed in your catchment data set ?

- Do not use excess of distributed information that you cannot justify.

- Keep in mind which are the objective of your modeling. Improving discharge prediction ? Improving calibration methods ? What else ?

- Do not give for granted that the first period in a data set is the calibration one.  Sometimes invert calibration and validation !!!

- Always use  objective indicator of godness of fit (GOF): but be aware that they can hide some important features, and the best GOFs performances not always means a bette prediction (Roger has a paper on it).

Finally he also, as well as me, felt the need for an infrastructure to support modeling. His institution promote Openfluid, a C++ framework.

Roger's bibliography:

Chahinian N, Moussa R, Andrieux P, Voltz M. 2005. Comparison of infiltration models to simulate flood events at the field scale. Journal of Hydrology, 306: 191-214.

Chahinian N, Voltz M, Moussa R, Trotoux G. 2006. Assessing the impact of hydraulic properties of a crusted soil on overland flow modelling at the field scale. Hydrological Processes, 20 : 1701-1722.

Charlier JB, Cattan P, Moussa R, Voltz M. 2008. Hydrologic behaviour and modelling of a volcanic tropical cultivated catchment. Hydrological Processes, 22 : 4355-4370.

Charlier JB, Moussa R, Cattan P, Cabidoche YM, Voltz M. 2009. Modelling runoff at the plot scale taking into account rainfall partitioning by vegetation: application to stemflow of banana (Musa spp.) plant. Hydrology and Earth System Sciences, 13, 2151-2168.

Cheviron B, Gumiere SJ, Le Bissonnais Y, Moussa R, Raclot D, 2010. Sensitivity analysis of distributed erosion models: Framework. Water Resources Research, vol. 46, W08508, 13 p.

Gomez-Delgado F, Roupsard O, Le Maire G, Taugourdeau S, Bonnefond JM, Perez A, van Oijen M, Vaast P, Rapidel B, Voltz M, Imbach P, Harmand JM, Moussa R. 2011. Modelling the hydrological behaviour of a coffee agroforestry basin in Costa Rica. Hydrology and Earth System Sciences, 15, 369–392.

Gumiere S, Raclot D, Cheviron B, Davy G, Louchart X, Fabre JC, Moussa R, Le Bissonnais Y, 2011. MHYDAS-Erosion a distributed single-storm water erosion model for agricultural catchment. Hydrological Processes,  in Press .

Lagacherie P, Rabotin M, Colin F, Moussa R,  Voltz M, 2010. Geo-MHYDAS: A  discretization procedure of Cultivated Landscapes for distributed hydrological modelling. Computers & Geosciences, 36 (2010) 1021–1032.

Moussa R. 2008a. Effect of channel network topology, basin segmentation and rainfall spatial distribution on the GIUH transfer function. Hydrological Processes, 22 : 395-419

Moussa R. 2008b. What controls the width function shape, and can it be used for channel network comparison and regionalization?. Water Resources Research, 44, 20 p., W08456.

Moussa R. 2010. When monstrosity can be beautiful while normality can be ugly: assessing the performance of event-based flood models. Hydrological Sciences Journal, 55(6), 1074 – 1084.

Moussa R, Chahinian N. 2009. Comparison of different multi-objective calibration criteria using a conceptual rainfall-runoff model of flood events. Hydrology and Earth System Sciences, 13, 519-535.

Moussa R, Voltz M, Andrieux P. 2002. Effects of the spatial organization of agricultural management on the hydrological behaviour of a farmed catchment during flood events. Hydrological Processes 16 : 393-412 (DOI: 10.1002/hyp.333).

Moussa R, Chahinian N, Bocquillon, C. 2007. Distributed hydrological modelling of a Mediterranean mountainous catchment - model construction and multi-site validation. Journal of Hydrology 337: 35-51.

Moussa R, Colin F, Rabotin M. 2011. Invariant morphometric properties of headwater subcatchments. Water Resources Research, in Press.

 Bibliography of others will follow. Anyone of them is a champion in publishing. So there is lot to read.

Thursday, October 6, 2011

I really loved his computers which I used since the beginning of my reasearch life

since 89, I guess. I had very little trouble with them, and could dedicate the time that others were spending to fix their motherboards, cards, software and hardware to do hydrology, or just enjoy life.

He is narating here:

And talking:

What he accomplished, in the word of Anton Ego:

Yes, I think I do. After reading a lot of overheated puffery about your new cook, you know what I'm craving? A little perspective. That's it. I'd like some fresh, clear, well seasoned perspective. Can you suggest a good wine to go with that?


Sunday, October 2, 2011

Presentation about landslides triggering given at IWL2

I was trying to convey the idea that landslide triggering is tricky and complex. But simple settings have a simple behavior, especially when we look at statistics. Nevertheless complexity is behind the curtain. The right one, I mean, that depends on vegetation distribution, soils use, heterogenous soil depth, and the fact that landslides are a very local phenomenon.

Here you will find the presentation. Hopefully a paper will come out from it.