Tutorial Basic Users
(→Anwendung des JAMS Data Explorers (JADE) für die Ergebnisauswertung) |
|||
| Line 281: | Line 281: | ||
[[Bild:toShape.jpg]] | [[Bild:toShape.jpg]] | ||
| − | == | + | ==Application of the JAMS Data Explorer (JADE) for Result Analysis== |
| + | |||
| + | By using the JAMS Data Explorer (JADE), on the one hand, input data which were the basis for the modeling can be displayed. On the other hand, the JADE tool can evaluate and visualize calculated output data. | ||
| + | The JADE can be launched under the menu item Modelle/JAMS Data Explorer. The input data which were the basis for the modeling are automatically loaded (Input Data). Moreover, the output folder generated during the modeling will be loaded (Output-Data). In this way, all modeled data can be loaded, displayed and analysed. | ||
| − | |||
| − | |||
[[Bild:Modell_laden.jpg]] | [[Bild:Modell_laden.jpg]] | ||
| Line 365: | Line 366: | ||
[[Bild:Beispiel2.jpg]] | [[Bild:Beispiel2.jpg]] | ||
| − | |||
| − | |||
==Overview of Implemented Models== | ==Overview of Implemented Models== | ||
Revision as of 12:02, 2 February 2011
System Requirements, Download and Installation of JAMS/J2000
System Requirements
To run JAMS the Java Runtime Environment (JRE) Version 5.0 or higher is required. The installation file can be downloaded here: http://java.sun.com/javase/downloads/index.jsp.
JAMS/J2000 Download
JAMS is available to download at www.geoinf.uni-jena.de/5580.0.html as a package including the hydrological model J2000 and a test data set. To run JAMS an installation of Java (J2SE JRE, Version 1.5 or higher) is required.
In addition, all Java sources of JAMS, different standard components as well as all libraries needed are available on this website.
There are packages for Windows and Linux at disposal. If your operating system does not have a Java installation, JAMS/J2000 can be downloaded with a Java Virtual Machine.
The installation for Windows is carried out by an executable file which installs JAMS completely. For Linux a tgz-archive is available which includes all files needed. For the execution of JAMS on Linux an existing Java installation is required.
The installation of JAMS includes two executable files:
- jams.exe (jams on Linux) launches the JAMS-Launcher
- juice.exe (juice on Linux) launches the graphical model editor JUICE
JAMS/J2000 Installation on Windows
The installation program opens.
Click on next and accept the license agreement.
In the next step you can decide whether you would like to install two test data sets in addition to JAMS/J2000. In each of those data sets you can find an example of a model description and all necessary input data for the catchment area.
Please select the folder where the programm will be installed.
Please select a folder of the start menu to create shortcuts for JAMS Builder, JAMS Launcher, JAMS Remonte Launcher and test data sets.
Click on install to start the installation.
JAMS/J2000 has been successfully installed.
Introduction and Application of the JAMS Launcher
The JAMS Launcher is a graphical surface for editing model configurations and system configurations and a device for model execution. The design is dependent on the model configuration which is currently loaded. Automatically generated input components can serve to define initialization values for specific model parameters.
The JAMS launcher also enables loading and saving model and system configurations. It is automatically displayed when starting JAMS, providing that the feature "guiconfig" has the value "1" in the system configuration.
Structure and Capacity of the JAMS Launcher
The JAMS Launcher is a software tool which facilitates loading and parameterizing models as well as carrying out modelings. Additionally, modeling results can be visualized. The total runoff at the catchment outlet, soil water, snow water equivalent and the map of the catchment area can be examined.
You can start the launcher from the start menu or from a shortcut on your desktop. The following window will appear: Bild:JAMS.jpg
The JAMS Launcher has been opened and can now start the modeling.
Under the menu item Datei you can load, save and close models.
The menu item Extras can change settings, load and save settings or continue an interrupted model execution.
Under JAMS settings you can change libraries, output size, output of model protocols and information on model windows. Those settings can be saved for later modelings under Einstellungen speichern and can be loaded under Einstellungen laden.
Under the menu item Modell, you can start the modeling (Modell starten), you can see the current workspace (Workspace anzeigen) or you can see the docket of all ongoing processes (Prozessliste öffnen).
Under the menu item JAMS Data Explorer you can export the modeling results as well as the input data which are at disposal for the model into the tool JADE. This allows an intensive evaluation of results and an analysis of the measured data used for the modeling.
The menu items Lade Modellparameter and Speichere Modellparameter are used to manage model parameters. An edited parameter set can be saved to reuse it for new modelings. Bild:Systemleiste_Modell.jpg
Under the menu item Protokolle you can look at the Infoprotokoll (information protocol). It contains information on the model's author, its version or on its user as well as information on the model's efficiency (How well did the model display the measured data?), the model structure (Which components were used and how often?) and on the model runtime.
The Fehlerprotokoll (error log) records any errors which may have occurred.
Bild:Systemleiste_Protokolle.jpg
Under the menu item Hilfe (help) you can obtain online help of JAMS by clicking on JAMS online.
Selecting and Running a Model
If you want to run a model, select a model description in the JAMS Launcher under the item Datei/Modell laden (load file/model). Models can be read as *.xml or *.jam files. If you named the installation folder jams and installed a test data set when installing JAMS, then you will find an example of a model description (j2k_gehlberg.jam) in folder jams/data.
You can start the modeling by clicking on the green button in the menu bar.
Editing Model Parameters
Basic Settings
- Workspace directory: Sets the working directory. It has to contain three more folders: Parameter (for all parameter files), Data (for all data files) and Output (for all output files).
- Time interval: The time interval for the model run is selected.
- Caching: The results of some compute-intensive processes can be temporarily stored in hard disk and reused for further model runs. Therefore the model run is slightly faster. Attention: This feature is not completely safe yet and should only be applied by experienced users.
Diagrams and Maps
- Runoff plot: Activates the graphical display of the runoff modeled and measured during model run.
- Soil moisture plot: Activates the graphical display of the relative soil moisture during model run.
- Snow water equivalent: Activates the graphical display of the snow water equivalent during model run.
- Map enable: Enables the output of a cartographic display of selected state variables.
- Map attributes: A semicolon-separated list of state variables which are to be cartographically displayed.
- Map3D enable: Enables a 3D output of a cartographic display of selected state variables.
- Map3D attributes: A semicolon-separated list of state variables which are to be cartographically displayed (in 3D).
Initialising
- Multiplier for field capacity : The maximum storage capacity of the middle pore storages (MPS) can be increased (value > 1) or decreased (value < 1).
- Multiplier for air capacity: The maximum storage capacity of the large pore storages (LPS) can be increased (value > 1) or decreased (value < 1).
- initRG1: relative filling of the upper groundwater storage at beginning of model run (1 filled to capacity, 0 empty).
- initRG2: relative filling of the lower groundwater storage at beginning of model run (1 filled to capacity, 0 empty).
Regionalization
- number of closest stations for regionalization: Number n of stations used to calculate the value of an HRU (n stations which are closest to the HRU are selected).
- Power of IDW function for regionalization: Weighting factor used to exponentiate the distance of each station to the respective HRU.
- elevation correction on/off: Activates the elevation correction of the data values.
- r-sqrt threshold for elevation correction: Threshold value for the elevation correction of the data values. If the coefficient of determination of the regression relation between measured data of the stations and station elevations is smaller than this value, an elevation correction is not carried out.
Those settings (i.e. minimum temperature, maximum temperature, medium air temperature, precipitation, absolute air moisture, wind speed, sunshine duration) can be adjusted for every single input variable.
Radiation
- flowRouteTA [h]: runtime of the outflow route
Interception
- a_rain [mm]: Maximum storage capacity of the interception storage per m2 leaf area for rain
- a_snow [mm]: Maximum storage capacity of the interception storage per m2 leaf area for snow
Snow
- Component active: Activates the snow module.
- baseTemp [°C]: Temperature limit value for snow precipitation.
- t_factor [mm/°C]: Temperature factor for calculation of snowmelt runoff.
- r_factor [mm/°C]: Rain factor for calculation for calculation of snowmelt runoff.
- g_factor [mm]: Soil heat flux factor for calculation of snowmelt runoff.
- snowCritDens [g/cm³]: critical snow density
- ccf_factor [-]: factor for calculation of the cold content of snow cover
Soilwater
- MaxDPS [mm]: maximum hollow reserve
- PolRed [-]: polynomial reduction factor for reduction of potential evaporation with limited water supply.
- LinRed [-]: linear reduction factor for reduction of potential evaporation with limited water supply.
(Note: PolRed and LinRed do not represent alternatives. Only one can be attributed a value, the other one has to be 0.)
- MaxInfSummer [mm]: maximum infiltration during summer period
- MaxInfWinter [mm]: maximum infiltration during winter period
- MaxInfSnow [mm]: maximum infiltration with snow cover
- ImpGT80 [-]: relative infiltration capacity of areas with a sealing degree of > 80%
- ImpLT80 [-]: relative infiltration capacity of areas with a sealing degree of < 80%
- DistMPSLPS [-]: calibration coefficient for distribution of infiltration on soil storages LPS and MPS
- DiffMPSLPS [-]: calibration coefficient for the definition of the diffusion amount of the LPS storage in relation to MPS at the end of a time step
- OutLPS [-]: calibration coefficient for definition of LPS outflow
- LatVertLPS [-]: calibration coefficient for distribution of the LPS outflow on the lateral (interflow) and vertical (percolation) component
- MaxPerc [mm]: maximum percolation rate
- ConcRD1 [-]: retention coefficient for direct runoff
- ConcRD2 [-]: retention coefficient for interflow
Groundwater
- RG1RG2dist [-]: calibration coefficient for distribution of percolation water
- RG1Fact [-]: factor for runoff dynamics of RG1
- RG2Fact [-]: factor for runoff dynamics of RG2
- CapRise [-]: factor for the setting of capillary rise
Routing in the Flow
- flowRouteTA [h]: runtime of the outflow route
Visualization of Modeling Results
After the modeling has been successfully carried out, the following window opens automatically:
The ExeptionInfo contains information on the model's author, its version or on its user as well as information on the model's efficiency (How well did the model display the meadured data?), the model structure (Which components were used and how often?) and on the model runtime.
The Runoff Plot can compare the simulated runoff with the runoff actually measured. The precipitation is shown in this diagram as well.
The SWE Plot contains information on how much water is stored as snow.
Under the menu item Map the catchment area is shown with its modeling units. In the right window all those variables are listed which were selected in the JAMS Launcher/Plots & Maps/Map attributes. They can be evaluated in different maps.
The value of a single modeling unit variable can be shown by clicking on Modellierungseinheit.
The map can be exported with selected attributes as *.shp.
Application of the JAMS Data Explorer (JADE) for Result Analysis
By using the JAMS Data Explorer (JADE), on the one hand, input data which were the basis for the modeling can be displayed. On the other hand, the JADE tool can evaluate and visualize calculated output data. The JADE can be launched under the menu item Modelle/JAMS Data Explorer. The input data which were the basis for the modeling are automatically loaded (Input Data). Moreover, the output folder generated during the modeling will be loaded (Output-Data). In this way, all modeled data can be loaded, displayed and analysed.
Analyse der Eingabedaten für die Modellierung
Um die Eingangsdaten, die für die Modellierung verwendet worden sind, auszuwerten, wählen Sie einen Datensatz aus dem Input-Ordner mit Doppelklick aus. Die Auswahl wird dann im rechte Feld dargestellt.
In diesem Beispiel werden die Daten zum tmeam-Datensatz (Durchschnittstemperatur) angezeigt.
Das untere linke Feld enthält folgende Informationen zum Datensatz:
- Name der Datenreihe
- Datentyp
- Startpunkt der Zeitreihe
- Endpunkt der Zeitreihen
- Datendichte (Schrittweite zwischen zwei Datansätzen)
- Defaultwert, falls keine Daten vorliegen
Im rechtern Fenster werden die ausgewählten Daten dargestellt. Im Beispiel sieht man, dass für die Modellierung Messdaten an drei verschiedenen Stationen (Artern, Erfurt, Schmücke) genutzt worden sind.
Die rechte Legende enthält die folgende Funktionalität:
- Close Tab - Datensatz schließen.
- Time Plot - Darstellung des ausgewählten Datensatzes (mindestens einer Station) in einem Diagramm. Der Zeitpunkt entspricht dabei der X-Achse.
- Data Plot - Vergleichende Darstellung der Datensätze mindestens zwei verschiedener Stationen in einem Diagramm.
- Save Data - Der ausgewählte Datensatz kann gespeichert werden.
- Statistics - Statistische Auswertung des/der ausgewählten Datensatzes/Datensätze.
Beispiel
Im folgen Beispiel wird der Datensatz Erfurt durch klicken auf das Feld Erfurt ausgewählt.
Durch einen Klick auf den Button Time Plot wird die Zeitreihe für die Messstation Erfurt im Diagramm dargestellt.
Durch einen Klick auf den Button Statistics wird die Zeitreihe für die Messstation Erfurt statistisch ausgewertet.
Analyse der Ausgabedaten nach der Modellierung
Die Daten, die bei der Modellierung erzeugt werden, könne mithilfe des JADE analysiert und visualisiert werden. Mithilfe eines Doppelklicks auf den relevanten Datensatz, z.B. HRU-Loop, wird ein Fenster geöffnet, womit die Daten, die genauer analysiert werden sollen, ausgewählt werden können. Dieses wird im rechten oberen Teil der Programmoberfläche angezeigt und im folgenden Bild dargestellt.
Im ersten Fenster, können die Attribute gewählt werden, die analysiert werden sollen. In diesem Fall sind das area (Fläche der HRU), precip (Niederschlag), tmean (mittlere Temperatur), satMPS() und satLPS(). Für den Aggregationstype (Aggregation type) ist (1) eine Summenbildung, (2) eine Durchschnittsbildung und (3) eine nach Größe der HRU gewichtete Analyse möglich. In den Feldern Time Steps bzw. Entity IDs können Zeitschritte und Elemente (HRUs) ausgewählt werden. Durch den Shortcut Strg+A können alle Zeitschritte bzw. Elemente zusammen gewählt werden.
Folgende Buttons stehen zur Verfügung:
- Time Step - Listet die gewählten Daten des markierten Zeitschrittes für alle HRUs auf.
- Temp. Aggr. - Listet die gewählten Daten der markierten Zeitschritte in aggregierter Form für alle HRUs auf.
- Spatial Entity - Listet die gewählten Daten der markierten HRU für alle Zeitschritte auf.
- Spatial Aggr. - Listet die gewählten Daten der markierten HRUs in aggregierter Form für alle Zeitschritte auf.
- Crossproduct - Listet die gewählten Daten der/(des) markierten Zeitschritte/(s) für die markierte/(n) HRU/(s) auf.
- Time Filter - Der Time-Filter bietet die Möglichkeit aggregierte Daten einzelner Monate bzw. Jahre aufzulisten.
Für den aggregierten Novemberwert des Jahres 1996, würde man 1996-11-% in das Time-Filter Feld eintragen und dann den Botton Temp. Aggr. betätigen.
In den Feldern Month bzw. Years ist es möglich einzelne Monate bzw. Jahre zu analysieren.
- Monthly Aggr. - Listet die gewählten Daten des markierten Monats in aggregierter Form für alle HRUs auf.
- Time Step - Listet die gewählten Daten des markierten Jahres in aggregierter Form für alle HRUs auf.
- Reload Indes - Setzt die Auswahl zurück.
Beispiel
In diesem Beispiel wurden die Daten für die ausgewählten Attribute Niederschlag (precip) und Tagesdurchschittstemperatur (tmean) von 1. November 1996 bis 10. November 2006 für alle Modelleinheiten (HRUs) in aggregierter Form geladen. Als Attributtype wurde der Mittelwert gewählt (Aggregation type = mean), d.h. aus den Werten (01.11.1996, 02.11.1996,...,10.11.1996) der Durchschnittswert ermittelt. Close Tab, Time Plot, Data Plot, Save Data, Statistcs und Show on map wurden bereits unter dem Punkt Analyse der Eingabedaten für die Modellierung erläutert.
Ein Klick auf den Button Data Plot erzeugt folgendes Fenster, indem die Tagesdurchschnittstemperatur dem Niederschlag gegenübergestellt wird.
Overview of Implemented Models
This tutorial is concerned with the hydrological modelling system J2000. The hydrological modelling system J2000g was developed as a simplified model and the modelling system J2000-S additionally takes into account the water and nitrogen budget. Therefore, the models can be seen as a simplification and an extension of the J2000. The modeling systems are introduced below.
J2000
The hydrological modeling system J2000 allows a physically-based modeling of the water supply of big catchment areas. Alongside the simulation of hydrological processes, which influence the runoff generation and its concentration on the upper meso and macro level, it contains routines which can most certainly be used to regionalize measured climate and precipitation data selectively available. In addition, the calculation of actual stand evaporation is directly integrated into the model which allows an area-specific calculation of the evaporation according to different land use classes. Since the model is supposed to be suitable for modeling large catchment areas with a size of several 100km2, it is ensured that the modeling can be done with data available on a national scale. The simulation of different hydrological processes is carried out in closed program modules which are to a great extent independent of each other. This allows changing, substituting or adding single modules without having to structure the model once again from the start.
J2000-S
The water and mass transfer model J2000-S allows the simulation of the water and nitrogen budget of mesoscale catchments. The model is an extension of the model J2000 and therefore shares most components to describe the hydrological cycle. The description of the nitrogen budget contains additional components, naming soil temperature, soil nitrogen budget, land use management, plant growth and groundwater nitrogen budget. Additional modules were adjusted for requirements of the nitrogen budget.
J2000g
The model J2000g was developed as a simplified hydrological model to calculate temporally aggregated hydrological target figures distributed over time. Hydrological processes are calculated and displayed one-dimensionally for a random amount of points in space. Those modeling points allow an equal implementation of different distribution concepts (Response Units, grid cells, partial catchment areas) without any further model adjustment. The temporal discretization of the modeling can take place in daily steps or monthly steps. During the modeling the following processes are calculated for each time step: regionalization of climate data to the respective model units, calculation of global and net radiation as input for the calculation of evaporation, calculation of land-cover-specific potential evaporation according to Pernmam-Monteith, snow accumulation and snowmelt, soil water budget, groundwater recharge, retardation of runoff (translation and retention).
Preparation of Input Data
Application of the Implemented Grass-HRU Toolkit
Please see GRASS-HRU.
Sensitivity Analysis and Parameter Optimization
In order to apply hydrological models successfully it is necessary to define model parameters accurately. A direct measurement of the parameters is mostly not possible, too expensive or there is no clear physical relation. For those reasons the parameters are adjusted in a trial and error process in so far that the simulated factors (e.g. runoff) correspond best to the values measured. This task can be time-consuming and difficult if the corresponding model is complex or has a large number of parameters. Therefore it makes sense to automatize the optimization process with a software tool for the optimization of model parameters and sensitivity analysis which is available. Its functionality is described in detail in the Tutorial Calibration and Sensitivity Analysis.
Exercises for the Tutorial Basic Users
Übung 1/JADE
Aufgabe
Analysieren Sie den Datensatz Wind (wind) der für die Modellierung im Einzugsgebiet der Wilden Gera (mit Pegel Gehlberg) zur Verfügung gestanden hat. Wählen Sie die Datensätze für die Messstation Artern und Erfurt und lassen Sie sich die Daten in einem Diagramm vergleichend darstellen. Vergleichen Sie Median, Maximum sowie Minimum der Zeitreihen.
Lösung
(1) Klicken Sie auf den tmean Ordner aus den Input Daten.
(2) Markieren Sie die Datensätze Artern und Erfurt. (Strg + Klick auf den Datensatz) Diese sollten jetzt BLAU hinterlegt sein.
Bild:Auswahl_Arbtern_Erfurt.jpg
(3) Durch einen Klick auf den Button Data Plot werden die Zeitreihen für die Messstationen Artern und Erfurt im Diagramm gegenübergestellt.
(4) Durch einen Klick auf den Button Statistics werden die ausgewählten Zeitreihen statistisch ausgewertet. Hierraus können die Informationen für Median, Maximum und Minimum der Zeitreihen abgelesen werden.
Bild:Statistics_Artern_Erfurt.jpg
Übung 2/ JADE
Aufgabe
Analysieren Sie den Niederschlag sowie die Tagesdurchschnittstemperatur der Modellierungseinheit 9,10,11 (Entity IDs 9,10,11) der Modellierung mit dem Modellsystem J2000 im Einzugsgebiet der Wilden Gera mit Pegel Gehlberg. Bilden Sie dazu den Durchschnittswert der drei Modelleinheiten.
Lassen Sie sich die geladenen Daten in einem Diagramm anzeigen. Passen Sie die Namen der Achsen an die dargestellten Attribute an.
Zoomen Sie in den Zeitraum 01.07.1998 bis 31.12.1998 und stellen Sie diesen im Diagramm dar.
Lösung
(1) Klicken Sie die Attribute precip und tmean an.
(2) Als Aggregationstype muss mean gewählt werden.
(3) Wählen Sie die drei Modelleinheiten 9,10,11. (Tipp: Durch drücken der Taste Strg können mehrere Modelleinheiten ausgewählt werden.)
(4) Drücken Sie den Button Spatial Aggr..
Folgendes Fenster sollte angezeigt werden:
(5) Markieren Sie die Datensätze ID,precip und tmean. (STRG + Klick auf den Datensatz) Diese sollte jetzt BLAU hinterlegt sein.
(6) Drücken Sie den Button Time Plot um die gewählten Daten darzustellen. (Als X-Achse wird automatisch die Zeitachse gewählt!)
(7) Markieren Sie den Zeitraum 01.07.1998 bis 31.12.1998 mit der Maus.
(8) Der gewünschte Zeitraum wird automatisch dargestellt.
