CM Box User GuideMain Page | About | Special pages | Log in

Printable version | Disclaimer | Privacy policy | Current revision

(Difference between revisions)

Revision as of 19:26, 19 August 2006
Peter (Talk | contribs)
(Defining the grid)
← Previous diff
Revision as of 19:27, 19 August 2006
Peter (Talk | contribs)
(Step 1. Creating a grid based network.)
Next diff →
Line 86: Line 86:
-===Step 1. Creating a grid based network.===+===Creating a grid based network.===
The creation of a grid-based net work is best explained through an example. In the example below a grid for Southern Africa will be created. The creation of a grid-based net work is best explained through an example. In the example below a grid for Southern Africa will be created.

Revision as of 19:27, 19 August 2006

Setting up the monitoring network

Before setting up a monitoring network a decision has to be made on the type of network that is appropriate for the country or region at hand. This chapter outlines the practical steps to set up such a network given the choice made in the previous chapter.

Introduction to a station-based network

Most countries have an extensive network of meteorological stations. The most important types of monitoring stations are :

  • Synoptic stations. These stations provide hourly data on a real-time basis. Usually these data are fed into the global GTS system of WMO. Most of these stations provide weather data that enable users to calculate the Penman-Monteith reference evaporation (ET0). A subset of these stations do upper-air observations, which is less interesting for agrometeorological purposes.
  • Realtime Rainfall stations. These stations monitor rainfall and send in their data to the National Meteorological Office once a day.
  • Off-line rainfall stations that send in their daily data once a month or even once a year. These stations are important for climatic studies and the provision of normal data.


Which stations are to be included in the crop monitoring network?

Although the selection criteria differs from countrry to country a few guidelines are universal:

  • The station should be in an area where the monitoring crops are located.
  • A reasonably stable stream of data on de decadal basis (or daily, hourly basis) should reach the crop forecasting unit.
  • Stations should be evenly spread in crop growing areas. It does not make sense to include five stations in a 10 by 10 kilometer area (this is often the case in a capital city)
  • To monitor crops in a water balance model, the following data are needed for the selected stations:
    • Rainfall (actual and normal)
    • PET (Potential evapotranspiration)
    • Crops (which crops are planted and where they can be found)
    • Planting dates for crops (PLD)
    • Cycle length for all the crops (CYC)
    • Water Holding Capacity (WHC) of the soil.
    • Percentage Effective Rainfall (EFR)
    • Pre-season Crop coefficient (usually around 0.15)
    • Irrigation data including bund height (when applicable).

A station-based network: step by step

The following fields for every station are needed:

  1. ID (national or international number)
  2. Name (spelling is sometimes an issue)
  3. Longitude
  4. Latitude
  5. Altitude
  6. Province (optional)
  7. District (optional)

Step 1. Entering administrative areas

Most countries are subdivided in provinces and districts although the names for the administrative units may differ. Start AgrometShell. Open the menu item “Database – Manage Stations”. Press the “Administrative Levels” – button at the top of the screen. Screen 1 will appear. The top part enables the entry of provinces (administrative level 1), the bottom part is for the entry of administrative level 2 (districts) corresponding with the province.

The country codes used can be found here: Country codes


Screen 1: Entering administrative units


In this step we will start setting up the “rainfall monitoring network” list. Perform the following steps:

Step 2. Entering station data

In Agrometshell activate the “Database – Manage Station Lists” function, select <Create new list> and specify the name for the new list E.G. “crop monitoring network”. Press OK. A window will appear in which to enter the names of the new stations and all the attributes. In the first column the code of the station is specified. The second column specifies the station name. The third column specifies the latitude, the fourth the longitude, the fifth the altitude. The next three columns enable the user to specify the country and the administrative levels from “pick lists”. After entering all the stations the screen should look like this in screen 2.


Screen 2: Entering station attributes


Step 3. Viewing a station map

Activate the Database-Map function. Specify "Station Name" at the second option and press <Ok>. The map station map appears (screen 3).


Screen 3: Viewing the station map

Introduction to a grid-based network

Although a station-based network is the preferred type of network, a grid-based network is often the only possibility when:

  • A reliable source of ground data is not available.
  • Ground data do not arrive in time for crop forecasting
  • Ground data are sparse or geographically unevenly spread.

Luckily, through the years the quality of satellite derived data has improved dramatically and good data is available at low or no cost at all. Organizations ECMWF (European Centre for Medium-Range Weather Forecasts), EU/JRC, FAO provide good satellite data for use by national early warning units.

In a grid-based network, the water balance calculation is done with “fake” stations. These stations are laid out in a regular network. Agrometshell fully supports the use of a grid based network. The next paragraph will outline this.


Creating a grid based network.

The creation of a grid-based net work is best explained through an example. In the example below a grid for Southern Africa will be created.


Activate the Tools-Create Station List from Grid function. The first step is to specify the map that defines the boundaries of the study area (usually a country or a sub-region). In this example, a grid for Southern Africa is created. All Southern Africa countries can be found in the “world.bna” boundary file that can be found in the AgrometShell installation package.
In this window the grid is defined. At first select the polygons in the top box and select them with the down-arrow. Then specify the grid density. In this case 0.,5 degrees is chosen. Press <Redraw> to see the result. This means that a square grid will be created where stations are half a degree apart. Please note that the program predicts the creation of 1343 stations. Specify a list name to be generated and prefixes for the station names and ID’s in order to clearly distinguish the stations in the list.
After pressing the <Generate>, the list is filled and finally shown in the AgrometShell list editor.



Page generated in 0.217291 seconds.