New and Improved African Flood and Drought Monitoring and Early Warning Systems
African Flood and Drought Monitors website
The African Flood and Drought Monitor and its country level monitors, have been undergoing an array of improvements that combine to make them even more useful than ever. These improvements implemented in the development of two new national flood and drought monitoring and early warning systems for Zambia and Botswana. These improvements are focused on the web interface, its data/information and functionality, and have also been implemented in upgraded versions of the existing systems for Malawi, Mozambique, Namibia, South Africa, and Zimbabwe.
Overall, the FDMs have been improved in several ways, focused on adding new variables and flood and drought indices, and providing new functionality to allow the user to better identify and track flood and drought conditions, and download data. Details are below.
Data Download
Make data downloadable from the ‘Current conditions’ tab after selecting the river reach, station,
The system allows the user to download data as maps or time series from each of the tabs. Map data download is available for the currently displayed variables and selected date. Download format is GeoTiff which can be further displayed and analysed in GIS software. For times series data, the user can select a map region (e.g. district), river point or reach to display a time series of data for the selected variable. Time series data are downloaded as a “csv” format file which can be loaded into other software (e.g. R, Excel) for further display and analysis. Screenshots of maps, timeseries and tables can be used to capture images of the data for use in reports, presentations or other document types.
Current Conditions tab
Add a map to the ‘Current conditions’ tab to visualize the most critical conditions in the short-term forecast:
A “warnings” switch has been added to the “Current Conditions” tab which switches the view of the map data to show warning categories, generally based on high or low percentile values of the current variable. The categories represent conditions that are expected infrequently and are categorized as “moderate”, “severe” or “extreme”. For example, agricultural drought is represented by soil moisture values that are below the 20th (moderate), 10th (severe) or 5th (extreme) percentiles of the long-term climatology.
A separate “Current Warnings” tab has also been added which provides specific warning information, again based on categorized values of relevant variables, such as: high precipitation, pluvial flooding, fluvial flooding, and meteorological, agricultural and hydrological drought. These categories represent the maximum (most extreme) category over the forecast period and give a spatial view of where hazard conditions may occur. The evolution of the hazard (via the warning category) can also be visualized as a time-series for spatial areas (e.g. districts) or points/river reaches.
Append Short-Term Forecasts to the Historic Time Series
Append the short-term forecasts to the historic time series in the ‘Current conditions’ tab, which can be updated by clicking on the country, district/province, station or river reach:
The short-term forecast is now appended to the historic data in the time series chart shown in the “Current Conditions” tab to provide a seamless view of the evolution of the selected variable. The chart shows the evolution of the currently selected variable for the historic period (~2 years) and the future 9 days of the forecast. The historic data are shown as a single estimate; the forecast is shown as the ensemble mean, 25-75th and 5-95th percentile ranges to represent the most likely forecast (mean) and its uncertainty (percentile ranges). Downloaded data include the historic and forecast data.
Enlarge the Current Conditions Tab
Make the time series in the ‘Current conditions’ tab bigger by either increasing the size of the time series or adding a button to expand it to full screen.
A button has been added to the “Current Conditions” and “Seasonal Forecasts” tabs which expands the small time series chart to a larger window-width chart at the bottom of the map. The button toggles between the small and large charts. The larger chart may show more information (e.g. background climatology) depending on the variable selected, and the user can zoom in on different sub-time periods, and aggregate the view to weekly or monthly.
Change the Short-Term Forecasts tab to a Summary tab
Change the ‘Short-term forecasts’ tab to ‘Summary’ to reflect that this tab provides summary information for the country, districts/provinces, stations etc.
This tab is now the “Summary” tab and only shows tabular information for the districts or selected river gauging points. The district table shows a set of metrics averaged or accumulated over each district: historic rainfall for the past 10 and 90 days; the forecast ensemble mean rainfall for the next 9-days; the current area in soil moisture (agricultural) drought and the 9-day forecast area in drought. The river points table shows the current streamflow for each river gauging point and its equivalent percentile, as well as the 9-day forecast, again represented by the ensemble mean.
Add a map to the Seasonal Forecasts Tab
Add a map to the ‘Seasonal forecasts’ to visualize the most critical conditions in the seasonal forecast:
An updated “Seasonal Forecasts” tab now shows a wider range of variables and derived indices based on the 6-month precipitation forecasts, such as SPI-1, SPI-3, probability of precipitation below the median or lower tercile, probability of moderate, severe or extreme drought or pluvial conditions. The ensemble spread is provided for the variables, and the probability-based indices use the ensemble to estimate the probability of reaching climatological thresholds (e.g. pluvial/drought conditions, seasonal median or lower/upper tercile). The data and information are shown as maps and the user can select different dates in the forecast period. The data averaged over districts can be shown as time series. Both map and time series data can be downloaded.
User Guide
A user guide/tutorial is now available, which explains in detail the updated versions of the Flood and Drought Monitor platforms.
Summary
Flood and drought monitoring systems have been developed for Zambia and Botswana to complement existing systems in southern Africa for Malawi, Mozambique, Namibia, South Africa, and Zimbabwe. The existing systems have been updated to reflect the new web interface and warning information that have been developed for the Zambia and Botswana systems. The systems provide information on near real-time climate and hydrological conditions, and forecasts on short-term and seasonal scales. This is based on hydrological modelling that is forced by historic and near real-time climate model reanalysis and analysis data, and bias-corrected forecasts from ensemble weather and climate forecast systems. Evaluation of the system data (as described in deliverable 1) shows encouraging performance, with areas for potential future improvement, especially if there is access to available ground data. The new interface has been developed to provide improved access to the system data and information, new functionality to help users identify evolving flood and drought conditions, and targeted warning information (described in deliverable 2). The new interface is described in detail in the user guide. Overall, the new systems and the improved interfaces and data will provide better monitoring and early warning information for the region.