Climate Modelling and Prediction

Our Projects

Advocacy

News

Valid HTML 4.01 Transitional

Valid CSS!

Computational models of climate help understand the mechanisms at work and enable predictions. Our initial focus was on high resolution local area models that capture salient features of Sri Lanka’s climate. We developed a simple model for capturing the topographically induced rainfall. Separately, we developed a high-resolution wind climate model while ignoring the rainfall mechanisms. Thereafter we also partnered with Joshua Qian at the IRI in his work on a Regional Climate Model for Sri Lanka.

We have investigated the skill of predictions over Sri Lanka by the IRI’s seasonal forecast system. We collaborated to implement a statistical methodology to correct the predictions of global models of rainfall over Sri Lanka at high resolution.

Activity
Objectives
 Partners Status Next Steps
Mesoscale Wind Modelling

Generate 2-D wind climatologies

Iowa State University
Investigate causes of
model error in some
months. Draft papers
available.
Develop 2-D wind
resource assessment.
Refine papers for
publications.
     
Orographic Rainfall
Modelling
Extend 2-D orographic rainfall model to generate slices of prediction.
IRI

Obtain upper air data sets for work, obtain East-West slices of Sri Lankan topography.
Paper on orographic
rainfall over Sri
Lanka. Assess effects
of pollution on
orographic rainfall.
     
Regional Climate Modelling focused on Sri Lanka
Develop new methodologies for high-resolution climate prediction
IRI (Joshua Qian), Int. Center for Theoretical Physics, Met. Dept.
High Resolution climate model developed for Sri Lanka and shown to be essential for capturing regional variations.
Further development of nesting algorithms, model runs for extra years, implementation of model in Sri Lanka publications.
Statistical Down-scaling
Generate statistical predictions for all seasons for Sri Lanka
IRI October to December
season predictions
completed. Paper drafted.		Develop predictions for all seasons. Publish OND paper.

 

/files/fect//windmodel.png

Orographic Rainfall Modeling

Orographic rainfall is brought about by the uplift of moist air by mountains leading to precipitation. Orographic rainfall modelling is useful because, it is a means of understanding the physical mechanisms of precipitation; it provides a basis for interpolating and extrapolating rainfall observations; and it provides a simple downscaling mechanism between large scale atmospheric features such as circulation and moisture budgets and local rainfall. These large scale features may be obtained from radiosonde measurements, reanalyses data or GCM’s.
Orographic rainfall is particularly important for water resource studies in tropical regions as many rivers obtain a large part of their rainfall in the mountain regions. Sri Lanka is under the semi-annual monsoonal changes in wind directions. The mean wind direction from May to October is from the West and from December to March is from the North-East. The strongest westerly winds are in July and the strongest North-Easterly winds are during January. While the wind intensity in January is half of that during July, it is still critical in inducing rainfall in an otherwise dry part of Sri Lanka. Since orographic rainfall is caused by such intense lateral winds, we present an analysis of rainfall for July.

/files/fect//ortho.png

Orographic Rainfall Model for July

To model mountain induced rainfall, we first computed the vertical velocity across a mountain West-East cross-section; then estimated the moisture across the section and then established the extent of condensation and precipitation. Based on assumptions of distribution of rain droplets, we estimate where the rain shall fall.

This model may be tested against average rainfall along stations close to the cross-section when the westerlies are strongest. We have obtained the average monthly July rainfall for stations in the East-West cross-section going from Colombo to Pottuvil. The actual rainfall values and the model predictions are shown below.

 

/files/fect//otho2.png

 

/files/fect//ortho3.png

 

Regional Climate Modelling

Sri Lanka climate is downscaled from a global atmospheric general circulation model ECHAM4.5 using a regional climate model RegCM3.

/files/fect//regcm.png
The RegCM3 was run over South Asia with 100km and 20km grid size, with its domain and the topography shown above. Because of the strong orographic effect on the rainfall in Sri Lanka, fine resolution in the regional model is needed. Resolving the topography at 100km-grid resolution results only in a highest elevation of only 200m and at resolution of 20km-grid a peak of 1000m is obtained. The highest elevation in Sri Lanka is 2500m.
/files/fect//regcmresult.png
Summary: The precipitation and circulation produced by GCM’s is often unrealistic. Using a regional model, one is able capture the gross regional variations in precipitation and circulation. This work shows that a regional climate model of at least 20-km horizontal grid is needed to capture the regional variations of rainfall in a region with topography as in Sri Lanka.

Seasonal Climate Predictions - IRI

The International Research Institute for Climate Prediction (now IRI for Climate and Society) has been issuing global climate forecasts at quarterly intervals in an experimental mode since October 1997. Forecasts for rainfall and temperature are issued for the subsequent 3 months and 6 months. Forecasts are issued for rainfall and temperature in the categories of above normal, near-normal and below-normal. The cut-offs for each category are based on the wettest, normal and driest 10 episodes for the given season from 1960 to 1990.

These forecasts are based on global climate simulations by four Global Climate Models which are initialized based on observations of global ocean surface temperatures. These simulations are at a resolution of approximately 250 km. And thus Sri Lanka is captured in two grid boxes.

Seasonal climate forecasting is a new field and the skill obtained thus far is good for regions such as Indonesia, North-East Brazil, and East Africa. In general, the forecast skill for Asia and Europe is weaker than that for other continents. The regions with the most skill within Asia are South-East Asia and South Asia. However, the skill is likely to improve in the coming years.
/files/fect//predict01.png
Both modellers and potential users are concerned regarding the quality of forecasts - evaluating forecast quality is a simpler task with deterministic forecasts than probabilistic forecasts. Given that some weightage is afforded for all eventualities, essentially there can be no wrong probabilistic forecasts. However, if the forecasting system consistently affords higher weightage to categories that do not occur, then the forecasting system is less reliable. The reliability of the forecasting system can only be evaluated once there are a large number of forecasts and outcomes. However, the IRI forecasting system has been in place only for a limited time. Hence to provide a simple account of its reliability one is compelled to adopt a crude alternative - one based on replacing the probabilistic forecast with a deterministic forecast for its dominant category. Based on this approach, the track record for the IRI for Sri Lanka of the 3 month rainfall forecasts for the 12 forecasts that has been issued since 1997 is eight hits (observation coincides with dominant category of forecast), three 1-category misses (observation in neighbouring category from dominant category for temperature) and no 2-category misses.
/files/fect//predict02.png
/files/fect//hitmiss.png

Statistical Downscaling of Predictions

The large-scale atmospheric circulation across the Indian Ocean sector has a strong degree of predictability in the October-December season. This provides a basis for investigating predictability at spatial scale. We ask the question: for a given large-scale wind forecast across the region, what are the details of the rainfall pattern to expect across Sri Lanka? Two ways to answer this question are

Run a high resolution climate model driven with the large-scale wind fields from the Global Climate Model

Establish the statistical relationship between the details of the observed rainfall pattern and the large-scale wind forecast – using analysis over a large set of past years – and use these relationships to forecast each location, given a large-scale wind forecast.
Using the statistical approach, we find that there is good skill on the eastern side. This skill is physically based – since wetter years appear to be associated with SST that enhances the easterly surface wind component across the region – which will impinge on the eastern side of the island first and give rise to particularly enhanced precipitation on eastward facing slopes.

 

/files/fect//downlonscale.png
Map: The skill of the downscaling scheme that was shown is represented as a correlation score at right. The correlation skill ranges up to 0.6. The topographic contours are shown as well.

 

/files/fect//examplescale.png

Conference proceedings




Reports