Monthly rainfall hind-cast using machine learning algorithms for Coimbatore, Tamil Nadu

Due to current world climate change, the accuracy of predicting rainfall is critical. This paper presents an approach using four different machine learning algorithms, viz., Decision Tree Regression (DTR), Gradient Boosting (GB), Ada Boost (AB) and Random Forest Regression (RFR) techniques to improve the rainfall forecast performance. When historical events are entered into the model and get validated to realise how well the output suits the known results referred as Hind-cast. Historical monthly weather parameters over a period of 42 years (1976 to 2017) were collected from Agro Climate Research Centre, Tamil Nadu Agricultural University. The global climate driver’s viz., Southern Oscillation Index and Indian Ocean Dipole indices were retrieved from Bureau of Meteorology, Australia. K- means algorithm was employed for centroid identification (which select the rows with unique distinguished features) at 90 per cent of the original data for the period of 42 years by eliminating the redundancy nature of the datawhich were used as training set. The result indicated the supremacy and notable strength of RFR over the other algorithms in terms of performance with 89.2 per cent. The Co-efficient of Determination (R2) for the predicted and observed values was found to be 0.8 for the monthly rainfall from 2015 to 2017.

THE CLUSTER ANALYSIS APPROACH FOR CLASSIFICATION OF ANDHRA PRADESH ON THE BASIS OF RAINFALL

The ave rage linkllBe method of cluster analys is was appl ied (or classifying the di stricts of AndhraPrad esh on th e b asis of monthly rainfall recorded. in diff ere nt sea so ns . The meth od of cluste ring h as the ad va nta geof least subjectivity in clu ster formati on unlike th at of the pri nci pa l components method. Th e a nalysis was ca rriedou t seasonwise on th e basis of 30 )~ ars of monthly rain fa ll data covering th e yea rs 1961-62 to 1~91. [I was foundthat the di stri cts ofAndh ra Prad esh ca n be classified iuto 5 10 7 d usters which depend on the seaso n. Ea st and WeslGod avari districts of coastal Andh ra region exh ib ited a similar pat tern in the rainfall of all the seasons: cert ain d istricts c f Telangan a region also exh ibi ted a simila r pan crn in rai nfall of all th e seasons excepting the 5Ou th~SI mensoon.In ce rtai n clu eters . there was 8 representetien u t'Ji.oJlril,,·u from all th e three ~ii ons of the Stale, Seesonwiseclust erings a re also di scu ssed .

Behaviour of rainfall over lakes around Bombay

Analysis of monthly rainfall brings out two distinctly different areas of rainfall affinity-one comprising of the lakes situated in Greater Bombay and the other in Thane districts. Rainfall of different sub-periods/months was found to be independent of the rainfall of preceding sub-periods/months. Multiple regression equations between lake levels and monthly rainfall have been computed to predict the anticipated lake levels at the end of different months.

Variability and probabilistic estimates of rainfall extremes In Bangladesh during the southwest monsoon season

The present study describes the temporal and spatial distributions of mean monthly rainfall and its variability together with the spatial distributions of the probabilistic estimates of rainfall extremes over Bangladesh during the- southwest monsoon season. The- probabilistic rainfall extremes have been computed for IWO lime scales: (a) in I year out of 4 years, and (b) in 1 year out of 10 years -representing relatively less extreme events and extreme events respectively. The mean monthly rainfall increases from June to July at most places over Bangladesh and then decreases up to September. The variability of rainfall decreases with increasing rainfall up to July at many places and then increases up to September. The study also reveals that the mean rainfall and the- probabilistic rainfall extremes are maximum over the southern and north-eastern parts of the country where the variability of rainfall is low and the rainfall is reliable. There exists a belt of low rainfall over the- central part of Bangladesh roughly between 23oN and 24°N. The rainfall gradients are maximum over north-eastern Bangladesh and the gradients of the probabilistic high rainfall are more than those of the probabilistic low rainfall in this area.

An Assessment of Water Budget of Ibadan, Nigeria

The influence of climatic conditions of precipitation and evapotranspiration exercise great control on soil water budget. This is fundamental to crop production and hydrological processes. This study assessed the temporal variability of soil moisture condition of Ibadan, Nigeria using the water budget approach. Specifically the study analyzed the climatic variables of monthly rainfall and means monthly air temperature, computed the mean monthly evapotranspiration values, plots the water budget graph, and discussed the implications of the observed seasonal trend in water budget condition on agricultural activities and hydrological processes. Monthly rainfall and mean monthly air temperature data used were collected from the archives of the Nigeria meteorological agency for the period 2008-2020. Monthly potential evapotranspiration data used in the study was estimated from the mean monthly air temperature data. The monthly rainfall data and the monthly evapotranspiration data were used to plot the water budget graph. Results revealed temporal variability in soil moisture condition. Water deficit condition was observed between November and April while water surplus condition was observed between May and October. The highest water surplus condition was observed in September (111.9mm) while the highest deficit condition (-125.64mm) was observed in December. The month of October recorded the lowest water surplus condition (41.30mm) while the month of April recorded the lowest water deficit condition (-10.10mm). The implications of the observed seasonal variation in soil moisture status on agricultural activities and hydrological processes were discussed.

Artificial Neural Network for Markov Chaining of Rainfall Over India

Rainfall forecasting plays a significant role in water management for agriculture in a country like India where the economy depends heavily upon agriculture. In this paper, a feed forward artificial neural network (ANN) and a multiple linear regression model has been utilized for lagged time series data of monthly rainfall. The data for 23 years from 1990 to 2012 over Indian region has been used in this study. Convincing values of root mean squared error between actual monthly rainfall and that predicted by ANN has been found. It has been found that during monsoon months, rainfall of every n+3rd month can be predicted using last three months' (n, n+1, n+2) rainfall data with an excellent correlation coefficient that is more than 0.9 between actual and predicted rainfall. The probabilities of dry seasonal month, wet seasonal month for monsoon and non-monsoon months have been found.

Some hydrometeorological studies over Teesta basin in north Bengal

Based on data of 40 rainfall stations located within and in the neighbourhood of Teesta basin in north Bengal for period ranging between 7 & 23 years, hydrometeorological informations of the spatial distribution of monthly rainfall, umber of rainy days and extreme rainfall distribution over Teesta basin have been determined and presented on basin maps for the months of May to October. The average monthly areal precipitation depth as wi1l as extreme areal precipitation depth for a day have been discussed for 6 sectors of the basin. The pentads rainfall for 22 selected stations in the catchment during May to October have also been evaluated and discussed.

Koreksi Bias Data Hujan Luaran GCM ECHAM5 Untuk Prediksi Curah Hujan Bulanan dan Musiman Pulau Lombok

This study aims to evaluate the ability of the ECHAM5 GCM model output data in estimating monthly rainfall on the island of Lombok. The data used in this study are ECHAM5 monthly rainfall data and automatic rainfall recorder (ARR) measurement rain data for 2000-2018 obtained from ARR Gunung Sari. Correction of bias is conducted by using the mean ratio method and the regression method. The method that produces the best approach is then used to obtain rain data projections and a simple regression method. Evaluation and validation used the Pearson correlation coefficient (r), Root Mean Square Error (RMSE) and Nash-Sutcliffe Efficiency (NSE) values. The results obtained are that the daily and monthly rainfall data from the ECHAM5 model cannot be directly used to replace the rain measurement data because of its very low accuracy. The downscaling technique performed on daily and monthly rainfall data using the average ratio method does not show satisfactory performance where the efficiency figures produced are still low even gave a slight increasing number. However, the ECHAM5 model data can be used to obtain rainfall projections on a monthly and seasonal scale with a good and satisfactory correlation. Key words: mean ratio method; global climate model; ECHAM5; monthly rainfall.