Calibration of an oscillating nozzle‐type rainfall simulator

Nozzle‐type rainfall simulators are commonly used in hydrologic and soil erosion research. Simulated rainfall intensity, originating from the nozzle, increases as the distance between the point of measurement and the source is decreased. Hence, rainfall measured using rain gauges would systematically overestimate the rainfall received at the ground level. A simple model was developed to adjust rainfall measured anywhere under the simulator to plot‐wide average rainfall at the ground level. Nozzle height, plot width, gauge diameter and height, and gauge location are required to compute this adjustment factor. Results from 15 runs at different rain intensities and durations, and with different rain gauge layouts, showed that a simple average of measured rain would overestimate the plot‐wide rain by about 20 per cent. Using the adjustment factor to convert measured rainfall for individual gauges before averaging improved the estimate of plot‐wide rainfall considerably. For the 15 runs considered, overall discrepancy between actual and measured rain is reduced to less than 1 per cent with a standard error of 0·97 mm. This model can be easily tested in the ?eld by comparing rainfall depths of different sized gauges. With the adjustment factor they should all give very similar values. Copyright © 2003 John Wiley & Sons, Ltd.     

Genetic programming approach for flood routing in natural channels

In recognition of the non‐linear relationship between storage and discharge existing in most river systems, non‐linear forms of the Muskingum model have been proposed, together with methods to calibrate the model parameters. However, most studies have focused only on routing a typical hypothetical flood hydrograph characterized by a single peak. In this study, we demonstrate that the storage–discharge relationship adopted for the non‐linear Muskingum model is not adequate for routing flood hydrographs in natural channels, which are often characterized by multiple peaks. As an alternative, an evolutionary algorithm‐based modelling approach, i.e. genetic programming (GP), is proposed, which is found to route complex flood hydrographs accurately. The proposed method is applied for constructing a routing model for a channel reach along the Walla Walla River, USA. The GP model performs extremely well with a root‐mean‐square error (RMSE) of 0·73 m³ s^?1 as against an RMSE of 3·26 m³ s^?1 for routing the multi‐peaked hydrograph. The advantage of GP lies in the fact that, unlike other models, it establishes the routing relationship in an easy and simple mathematical form. Copyright © 2007 John Wiley & Sons, Ltd.     

Long‐term change of daily and multi‐daily precipitation in southern Sweden

Daily rain series from southern Sweden with records dating back to the 1870s have been analysed to investigate the trends of daily and multi‐day precipitation of different return periods with emphasis on the extremes. Probabilities of extreme storms were determined as continuously changing values based on 25 years of data. An extra set of data was used to investigate changes in Skåne, the southernmost peninsula of Sweden. Another 30‐year data set of more than 200 stations of a dense gauge network in Skåne was used to investigate the relation between very large daily rainfall and annual precipitation. The annual precipitation has increased significantly all over southern Sweden due to increased winter precipitation. There is a trend of increasing maximum annual daily precipitation at only one station, where the annual maximum often occurs in winter. The number of events with a short return period is increasing, but the number of more extreme events has not increased. Daily and multi‐daily design storms of long return periods determined from extreme value analysis with updating year by year are not higher today than during the last 100 years. The largest daily storms are not related to stations with annual rainfall but seem to occur randomly. Copyright © 2013 John Wiley & Sons, Ltd.     

Continuous daily hydrograph simulation using duration curves of a precipitation index

The paper describes a parsimonious approach for generating continuous daily stream‐flow time‐series from observed daily rainfall data in a catchment. The key characteristic in the method is a duration curve. It is used to convert the daily rainfall information from source rain gauges into a continuous daily hydrograph at the destination river site. For each source rain gauge a time‐series of rainfall related ‘current precipitation index’ is generated and its duration curve is established. The current precipitation index reflects the current catchment wetness and is defined as a continuous function of precipitation, which accumulates on rainy days and exponentially decays during the periods of no rainfall. The process of rainfall‐to‐runoff conversion is based on the assumption that daily current precipitation index values at rainfall site(s) in a catchment and the destination site's daily flows correspond to similar probabilities on their respective duration curves. The method is tested in several small catchments in South Africa. The method is designed primarily for application at ungauged sites in data‐poor regions where the use of more complex and information consuming techniques of data generation may not be justified. Copyright © 2000 John Wiley & Sons, Ltd.     

Stochastic evaluation of simple pairing approaches to reconstruct incomplete rainfall time series

Two-station pairing approaches are routinely used to infill missing information in incomplete rainfall databases. We evaluated the performance of three simple methodologies to reconstruct incomplete time series in presence of variable nonlinear correlation between data pairs. Nonlinearity stems from the statistics describing the marginal peak-over-threshold (POT) values of rainfall events. A Monte Carlo analysis was developed to quantitatively assess expected errors from the use of chronological pairing (CP) with linear and nonlinear regression and frequency pairing (FP). CP is based on a priori selection of regression functions, while FP is based on matching the probability of non-exceedance of an event from one time series with the probability of non-exceedance of a similar event from another time series. We adopted a generalized Pareto (GP) model to describe POT events, and a t-copula algorithm to generate reference nonlinearly correlated pairs of random temporal distributions distributed according with the GP model. The results suggest that the optimal methodology strongly depends on GP statistics. In general, CP seems to provide the lowest errors when GP statistics were similar and correlation became linear; we found that a power-2 function performs well for the selected statistics when the number of missing points is limited. FP outperforms the other methods when POT statistics are different and variables are markedly nonlinearly correlated. Ensemble-based results seem to be supported by the analysis of observed precipitation at two real-world gauge stations.     

Determination of bare surface soil moisture from combined temporal evolution of land surface temperature and net surface shortwave radiation

Land surface soil moisture (SSM) is an important variable for hydrological, ecological, and meteorological applications. A multi‐linear model has recently been proposed to determine the SSM content from the combined diurnal evolution of both land surface temperature (LST) and net surface shortwave radiation (NSSR) with the parameters TN (the LST mid‐morning rising rate divided by the NSSR rising rate during the same period) and t_d (the time of daily maximum temperature). However, in addition to the problem that all the coefficients of the multi‐linear model depend on the atmospheric conditions, the model also suffers from the problems of the nonlinearity of TN as a function of the SSM content and the uncertainty of determining the t_d from the diurnal evolution of the LST. To address these problems, a modified multi‐linear model was developed using the logarithm of TN and normalizing t_d by the mid‐morning temperature difference instead of using the TN and t_d. Except for the constant term, the coefficients of all other variables in the modified multi‐linear model proved to be independent of the atmospheric conditions. Using the relevant simulation data, results from the modified multi‐linear model show that the SSM content can be determined with a root mean square error (RMSE) of 0.030m³/m³, provided that the constant term is known or estimated day to day. The validation of the model was conducted using the field measurements at the Langfang site in 2008 in China. A higher correlation is achieved (coefficient of determination: R² = 0.624, RMSE = 0.107m³/m³) between the measured SSM content and the SSM content estimated using the modified multi‐linear model with the coefficients determined from the simulation data. Another experiment is also conducted to estimate the SSM content using the modified model with the constant term calibrated each day by one‐spot measurements at the site. The estimation result has a relatively larger error (RMSE = 0.125m³/m³). Additionally, the uncertainty of the determination of the coefficients is analysed using the field measurements, and the results indicate that the SSM content obtained using the modified model accurately characterizes the surface soil moisture condition. Copyright © 2012 John Wiley & Sons, Ltd.     

Comparison of radar and gauge precipitation data in watershed models across varying spatial and temporal scales

Precipitation is a key control on watershed hydrologic modelling output, with errors in rainfall propagating through subsequent stages of water quantity and quality analysis. Most watershed models incorporate precipitation data from rain gauges; higher‐resolution data sources are available, but they are associated with greater computational requirements and expertise. Here, we investigate whether the Multisensor Precipitation Estimator (MPE or Stage IV Next‐Generation Radar) data improve the accuracy of streamflow simulations using the Soil and Water Assessment Tool (SWAT), compared with rain gauge data. Simulated flows from 2002 to 2010 at five timesteps were compared with observed flows for four nested subwatersheds of the Neuse River basin in North Carolina (21‐, 203‐, 2979‐, and 10 100‐km² watershed area), using a multi‐objective function, informal likelihood‐weighted calibration approach. Across watersheds and timesteps, total gauge precipitation was greater than radar precipitation, but radar data showed a conditional bias of higher rainfall estimates during large events (>25–50 mm/day). Model parameterization differed between calibrations with the two datasets, despite the fact that all watershed characteristics were the same across simulation scenarios. This underscores the importance of linking calibration parameters to realistic processes. SWAT simulations with both datasets underestimated median and low flows, whereas radar‐based simulations were more accurate than gauge‐based simulations for high flows. At coarser timesteps, differences were less pronounced. Our results suggest that modelling efforts in watersheds with poor rain gauge coverage can be improved with MPE radar data, especially at short timesteps. Published 2013. This article is a U.S. Government work and is in the public domain in the USA.     

The Use of Radar and Gauge Measurements to Estimate Areal Precipitation for Several Czech River Basins

Two methods estimating areal precipitation for selected river basins in the Czech Republic are compared. The methods use radar precipitation (the radar-derived precipitation estimate based on column maximum reflectivity) and data from 81 on-line rain gauges routinely provided by the Czech Hydrometeorological Institute. Data from a dense network of climatological rain gauges (the average inter-station distance is approximately 8 km), the measurements of which are not available in real time, are utilized for the verification. The mean areal precipitation, which is used as the ground truth, is obtained by the weighted interpolation of the dense rain gauge network. The accuracy of the methods is evaluated by the root-mean-square-error.The first, pixel-related method merges radar precipitation with rain gauge data to obtain adjusted pixel values. The adjusting procedure combines radar and gauge values in one variable that is interpolated into all radar pixels. The adjusted pixel precipitation is calculated from radar precipitation and from the value of the combined variable. The areal estimates are determined by adding the corresponding pixel values. The second method applies a linear regression model to describe the relationship between the areal precipitation (dependent variable) and its estimates, which are determined from (i) non-adjusted radar precipitation and (ii) on-line rain gauge measurements interpolated into pixels. Classical linear regression, ridge regression and robust regression models are tested.Both the methods decrease the average areal error in comparison with the reference method, which uses the on-line rain gauge data only. The decrease is about 10% and 15% for the pixel-related and regression methods, respectively. When the estimates of the pixel-related method are included as predictors into the regression method then the improvement of accuracy is almost 25%.     

Probability of combined high sea levels and large rains in Malmö, Sweden,southern Öresund

Observations of sea level and precipitation in Malmö, Sweden in the southeastern part of the sound Öresund have been used to estimate the probabilities of local compound events of high sea level and large daily and hourly rains. There are observations of sea level and daily rains extending back to 1930. The observations of short‐term rainfall are from 1980 and onwards. Most large rainfalls come in the summer, while the highest sea levels are in the autumn and in the winter. The highest observed sea level is about 130 cm above mean sea level, and the largest daily rain is close to 100 mm. However, the highest sea level observed during a day with rainfall corresponding to the 1‐year rain is less than 60 cm. The highest sea level observed during an hour with 1‐year hourly rainfall is 30 cm. From the statistics of daily rains, hourly rains and sea level, extreme values for each of them have been computed. For events with frequency higher than one per four years the probabilities of combined events sea level – rainfall are determined directly from the observations. For more rare events, marginal distributions of sea level and rainfall are determined. Copulas and conditional probabilities are used. When the sea level exceeds 20 cm above mean sea level, daily rains exceeding 10 mm are almost independent of the sea level and so are hourly rains exceeding 5 mm. It is extremely rare that large rains occur when the sea level is very high. The combination of 1‐year rainfall and the 1‐year sea level has a return period of more than 200 years.     

Winter rainfall interception by two mature open‐grown trees in Davis,California

A rainfall interception measuring system was developed and tested for open‐grown trees. The system includes direct measurements of gross precipitation, throughfall and stemflow, as well as continuous collection of micrometeorological data. The data were sampled every second and collected at 30‐s time steps using pressure transducers monitoring water depth in collection containers coupled to Campbell CR10 dataloggers. The system was tested on a 9‐year‐old broadleaf deciduous tree (pear, Pyrus calleryana ‘Bradford’) and an 8‐year‐old broadleaf evergreen tree (cork oak, Quercus suber) representing trees having divergent canopy distributions of foliage and stems. Partitioning of gross precipitation into throughfall, stemflow and canopy interception is presented for these two mature open‐grown trees during the 1996–1998 rainy seasons. Interception losses accounted for about 15% of gross precipitation for the pear tree and 27% for the oak tree. The fraction of gross precipitation reaching the ground included 8% by stemflow and 77% by throughfall for the pear tree, as compared with 15% and 58%, respectively, for the oak tree. The analysis of temporal patterns in interception indicates that it was greatest at the beginning of each rainfall event. Rainfall frequency is more significant than rainfall rate and duration in determining interception losses. Both stemflow and throughfall varied with rainfall intensity and wind speed. Increasing precipitation rates and wind speed increased stemflow but reduced throughfall. Analysis of rainfall interception processes at different time‐scales indicates that canopy interception varied from 100% at the beginning of the rain event to about 3% at the maximum rain intensity for the oak tree. These values reflected the canopy surface water storage changes during the rain event. The winter domain precipitation at our study site in the Central Valley of California limited our opportunities to collect interception data during non‐winter seasons. This precipitation pattern makes the results more specific to the Mediterranean climate region. Copyright © 2000 John Wiley & Sons, Ltd.     

Ensemble rainfall forecasting with numerical weather prediction and radar‐based nowcasting models

The overall objective of this study is to improve the forecasting accuracy of the precipitation in the Singapore region by means of both rainfall forecasting and nowcasting. Numerical Weather Predication (NWP) and radar‐based rainfall nowcasting are two important sources for quantitative precipitation forecast. In this paper, an attempt to combine rainfall prediction from a high‐resolution mesoscale weather model and a radar‐based rainfall model was performed. Two rainfall forecasting methods were selected and examined: (i) the weather research and forecasting model (WRF); and (ii) a translation model (TM). The WRF model, at a high spatial resolution, was run over the domain of interest using the Global Forecast System data as initializing fields. Some heavy rainfall events were selected from data record and used to test the forecast capability of WRF and TM. Results obtained from TM and WRF were then combined together to form an ensemble rainfall forecasting model, by assigning weights of 0.7 and 0.3 weights to TM and WRF, respectively. This paper presented results from WRF and TM, and the resulting ensemble rainfall forecasting; comparisons with station data were conducted as well. It was shown that results from WRF are very useful as advisory of anticipated heavy rainfall events, whereas those from TM, which used information of rain cells already appearing on the radar screen, were more accurate for rainfall nowcasting as expected. The ensemble rainfall forecasting compares reasonably well with the station observation data. Copyright © 2012 John Wiley & Sons, Ltd.     

Cokriging for enhanced spatial interpolation of rainfall in two Australian catchments

Rainfall data in continuous space provide an essential input for most hydrological and water resources planning studies. Spatial distribution of rainfall is usually estimated using ground‐based point rainfall data from sparsely positioned rain‐gauge stations in a rain‐gauge network. Kriging has become a widely used interpolation method to estimate the spatial distribution of climate variables including rainfall. The objective of this study is to evaluate three geostatistical (ordinary kriging [OK], ordinary cokriging [OCK], kriging with an external drift [KED]), and two deterministic (inverse distance weighting, radial basis function) interpolation methods for enhanced spatial interpolation of monthly rainfall in the Middle Yarra River catchment and the Ovens River catchment in Victoria, Australia. Historical rainfall records from existing rain‐gauge stations of the catchments during 1980–2012 period are used for the analysis. A digital elevation model of each catchment is used as the supplementary information in addition to rainfall for the OCK and kriging with an external drift methods. The prediction performance of the adopted interpolation methods is assessed through cross‐validation. Results indicate that the geostatistical methods outperform the deterministic methods for spatial interpolation of rainfall. Results also indicate that among the geostatistical methods, the OCK method is found to be the best interpolator for estimating spatial rainfall distribution in both the catchments with the lowest prediction error between the observed and estimated monthly rainfall. Thus, this study demonstrates that the use of elevation as an auxiliary variable in addition to rainfall data in the geostatistical framework can significantly enhance the estimation of rainfall over a catchment.     

Spatially distributed long‐term hydrologic simulation using a continuous SCS CN method‐based hybrid hydrologic model

A continuous Soil Conservation Service (SCS) curve number (CN) method that considers time‐varied SCS CN values was developed based on the original SCS CN method with a revised soil moisture accounting approach to estimate run‐off depth for long‐term discontinuous storm events. The method was applied to spatially distributed long‐term hydrologic simulation of rainfall‐run‐off flow with an underlying assumption for its spatial variability using a geographic information systems‐based spatially distributed Clark's unit hydrograph method (Distributed‐Clark; hybrid hydrologic model), which is a simple few parameter run‐off routing method for input of spatiotemporally varied run‐off depth, incorporating conditional unit hydrograph adoption for different run‐off precipitation depth‐based direct run‐off flow convolution. Case studies of spatially distributed long‐term (total of 6 years) hydrologic simulation for four river basins using daily NEXRAD quantitative precipitation estimations demonstrate overall performances of Nash–Sutcliffe efficiency (E_NS) 0.62, coefficient of determination (R²) 0.64, and percent bias 0.33% in direct run‐off and E_NS 0.71, R² 0.72, and percent bias 0.15% in total streamflow for model result comparison against observed streamflow. These results show better fit (improvement in E_NS of 42.0% and R² of 33.3% for total streamflow) than the same model using spatially averaged gauged rainfall. Incorporation of logic for conditional initial abstraction in a continuous SCS CN method, which can accommodate initial run‐off loss amounts based on previous rainfall, slightly enhances model simulation performance; both E_NS and R² increased by 1.4% for total streamflow in a 4‐year calibration period. A continuous SCS CN method‐based hybrid hydrologic model presented in this study is, therefore, potentially significant to improved implementation of long‐term hydrologic applications for spatially distributed rainfall‐run‐off generation and routing, as a relatively simple hydrologic modelling approach for the use of more reliable gridded types of quantitative precipitation estimations.     

Spatial interpolation using nonlinear mathematical programming models for estimation of missing precipitation records

Abstract

New mathematical programming models are proposed, developed and evaluated in this study for estimating missing precipitation data. These models use nonlinear and mixed integer nonlinear mathematical programming (MINLP) formulations with binary variables. They overcome the limitations associated with spatial interpolation methods relevant to the arbitrary selection of weighting parameters, the number of control points within a neighbourhood, and the size of the neighbourhood itself. The formulations are solved using genetic algorithms. Daily precipitation data obtained from 15 rain gauging stations in a temperate climatic region are used to test and derive conclusions about the efficacy of these methods. The developed methods are compared with some naïve approaches, multiple linear regression, nonlinear least-square optimization, kriging, and global and local trend surface and thin-plate spline models. The results suggest that the proposed new mathematical programming formulations are superior to those obtained from all the other spatial interpolation methods tested in this study.

Editor D. Koutsoyiannis; Associate editor S. Grimaldi

Citation Teegavarapu, R.S.V., 2012. Spatial interpolation using nonlinear mathematical programming models for estimation of missing precipitation records. Hydrological Sciences Journal, 57 (3), 383–406.     

Nonlinear analysis of rainfall dynamics in California's Sacramento Valley

This study investigates the dynamic nature of rainfall observed at the Sustainable Agriculture Farming Systems (SAFS) site in California's Sacramento Valley, which was established to study the benefits of winter cover cropping in Mediterranean irrigated‐arid systems. Rainfall data of four different temporal scales (i.e. daily, weekly, biweekly, and monthly) are analysed to determine the dynamic nature of precipitation in time. In an arid climate with seasonal precipitation this has large implications for land and water management, both in the short term and in the long term. A nonlinear dynamic technique (correlation dimension method) that uses the phase‐space reconstruction and dimension concepts is employed. Bearing in mind the possible effects of the presence of zeros (i.e. no rain) on the outcomes of this analysis, an attempt is also made to compare the dynamic nature of all‐year rainfall and winter rainfall. Analysis of 15 years of data suggests that rainfall dynamics at this site are dominated by a large number of variables, regardless of the scales and seasons studied. The dimension results also suggest that: (1) rainfall dynamics at coarser resolutions are more irregular than that at finer resolutions; (2) winter rainfall has a higher variability than all‐year rainfall. These results are indeed useful to gain information about the complexity of the rainfall process at this site with respect to (temporal) scales and seasons and, hence, the appropriate model (high‐dimensional) type. However, in view of the potential effects of certain rainfall data characteristics (e.g. zeros, measurement errors, scale effects) on the correlation dimension analysis, the discussion also emphasizes the need for further verification, and possibly confirmation, of these results. Copyright © 2005 John Wiley & Sons, Ltd.     

Bedload transport of small rivers in Malaysia

Numerous time-consuming equations, based on the relationship between the reliability and representativeness of the data utilized in defining variables and constants, require complex parameters to estimate bedload transport. In this study the easily accessible data including flow discharge, water depth, water surface slope, and surface grain diameter （ds0） from small rivers in Malaysia were used to estimate bedload transport. Genetic programming （GP） and artificial neural network （ANN） models are applied as complementary tools to estimate bed load transport based on a balance between simplicity and accuracy in small rivers. The developed models demonstrate higher performance with an overall accuracy of 97% and 93% for ANN and GP, respectively compared with other traditional methods and empirical equations.     

Downslope soil detachment–transport on steep slopes via rain splash

This study developed a one‐dimensional model of downslope rain splash transport based on field experiments and previous studies. The developed model considers soil detachment processes, ground cover, probability densities, and the effect of overland run‐off in preventing detachment. Field monitoring was conducted to observe precipitation run‐off, ground cover, and sediment production on steep hillslopes. Field‐observed data were used to develop the splash detachment rate equation, probability densities for splash transport, and the maximum splash transport distance. Observed and estimated splash transport showed overall agreement, with some differences for small storm events or events with relatively low intensity, probably caused by variation of overland run‐off depth and connectivity as well as differences in soil surface cohesion at various degrees of wetness. Our model can provide insights on the interactions among rainfall intensity, soil surface condition, soil wetness, and splash transport on forested hillslopes. Copyright © 2011 John Wiley & Sons, Ltd.     

Evaluation and hydrologic validation of TMPA satellite precipitation product downstream of the Pearl River Basin,China

With high spatio‐temporal resolution and wide coverage, satellite‐based precipitation products can potentially fill the deficiencies of traditional in situ gauge precipitation observations and provide an alternative data source for ungauged areas. However, due to the relatively poor accuracy and high uncertainty of satellite‐based precipitation products, it remains necessary to assess the quality and applicability of the products for each investigated area. This study evaluated the accuracy and error of the latest Tropical Rainfall Measuring Mission Multi‐satellites Precipitation Analysis 3B42‐V7 satellite‐based precipitation product and validated the applicability of the product for the Beijiang and Dongjiang River Basins, downstream of the Pearl River Basin in China. The study first evaluated the accuracy, error, and bias of the 3B42‐V7 product during 1998–2006 at daily and monthly scale via comparison with in situ observations. The study further validated the applicability of the product via hydrologic simulation using the variable infiltration capacity hydrological model for three hydrological stations in the Beijiang River Basin, considering two scenarios: a streamflow simulation with gauge‐calibrated parameters (Scenario I) and a simulation after recalibration with the 3B42‐V7 product (Scenario II). The results revealed that (a) the 3B42‐V7 product produced acceptable accuracy both at the daily scale and high accuracy at the monthly scale while generally tending to overestimate precipitation; (b) the product clearly overestimated the frequency of no rainfall events at the grid cell scale and light rainfall (<1 mm/day) events at the region scale and also overestimated the amount of heavy rain (25–50 mm/day) and hard rain (≥50 mm/day) events; (c) under Scenario I, the 3B42‐V7 product performed poorly at three stations with gauge‐calibrated parameters; under Scenario II, the recalibrated model provided significantly improved performance of streamflow simulation with the 3B42‐V7 product; (d) the variable infiltration capacity model has the ability to reveal the hydrological characteristics of the karst landform in the Beijiang Basin when using the 3B42‐V7 product.     

Effects of temporal scales and space mismatches on the TRMM 3B42 v7 precipitation product in a remote mountainous area

The accurate measurement of precipitation is essential to understanding regional hydrological processes and hydrological cycling. Quantification of precipitation over remote regions such as the Tibetan Plateau is highly unreliable because of the scarcity of rain gauges. The objective of this study is to evaluate the performance of the satellite precipitation product of tropical rainfall measuring mission (TRMM) 3B42 v7 at daily, weekly, monthly, and seasonal scales. Comparison between TRMM grid precipitation and point‐based rain gauge precipitation was conducted using nearest neighbour and bilinear weighted interpolation methods. The results showed that the TRMM product could not capture daily precipitation well due to some rainfall events being missed at short time scales but provided reasonably good precipitation data at weekly, monthly, and seasonal scales. TRMM tended to underestimate the precipitation of small rainfall events (less than 1 mm/day), while it overestimated the precipitation of large rainfall events (greater than 20 mm/day). Consequently, TRMM showed better performance in the summer monsoon season than in the winter season. Through comparison, it was also found that the bilinear weighted interpolation method performs better than the nearest neighbour method in TRMM precipitation extraction.