Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Evaluation of Satellite Precipitation Products for Estimation of Floods in Data-Scarce Environment

Evaluation of Satellite Precipitation Products for Estimation of Floods in Data-Scarce Environment Hindawi Advances in Meteorology Volume 2023, Article ID 1685720, 17 pages https://doi.org/10.1155/2023/1685720 Research Article Evaluation of Satellite Precipitation Products for Estimation of Floods in Data-Scarce Environment 1 2 1 1 Muhammad Masood , Muhammad Naveed , Mudassar Iqbal , Ghulam Nabi , 1 1 1 Hafiz Muhammad Kashif , Muhammad Jawad , and Ahmad Mujtaba Centre of Excellence in Water Resources Engineering, University of Engineering & Technology GT-Road, Lahore 54890, Pakistan Department of Civil Engineering, University of Engineering & Technology GT-Road, Lahore 54890, Pakistan Correspondence should be addressed to Muhammad Masood; chmasoud@gmail.com Received 22 November 2022; Revised 22 March 2023; Accepted 11 April 2023; Published 3 May 2023 Academic Editor: Tomeu Rigo Copyright © 2023 Muhammad Masood et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Utilization of satellite precipitation products (SPPs) for reliable food modeling has become a necessity due to the scarcity of conventional gauging systems. Tree high-resolution SPPs, i.e., Integrated Multi-satellite Retrieval for GPM (IMERG), Global Satellite Mapping of Precipitation (GSMaP), and Climate Hazards Group InfraRed Precipitation with Station (CHIRPS), data were assessed statistically and hydrologically in the sparsely gauged Chenab River basin of Pakistan. Te consistency of rain gauge data was assessed by the double mass curve (DMC). Te statistical metrics applied were probability of detection (POD), critical success index (CSI), false alarm ratio (FAR), correlation coefcient (CC), root mean square error (RMSE), and bias (B). Te hydrologic evaluation was conducted with calibration and validation scenarios for the monsoon fooding season using the Integrated Flood Analysis System (IFAS) and fow duration curve (FDC). Sensitivity analysis was conducted using ±20% cal- ibrating parameters. Te rain gauge data have been found to be consistent with the higher coefcient of determination (R ). Te mean skill scores of GSMaP were superior to those of CHIRPS and IMERG. More bias was observed during the monsoon than during western disturbances. Te most sensitive parameter was the base fow coefcient (AGD), with a high mean absolute sensitivity index value. During model calibration, good values of performance indicators, i.e., R , Nash−Sutclife efciency (NSE), and percentage bias (PBIAS), were found for the used SPPs. For validation, GSMaP performed better with comparatively higher values of R and NSE and a lower value of PBIAS. Te FDC exhibited SPPs’ excellent performance during 20% to 40% exceedance time. Pakistan’s history has emphasized the critical need for an 1. Introduction efcacious and steadfast food warning system [7]. Improved A food is a frequently recurring destructive natural hazard food modeling and remote sensing necessitate enhanced that humanity encounters [1]. Te adverse fooding con- food analysis methodology to send timely warnings to ditions have resulted in a signifcant deterioration of envi- populations and better reservoir operations [8]. Te early ronmental sustainability [2] throughout the world, wreaking warning system in Pakistan has limited capabilities. Al- havoc with costly infrastructure, food production, and though substantial improvements in food forecasting have human lives [3]. Globally, climate change and anthropogenic been made by utilizing the weather radar and telemetric activities have increased the frequency and intensity of systems in the warning system, there is still a need for many foods and exacerbated riverine food threats in several world eforts to advance the food forecasting and warning system places [4]. Pakistan is located in a highly vulnerable region to [9, 10]. climate change where foods occur almost every year since Precipitation is an important hydrological parameter the last three decades [5, 6]. Te devastating food in used for watershed management, food forecasting, and 2 Advances in Meteorology necessitated hydrological modeling. In Pakistan, limited climatological assessment [11]. It is also the most complex parameter because of its excessive spatiotemporal variations studies have been conducted to assess satellite precipitation products’ efectiveness, particularly utilizing the distributed that traditional rain gauges and radar networks cannot re- cord due to their sparsity [12]. Accurate and precise pre- IFAS model. In [38], the authors highlighted the scarcity of cipitation data with fne spatiotemporal resolution is hydrological data and the signifcance of upstream fow important for watershed management and food analysis boundary conditions when barrage operation standards are [13]. It necessitates the need for such techniques that sup- unknown in the Indus River. In [31], the authors used plements the rain gauge observations and provides excep- a lump and regional calibration approach to model the tional precipitation data to support hydrological modeling Jhelum river basin. In [39], the authors pointed out the difculty of food modeling at the confuence point of the issues [14–16]. Remote sensing satellites use refected light to detect, Chenab and Jhelum basins. In [38], the authors explored that the performance of IFAS can be improved by utilizing local collect, measure, and record the electromagnetic energy from the earth’s surface [17]. Advances in satellite remote soil texture data in the Indus River. In [38], the authors evaluated the precipitation results from diferent sources sensing have made it an excellent data source, as it can provide metrological data to support hydrological modeling for modeling the Indus River’s middle reach. Aziz [40] issues [18–20]. Moreover, precipitation data obtained from demonstrated that IFAS could be used for hydrological remote sensing have the potential to supplement the tra- modeling of the Kabul River with data scarcity. In [41], the ditional rain gauge system [21]. authors investigated that integrating satellite and gauge Recent studies have shown that the precipitation data rainfall data can enhance food forecasting in the estimated through satellite-based observations contributed Philippines-Cagayan River catchment. In [42], the authors recommended that improved satellite precipitation data be well to detecting rainfall distribution and severity in data- scarce regions [22, 23]. However, the satellite products may used to enhance food prediction in the Dungun River basin, Malaysia. have errors due to indirect estimation, sampling uncertainty, and retrieval algorithms [24–27]. Te properties of these Assessment of SPPs with a fully distributed hydrological model under diferent calibration scenarios is yet to be errors signifcantly vary in contrasting climates, storms, seasons, and altitudes [11, 28]. Terefore, it is essential to evaluated in the study area. In addition, the representation of validate the accuracy of satellite precipitation products and hydrological signatures with diferent rainfall data sets is yet suitability for a broader range of environments. Satellite to be explored. It mandates the investigation of satellite rainfall data can be validated using statistical analysis re- precipitation data sets using a distributed hydrological garding ground-based on gauge data and a proper hydro- model for diferent applications. In this study, three SPP- logical modeling framework [29]. Statistical analysis based datasets, i.e., IMERG, GSMaP, and CHIRPS, have been evaluated statistically and hydrologically in a data-scar determines the accuracy and consistency of satellite pre- cipitation data, while hydrological simulation elucidates the region, i.e., the Chenab River catchment of Pakistan. Te study utilized the IFAS model to generate streamfow for usefulness and application of the same datasets [30]. For reliable hydrological modeling, a proper calibration tech- a sparsely gauged catchment by using satellite precipitation nique, parametric sensitivity, and model capability are of datasets and derived hydrological signatures. primary consideration. A suitable calibration technique is critical because errors in model calibration and input 2. Study Area and Data Description datasets contribute to incorrect outcomes [31]. Similarly, parametric sensitivity produces only important parameters, 2.1. Study Area. Te Chenab River starts in Himachal reduces the analysis time, and contributes to modeling Pradesh, India, at the confuence of the Bhaga and Chandra calibration [32]. streams and fows across Indian-controlled Kashmir to Furthermore, the ability of the hydrological model to Pakistan [43]. Te catchment of the Chenab River covers an simulate water fow can be examined using the FDC, “a key area of about 26,000 km up to the Marala Barrage. It runof variability signature” [33]. Te FDC ofers additional embraces 97% of this catchment area in India, while only 3% details on the basins’ hydrological modeling and underlying in Pakistan up to the Marala barrage [44]. In Pakistan, there processes [34, 35]. Generally, lumped, semidistributed, and are four streamfow gauge stations on the river, i.e., Marala distributed models are used for a watershed’s hydrological barrage, Khanki barrage, Qadirabad barrage, and Trimmu modeling. Lumped models consider spatially uniform wa- barrage. For the present research, the study area ranged from tershed characteristics. Te semidistributed model divides the Marala barrage to the Trimmu barrage (Figure 1). the watershed into subbasins with unique hydrological re- Since the Chenab River and Jhelum River converged at sponses [36]. Conversely, distributed models consider the the Trimmu barrage, hydrological modeling at the con- spatial and temporal variation of physical properties in the vergence point is not possible [39]. So, the study considered watershed. Tese models can interpolate the rainfall data and the assessment of outfow at Qadir Abad barrage with an predict water fow at ungauged locations. However, these assumption of free fow at Khanki barrage during the models require a considerable input data set for fow monsoon fooding season. Te selected catchment is situated ° ° ° ° estimations [37]. between latitudes 72 –78 E and longitudes 32 –34 N, Recently, studies reported that statistical assessment of spanning over ∼16,000 km , with a gradient of 0.4 m/km precipitation data had not yielded reliable results that downstream of plain areas [43]. Advances in Meteorology 3 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 40000 120000 200000 280000 360000 440000 520000 600000 680000 760000 840000 920000 W E 1 cm = 49 km Kilometers 0 30 60 120 180 240 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 40000 120000 200000 280000 360000 440000 520000 600000 680000 760000 840000 920000 Asia Elevations 130 - 920 Pakistan_and_Kashmir_Boundary 930 - 2,200 Study_Area_Boundary 2,300 - 3,400 Headworks 3,500 - 4,500 Upstream Raingages 4,600 - 7,100 Downstream Raingages Streams Figure 1: Location map of the Chenab River Basin showing catchment area, upstream and downstream gauging stations of Marala Barrage. Te crosssectional characteristics of the basin comprise Satellite precipitation datasets at the daily time scale 85 km in length with an average width of 800 m. Te scarcity (Table 1) consisting of GSMaP, IMERGE, and CHIRPS and sparsity of meteorological gauging stations are big con- were downloaded from the Japan Aerospace Exploration cerns in this region. Tere are only four rain gauges organized Agency (JAXA) (https://sharaku.eorc.jaxa.jp/GSMaP/), by the Pakistan Meteorological Department (PMD), which do National Aeronautics and Space Administration (NASA) (https://pmm.nasa.gov/data-access/downloads/gpm), and not meet the requirements of the World Meteorological Or- ganization (WMO) and are inadequate for hydrological the University of California, Santa Barbara’s Climate modeling for watershed management. Te central hydrology of Hazards Group (UC Santa Barbara) (https://chg.geog. the catchment is controlled by the summer monsoon and ucsb.edu/data/chirps/), respectively. GSMaP consists of winter seasons, where the summer monsoon season dominates four types of products; two real-time (GSMaP-NRT, and has triggered signifcant fooding in this region. GSMaP-Gauge, and NRT) and two postreal-time (GSMaP-Gauge, GSMaP-MVK). In the present work, 2.2. Data Description. Te data for the research was collected GSMaP-Gauge NRT (version 6) was used. In order to at a daily scale for the years 2015–2020 and consisted of formulate the GSMaP-Gauge NRT precipitation pre- a gauge rainfall dataset, observed streamfow, and a satellite dictions with a 4 h latency period, the error parameters precipitation dataset. In addition, the topographical data estimated for the postreal-time product of GSMaP-Gauge comprises a digital elevation model, land use, and soil type. are utilized. GSMaP-Gauge also employed a blending of Daily rainfall data for the selected rain gauge stations were passive microwave (PMW) and infrared (IR) data along acquired from the PMD. Streamfow data for stream gauging with a unifed gauge-based analysis of the global daily stations were collected from Pakistan’s Flood Forecasting precipitation dataset from the Climate Prediction Center Division (FFD). (CPC) [22]. 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 3240000 3300000 3360000 3420000 3480000 3540000 3600000 3660000 3720000 3780000 3840000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 3210000 3270000 3330000 3390000 3450000 3510000 3570000 3630000 3690000 3750000 3810000 4 Advances in Meteorology Table 1: Specifcations of the three SPPs used in the study. Satellite precipitation datasets Specifcation GSMaP IMERG CHIRPS Developer JAXA NASA USGS and CHC ° ° ° Spatial resolution 0.1 0.1 0.05 Temporal resolution Daily Daily Daily Coverage 60 N-60 S 90 N-90 S 50 N-50 S Data availability 2014-onwards 2014-onwards 1981-onwards Latency 4 hours 14 hours 48 hours Frequency 1 hour 0.5 hour 24 hours Algorithm PMW and IR GMI and DPR CCD and gauge Version 6 6 2 Similarly, Global Precipitation Measurement (GPM) is precipitation records so that inconsistency does not sig- a multinational satellite project to integrate and enhance nifcantly infuence the average in one of the station’ records. precipitation observations from diferent satellites. IMERG If there are less than 10 stations located in a specifc region, is a GPM-based level 3 multisatellite precipitation algorithm the consistency of each station must be examined. Terefore, that incorporates all passive microwave and infrared-based at all four gauging stations located in the study area, the observations in the constellation. Typically, three products DMC technique was applied to examine the consistency of of IMERG precipitation (early-run, late-run, and fnal-run) annual rainfall data. are mostly considered. Early-run uses only forward morphing with a 4 hour latency period; however, late-run 3.2. Evaluation Statistics. Te efcacy of selected satellite and fnal-run use forward and backward morphing with precipitation products was assessed against four-gauge latency periods of 14 hours and 3.5 months, respectively station records with categorical and continuous metrics [45]. IMERG combines passive microwave, propagated pulse on a daily, ten daily, monthly, and seasonal scales from 2015 width modulation, and infrared radiation-based observa- to 2020. Diferent approaches have been used by comparing tions by the Kalman flter method to obtain precise esti- point precipitation data observed by rain gauges with pixel mation [46]. Tis research work utilized daily IMERG- precipitation data recorded by remote sensing satellites. late-run version 6 for statistical and hydrological assess- Usually, such procedures are based upon the upscale in- ments in the study area. terpolation of point values to grid scale data and the Moreover, CHIRPS is a quasi-global rainfall dataset that downscaling of grid data towards point values. Te rean- consists of three types of temporal data (daily, pentanal alyzed gridded data always vary from the station observa- (5 days), and monthly) with two types of spatial resolution tions and vice versa in several aspects. To avoid inaccuracies ° ° (0.05 , 0.25 ). Te daily data are considered real-time data caused by such upscale interpolation methods and down- with a latency period of 2 days, while pentanal and monthly scaling, a more direct approach has been proposed and used. data are considered post-real-time datasets with a 21 days In this approach, precipitation observed at stations falling latency period. Its algorithm is based on cold cloud duration within a grid cell will be averaged to obtain an estimate for (CCD) and ground gauge observations to approximate the the observed precipitation at the center of that grid cell and rainfall. Tis study employed the daily 0.05 grid CHIRPS then compared to the gridded value [21]. Tis approach has version 2.0 dataset [47]. been used in this study for comparison between rain gauge Furthermore, digital elevation model (DEM) was used to and satellite data sets. For this purpose, satellite precipitation represent the catchment’s topography and delineate the data at the daily time scale was downloaded and then watershed. Te present study collected DEM and Global converted into ten daily, monthly, and seasonal scales. In Map’s land cover data (Version 2) from the International selecting the tile of satellite precipitation data, PMD gauge Steering Committee for Global Mapping (ISCGM). Finally, value recording time (8:00 am daily) was kept in focus. Event the soil type data based on the Digital Soil Map of the World detection capability was evaluated with categorical metrics (DSMW), provided by the Food and Agriculture Organi- that include POD, FAR, and CSI. POD refects the ratio of zation (FAO), was used. accurately identifed rainfall events by the satellite con- cerning gauge rainfall data. FAR demonstrates the fraction of rainfall events in which the satellite predicts precipitation 3. Research Methodology while the rain gauge does not observe it. CSI represents 3.1. Double Mass Curve Analysis. DMC is employed to in- typically the fraction of rainfall occurrence accurately rec- spect the consistency of the hydrologic data and to adjust the ognized by the satellite. Continuous metrics measure the inconsistent precipitation data. In this graphical approach, quantitative diference between observed and predicted the cumulative data of a single station is compared with the precipitation. Tese metrics include bias, CC, and RMSE. pattern composed of cumulative data from other stations in Bias is the mean discrepancy between satellite estimation the area. Likewise, for the used pattern, enough gauging and rain gauge data. Depending on the quality of the rainfall stations must be included while checking the consistency of data, its value could be positive or negative, indicating Advances in Meteorology 5 overestimation and underestimation, respectively. CC esti- 3.4. Model Formulation. For the development of the IFAS mates the degree of agreement between the satellite and rain model, the extent of the target study area was defned by determining the latitude and longitude of the selected gauge precipitation data. RMSE depicts the mean dispersion of predicted precipitation around the known value of gauge catchment. Te IFAS model with a two-layered and three- observations. It is used to evaluate the precision of the tank structure was created by customizing the digital-based rainfall dataset [48]. From historical data, two rainy seasons, land cover, elevation, and soil type data to the appropriate monsoon (June to September) and westerly disturbance grid size. Te shapefle of the study area was imported into (November to February), have been established in the study the basin manager function of IFAS to defne basin and sub area: these were considered for seasonal evaluation at the basin boundaries. Te surface tank parameters were esti- daily scale. Tese statistical metrics are given as follows: mated utilizing the land cover data, while the aquifer tank parameters were tuned according to soil type data. Te POD � , essential aspect of hydrological modeling in a basin is ac- H + M curately estimating runof and water level initial conditions in the river course that afect the parameter optimization of FAR � , the model [31]. Te model was run six months before the H + F calibration of the food event to generate proper initial conditions until hydrological equilibrium was achieved. Te CSI � , principle of equifnality dictates that many combinations of H + M + F parameters are possible that give good agreement with the (1) 􏽐 (Si − Gi) i�1 observed streamfow data. Boundary conditions are essen- BIAS � , tial, especially when hydrological data is scarce and stan- 􏽳������������� dards for barrage operations are unknown. IFAS has an n 2 (Si − Gi). integrated water resources management (IWRM) interface 􏼐􏽐 i�1 RMSE � , that contains various techniques to incorporate barrage operating tasks. Te discharge fle technique was applied using the IWRM function to give the boundary condition in 􏽐 (Gi − Gm)(Si − Sm) I�1 􏽱������������� �􏽱������������� CC � , this study condition. Tis technique employed daily dis- n 2 n 2 􏽐 (Gi − Gm). 􏽐 (Si − Sm). i�1 i�1 charge data in the CSV fle to represent barrage operations. Marala barrage outfows were considered boundary con- where H, M, and F exhibit the number of hit, miss, and false ditions in this research due to data scarcity and unknown alarm events, while Gi and Si denote the gauge and satellite barrage operations upstream of the catchment. precipitation, Gm and Sm represent the mean of gauge and satellite precipitation data, and N indicates the total number of events used for evaluation. 3.5. Calibration Scenarios and Model Performance Indicators. Since the outputs of hydrological models are rarely capable of accurately refecting nature in its completeness, their 3.3. Explication of Hydrological (IFAS) Model. IFAS is a succinct runof analysis toolkit designed for food pre- performance must be evaluated before they can be employed in any decision-making process. Te IFAS is designed for diction in basins with insufcient hydrological and geo- physical information in developing countries. It is food analysis; therefore, it was calibrated and validated categorized as a physically distributed framework that can utilizing precipitation data collected during the monsoon integrate gauge rainfall data, satellite-based precipitation seasons (July to October) of 2015 and 2017, considering data, evaporation data, snowmelt data, and geophysical data medium and high fooding years, respectively (PMD/FFD). to simulate river course fow. It integrates grid-based Te model was calibrated individually using CHIRPS, datasets of topography, geology, and land cover to esti- GSMaP, and IMERG precipitation datasets. All datasets were validated against each calibration scenario. Te cali- mate the parameters of the physical conditions of a basin [49]. Te model can generate the channel network using bration of the model was achieved through a trial-and-error process. Te model performance was evaluated using model topographical data to defne the basin, sub-basin boundaries, 2 2 fow direction, and drainage patterns. It employs the Public performance indicators (MPI), R , NSE, and PBIAS. Te R Works Research Institute Distributed Hydrological Model is a statistical indicator representing the fraction of the (PWRI-DHM), consisting of a two- or three-tank structure dependent variable’s variance predicted by the independent and a routing model for runof simulation. Te three-tank variable. Te ideal value is 1, while a lower value than 1 structure comprises a surface, sub surface, and aquifer, while reveals the variation of model output. Te model perfor- the routing model comprises a kinematic hydraulic river mance with R > 0.5 is acceptable. Likewise, the NSE is the course routing tank. PWRI-DHM uses a nonlinear re- most often used method for determining correlation to test lationship to calculate each cell’s outfow based on the tank the efcacy of hydrological models. Te literature reveals model philosophy, considering Manning’s equation, Darcy’s a variety of acceptable, very good, and excellent value cat- law, and hyperbolic approximations. It uses a kinematic egories for NSE. Te calibration tolerance criteria are very wave equation to calculate the discharge in the river course subjective. Calibration with NSE is generally perceived as tank [38, 50, 51]. good if it is higher than 0.6 and excellent if it is more 6 Advances in Meteorology signifcant than 0.8 in the literature. Te model validation streamfow was taken as the baseline and the variation in criteria are less restrictive than the calibration levels. A value simulated fow was evaluated. Furthermore, the ability of of NSE greater than 0.5 is acceptable for validation, while each precipitation dataset to generate high and medium NSE greater than 0.7 is considered highly excellent. Te fows was examined through dependable fow exceedance, PBIAS examines the average tendency of the simulated fows where the extreme food events were represented in the to be greater or smaller than their observed fows. Te range of Q5–Q25 dependable fows, the medium fow re- perfect value of PBIAS is 0, and lower values represent quired for irrigation was designated by Q50 dependable accurate model reproduction [52]. fows, and Q70 dependable fows correspond to the water Tese MPIS are given as follows: availability for domestic supply. 􏽐 (Oi − Om)(Pi − Pm) I�1 􏽱������������ �􏽱������������ R � , 4. Results and Discussion n n 􏽐 (Oi − Om) 􏽐 (Pi − Pm) i�1 i�1 4.1. Consistency of Gauge Rainfall Data. Double mass n 2 analysis has been used to check the consistency of rainfall (Oi − Pi). i�1 (2) NSE � 1 − , data records at four stations, i.e., Gujrat, Sialkot, Sargodha, n 2 􏽐 (0i − Om). i�1 and Jhang. Te cumulative data of a single station was compared with the cumulative data of other stations in the 􏽐 (Oi − Pi) i�1 respective area. PBIAS � 100 ∗ , 􏽐 (0i) Te straight line shows data consistency, whereas any i�1 change in the straight line manifests a change in the data where Oi and Pi represent the observed and simulated fows, collection method that afects the relationship. For any Om and Pm are denoted by the mean values of the observed station, the rise of the curve from the trend line shows that and simulated fows, and n is the total number of events. there was more annual rainfall than in other stations. For all gauging stations, the R values were 0.98 to 0.99, and the 3.6. Implications of Sensitivity Analysis Technique. Te annual rainfall data were consistent with the DMC technique sensitivity analysis helps to examine the nonlinear variation (Figure 2). where CAR � is the cumulative annual rainfall. of highly uncertain parameters in the complex models. Te IFAS model was investigated to determine the most sensitive 4.2. Statistical Evaluation of SPPs at Diferent Temporal Scales. calibrating parameters severely afecting the calibrating Tis study identifed and quantifed the errors associated hydrograph. In this regard, the values of all calibrated pa- with satellite datasets. Te efcacy of selected satellite rameters were frst increased and then decreased by 20%, one products (CHIRPS, IMERG, and GSMaP) was assessed by one of their calibrated values. C is the 20 percent +20% statistically at daily, 10-daily, monthly, annual, and seasonal change in the calibrated parameter value and is determined scales using precipitation data recorded at PMD stations. using equation (3). So are the simulation results of a cali- brated hydrograph, and S is the change that occurred in +20% the simulation results when one of the calibrating param- 4.2.1. Daily and 10-Daily Scale. Statistical evaluations of eters’ values changed to +20%. Te mean percentage change selected SPPs at daily and 10-daily levels are presented in in the simulation results by increasing or decreasing the Table 2. In the case of categorical metrics, the mean POD of calibrating parameter value to 20% is called the sensitivity GSMaP was better than CHIRPS and IMERG on both index (I) and can be determined using equation (4). Te temporal scales. Te mean POD for CHIRPS and IMERG mean absolute sensitivity index (MASI) can be determined was lower by 50.79% and 22.22% for the daily scale with using equation (5). reference to GSMaP. In terms of mean FAR values, IMERG and GSMaP showed good agreement, and CHIRPS S − S 􏼁 0 ±20 C � 􏼨 􏼩 × 100, (3) underperformed. CSI gives more stable results due to the ±20% characteristics of the blending of POD and FAR. Te per- formance of GSMaP, IMERG, and CHIRPS was improved by C + C + −20 54%, 58%, and 55%, respectively, for CSI at the 10-daily (4) I � , ±20% scale. Remarkably better values of categorical metrics were 􏼌 􏼌 􏼌 􏼌 given by all the used satellite products at the 10 daily time 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 􏽮 C + C 􏽯 􏼌 􏼌 􏼌 􏼌 +20 −20 scale as compared to the daily time scale. (5) MASI � . In the case of continuous metrics, the mean BIAS of GSMaP was better than the other two products at daily scale. A slight diference was observed between the mean RMSE 3.7. Flow Assessment Using Hydrological Signature. Te FDC values of the selected products on both temporal scales. is an infuential streamfow variability signature that de- However, it was noted that the used SPPs showed less errors scribes hydrological behavior. It is the graphical represen- (BIAS and RMSE) on a daily scale, compared to 10 daily. tation of fows and the percentage of time that the fows Probably this was due to a reduction in sample size at a larger equal or surpass each other. Satellite precipitation datasets time scale as compared to a smaller time scale. All SPPs did were evaluated through the FDC, in which the observed not show good agreement with rain gauge data at the daily Advances in Meteorology 7 5000 5000 4000 4000 Y = 0.9376X + 69.259 Y = 0.8308X + 85.058 R = 0.9921 R = 0.991 3000 3000 2000 2000 1000 1000 0 0 0 1000 2000 3000 4000 5000 0 1000 2000 3000 4000 5000 CAR Average of 3 stations (mm) CAR Average of 3 stations (mm) (a) (b) 3000 7000 Y = 1.6996 -157.9 Y = 0.628X + 63.17 5000 R = 0.989 2000 R = 0.9854 0 0 0 1000 2000 3000 4000 0 1000 2000 3000 4000 CAR Average of 3 stations (mm) CAR Average of 3 stations (mm) (c) (d) Figure 2: Consistency assessment of rain gauge data in the study area based on DMA. (a) Gujrat (GRT) Gauge Station. (b) Sargodha (SGD) Gauge Station. (c) Jhang (JNG) Gauge Station. (d) Sialkot (SKT) Gauge Station. Table 2: Evaluation of the used SPPs based on mean values of categorical and continuous metrics at daily and 10-daily time scales. Categorical metrics Continuous metrics Assessment scale Satellite product POD FAR CSI BIAS (mm) RMSE (mm) CC GSMaP 0.63 0.63 0.29 0.13 8.32 0.3 Daily scale CHIRPS 0.31 0.74 0.17 0.44 9.24 0.21 IMERG 0.49 0.68 0.24 0.08 9.57 0.25 GSMaP 0.97 0.14 0.83 1.3 23.75 0.71 10-daily scale CHIRPS 0.86 0.18 0.72 0.99 27.55 0.62 IMERG 0.97 0.16 0.82 4.42 23.35 0.74 a similar temporal pattern in monthly estimates from Jan- scale. Te correlation coefcient was low at 0.21–0.30 at the daily scale, while it was high at 0.71–0.74 at the 10-daily scale, uary to December 2015. All SPPs signifcantly overestimated showing better performance of SPPs at a larger time scale, precipitation for July through September 2020. identical with categorical metrics. Overall, the statistical On an annual time scale, a comparison was made be- performance of SPPs was lower on a daily scale and higher tween the rain gauge values and the used SPP values shown on a 10-daily scale. in Figure 4. Te annual average precipitation in the study area, as estimated from observations from 2015 to 2020 from four gauging stations, was 691 mm/year. It has been ob- 4.2.2. Monthly and Annual Scale. Figure 3 shows a monthly served that the selected satellite-based precipitation products overestimated the annual precipitation amounts. IMERG comparison of IMERG, GSMaP, and CHIRPS precipitation observations with the reference data for the entire study and GSMaP showed overestimations of 23.47% and 7.17%, respectively, while CHIRPS showed an overestimation of period (January 2015 to December 2020). Te GSMaP precipitation product represented the best monthly pre- 1.08% with reference to rain gauge values. cipitation temporal pattern. However, both IMERG and Te result showed some diference in estimating pre- CHIRPS were also capable of representing the temporal cipitation magnitudes by the IMERG products over the variability of observed precipitation over the study area, Chenab River basin of Pakistan, but the performance of albeit with notable overestimation. In July and August of CHIRPS and GSMaP encourages the utilization of SPPs in 2016, 2017, and 2018, all precipitation data sources (gauges, the study area at an annual time scale. Several researchers IMERG, GSMaP, and CHIRPS) revealed increased pre- have also reported identical fndings in diferent regions of cipitation magnitude. Almost all data sources exhibited the world. CAR JNG station (mm) CAR GRT station (mm) CAR SKT station (mm) CAR SGD station (mm) 8 Advances in Meteorology Jan-15 May-16 Oct-17 Feb-19 Jun-20 Gauge GSMaP IMERG CHIRPS Figure 3: Comparison among precipitation estimates by the PMD gauges and the three SPPs at the monthly time scale in the study area. varied from 0.21, 27, and 0.29 for CHIRPS, IMERG, and GSMaP, respectively. Intercomparing between SPP revealed that GSMaP depicted higher values of CC. Te statistical performance in terms of BIAS, RMSE, and CC of the selected products revealed that the overall efciency of GSMaP has remained higher than the other two products in the mon- soon season. 4.3.2. Winter Western Disturbance Season. Figure 6 shows the performance of SPPs towards estimation of precipitation during winter due to the western disturbance season (westerly waves) for the entire study period based on daily precipitation data. Te event detection capability revealed that the POD of GSMaP was higher and better than the other Gauge CHIRPS GSMaP IMERG two products. Te CHIRPS underperformed in terms of POD values. In the case of FAR, GSMaP outerperformed Figure 4: Comparison among average annual precipitation esti- than IMERG and CHIRPS. In the case of CSI, all selected mated by the PMD gauges and the three SPPs in the study area. SPPs revealed better performance during western distur- bances. In the case of bias value, IMERG overestimated the 4.3. Evaluation of SPPs at the Seasonal Scale precipitation, while CHIRPS showed excellent performance. While considering the results of RMSE, an agreement was 4.3.1. Summer Monsoon Season. Figure 5 shows the per- observed between the median values of CHIRPS and formance of SPPs through box plots during the monsoon GSMaP. Te box plot results showed the RMSE values season for the study period. Te categorical detection indices ranged from 2.5 to 7.5 mm/day for the selected satellite revealed that POD results were found in the range of 0.26 to products, and higher values were produced by the IMERG. A 0.85. Te precipitation detection capability of GSMaP was strong agreement between CHIRPS and IMERG was ob- better with a POD of 0.75, followed by IMERG and CHIRPS served for the CC results. Intercomparison revealed that with values of 0.56 and 0.34, respectively. Notably, all SPPs SPPs showed comparatively better statistical performance revealed large FAR values during the monsoon season. Te during western disturbance than monsoon season. performance of the selected SPPs was lower in terms of CSI. Conclusively, the statistical performance of GSMaP is Inter comparing the results of CSI revealed the better better than other SPPs, as also reported in [22, 53, 54], in performance of GSMaP for the monsoon period. In the case other regions of the world. of bias, an agreement was observed between CHIRPS and IMERG, while GSMaP revealed better performance. Te results of RMSE indicated that the frst quartile of daily data 4.4. Parametric Sensitivity Assessment. Figure 7 shows the was found in the range of 11 to 20 mm per day for the sensitivity analysis of the IFAS model for surface, aquifer, selected three products. Te values of the second and third and river course tank parameters based on the mean ab- solute sensitivity index (MASI). In the case of surface tank quartiles of RMSE were observed at about 11 to 17 mm/day. For the CC results, it was observed that the median values parameters, the surface tank height (HFMND) and fnal Average Annual Precipitation (mm/year) Precipitation (mm/month) Advances in Meteorology 9 0.85 0.80 0.75 0.75 0.65 0.70 0.55 0.65 0.45 0.60 0.35 0.25 0.55 CHIRPS CHIRPS GSMAP GSMAP POD FAR (a) (b) 0.40 2.00 0.35 1.00 0.30 0.25 0.00 0.20 0.15 -1.00 CHIRPS CHIRPS GSMAP GSMAP CSI BIAS (c) (d) 20.0 0.35 18.0 0.30 16.0 0.25 14.0 0.20 12.0 10.0 0.15 CHIRPS CHIRPS GSMAP GSMAP RMSE CC (e) (f ) Figure 5: Statistical (continuous and categorical) assessment results of the used SPPs at a seasonal scale during the summer monsoon period. (a) POD. (b) FAR. (c) CSI. (d) BIAS. (e) RMSE. (f ) CC. 0.90 0.65 0.85 0.55 0.80 0.45 0.75 0.35 0.70 0.25 0.65 0.60 0.15 CHIRPS CHIRPS GSMAP GSMAP FAR POD (a) (b) 0.60 0.30 0.40 0.20 0.20 0.10 0.00 0.00 CHIRPS CHIRPS GSMAP GSMAP CSI BIAS (c) (d) Figure 6: Continued. CSI POD CSI POD RMSE BIAS CC FAR BIAS FAR 10 Advances in Meteorology 8.0 0.35 6.0 0.25 4.0 0.15 2.0 0.05 CHIRPS CHIRPS GSMAP GSMAP RMSE CC (e) (f ) Figure 6: Statistical (continuous and categorical) assessment results of the used SPPs at seasonal scale during the winter westerly wave period. (a) POD. (b) FAR. (c) CSI. (d) BIAS. (e) RMSE. (f ) CC. infltration capacity of soil (SKF) are the most sensitive and-error method. Another surface tank parameter parameters, with a mean absolute sensitivity index (MASI) (FALFX) was tuned from 0 to 1 to control the subsurface of 8 and 3.5, respectively. It was observed that other surface fow to calibrate the model. Te values of FALFX parameters tank parameters, i.e., SNF, HFOD, and HFID, do not sig- were subsequently decreased to adjust the hydrograph in the calibration process. nifcantly impact the calibration of the model and show a lower mean absolute sensitivity index. Similarly, the Tree diferent calibration scenarios were established to investigate the capacity of selected SPPs to calibrate the IFAS evaluation of aquifer tank parameters indicated that the parameters, i.e., the efect of storage height to generate base model and to examine their efectiveness for diferent ap- plications. In the frst scenario, the model was calibrated fow (AGD) and the initial value used for calculation (HIGD), are the two most sensitive parameters, with MASI utilizing the CHIRPS satellite precipitation data, and then values of 452 and 99, respectively. Te aquifer parameter the validation process was completed using the GSMaP and HIGD depicted a direct relationship with the change in IMERG. For the second scenario, the IFAS model was simulation results of the hydrograph, while HCGD and calibrated utilizing the GSMaP, and the model was validated AUD do not infuence the calibration of the model. by using CHIRPS and IMERG for evaluation. In the third Meanwhile, the analysis of the river course tank pa- scenario, the IMERG precipitation dataset was utilized to calibrate the IFAS model and then validated against GSMaP rameters indicated that the parameters related to the co- efcients of a crosssection of a river, i.e., RLCOF and RBS, and CHIRPS. Calibration and validation of the IFAS model were evaluated using the model performance indicators, i.e., were sensitive to the simulated hydrograph. Te aquifer tank parameters are more sensitive than any other tank param- NSE, R2, and PBIAS. Te model’s performance on each scenario and comparison among the performance of the eters also reported by [55], and river tank parameters played signifcantly less in calibration. Terefore, it is suggested to three scenarios are presented in Table 3. introduce the option of actual groundwater conditions in the For the frst calibration scenario, the statistical perfor- IFAS model for the target area. mance indicators R , NSE, and PBIAS were 0.89, 0.86, and Te IFAS model was calibrated and validated for the −0.16, respectively. Te intercomparison results of model river Chenab at Qadir Abad barrage outlet for the monsoon validation for this scenario revealed the better performance periods of 2015 and 2017, respectively, utilizing the selected of the GSMaP dataset with R , NSE, and PBIAS values of SPPs. In the calibration process, initially the default pa- 0.85, 0.83, and 0.16, respectively. Te IMERG and CHIRPS rameters for surface and aquifer tanks were used to run the datasets showed slightly lower performance during the model. Te surface parameters were based on digital land model validation process. From the graphical presentation cover data, while the aquifer parameters were based on soil of scenario 1 in Figure 8, some variations in simulating low type data for the selected basin. Te parameters were tuned and high fows were observed by the SPPs. and optimized with the trial-and-error technique to bring For the second calibration scenario, the statistical per- them into sound agreement with the observed fow data. Te formance indicators (R , NSE, and PBIAS) were 0.97, 0.96, critical parameters considered for a successful calibration of and −0.03, respectively. According to the calibration criteria, the model are the coefcient of base fow regulation (AGD) this scenario displayed excellent performance, demon- for the aquifer tank, the surface tank height (HFMND), the strating that GSMaP precipitation data resulted in a robust fnal infltration capacity of the soil (SKF), and the initial and trustworthy testing model with utility and accuracy that height of infltration (HFOD) for the surface tank. Since could be used to check and compare the results produced HFOD is a surface parameter, it signifcantly infuences the from the IMERG and CHIRPS precipitation models. adjustment of the peak of the hydrograph. Due to the surface Comparison of model validation results revealed that the tank’s fve distinct feature classes, successful peak calibration GSMaP dataset outperformed the other datasets, with R , requires fne-tuning of the land cover parameter. Land cover NSE, and PBIAS values of 0.9, 0.89, and 0.14, respectively. A classes from the IFAS graphical module were used to cali- strong agreement was observed between IMERG and brate the model, which was then fne-tuned using a trial- CHIRPS-based simulated fows. Te ability of the GSMaP RMSE CC Advances in Meteorology 11 0 0 HFMND SKF HFMXD FALFX SNF HFOD HIFD AGD HIGD HCGD AUD (a) (b) RLCOF RBS RNS RRID RBW RGWD RHW RHS RBH RBET (c) Figure 7: Sensitivity assessments among various parameters of the IFAS model. (a) Surface parameters. (b) Aquifer parameters. (c) River course parameters. Table 3: Performance evaluation comparison among the used calibration and validation scenarios. Scenario 1 Scenario 2 Scenario 3 Year CHIRPS rainfall model GSMaP rainfall model IMERG rainfall model 2 2 2 SPP NSE PBIAS R SPP NSE PBIAS R SPP NSE PBIAS R Calibrated 2015 CHIRPS 0.86 −0.16 0.89 GSMaP 0.96 −0.03 0.97 IMERG 0.91 −0.11 0.92 Validated 2017 CHIRPS 0.78 0.17 0.83 GSMaP 0.80 0.25 0.89 CHIRPS 0.80 0.17 0.85 Validated 2017 GSMaP 0.80 0.16 0.84 IMERG 0.78 0.16 0.84 GSMaP 0.82 0.16 0.86 Validated 2017 IMERG 0.79 0.15 0.82 CHIRPS 0.77 0.26 0.88 IMERG 0.81 0.16 0.85 6000 6000 4000 4000 2000 2000 0 0 25-Jun-15 14-Aug-15 3-Oct-15 22-Nov-15 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME Time Observed-Flow Observed Flow CHIRPS-Flow CHIRPS Flow (a) (b) Figure 8: Continued. MASI Discharge (m3/s) MASI Discharge (m3/s) MASI 12 Advances in Meteorology 6000 6000 4000 4000 2000 2000 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 Time Time Observed Flow Observed Flow GSMaP Flow IMERG Flow (c) (d) Figure 8: Graphical representations of calibration and validation scenario 1. (a) Calibration-CHIRP. (b) Validation CHIRPS. (c) Validation GSMaP. (d) Validation IMERG. 6000 6000 2000 2000 0 0 25-Jun-15 14-Aug-15 3-Oct-15 22-Nov-15 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME TIME Observed Flow Observed Flow GSMaP Flow CHIRPS Flow (a) (b) 6000 6000 4000 4000 2000 2000 0 0 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME TIME Observed Flow Observed Flow IMERG Flow GSMaP Flow (c) (d) Figure 9: Graphical representations of calibration and validation scenario 2. (a) Calibration-GSMaP. (b) Validation CHIRPS. (c) Validation IMERG. (d) Validation GSMaP. dataset to optimize the parameters and calibrate the model For the third calibration scenario, the statistical per- formance indicators R was better when compared with the CHIRPS precipitation , NSE, and PBIAS for model cali- model. A graphical presentation of scenario 2 is shown in bration were observed at 0.92, 0.91, and −0.11, respectively, Figure 9. It depicted a trend identical to scenario 1, but a bit which exhibited excellent performance of this model improved simulation was observed in predicting low and according to the calibration rating described by [52]. Te R high fows. of the GSMaP, IMERG, and CHIRPS datasets were 0.87, Discharge (m3/s) 3 3 Discharge (m /s) Discharge (m /s) 3 3 Discharge (m /s) Discharge (m /s) Discharge (m3/s) Advances in Meteorology 13 6000 6000 4000 4000 2000 2000 0 0 25-Jun-15 14-Aug-15 3-Oct-15 22-Nov-15 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME Time Observed-Flow Observed Flow IMERG-Flow CHIRPS Flow (a) (b) 6000.0 6000.0 4000.0 4000.0 2000.0 2000.0 0.0 0.0 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 Time TIME Observed Flow Observed Flow GSMaP Flow IMERG Flow (c) (d) Figure 10: Graphical representations of calibration and validation scenario 3. (a) Calibration-IMERG. (b) Validation CHIRPS. (c) Validation GSMaP. (d) Validation CHIRPS. 0 0 0 10 20 30 40 50 60 70 80 90 100 Exceeded Time (%) Exceedance Time (%) Observed GSMaP Observed GSMaP CHIRPS IMERG CHIRPS IMERG Figure 12: Assessment of the used SPPs in the term Figure 11: Assessment of the used SPPs in terms of hydraulic signature through FDC. dependable fows. values of 0.97, 0.96, and −0.03, respectively. For PBIAS 0.85, and 0.84, respectively, which depicts better GSMaP product performance for the third scenario also. A graphical evaluation, GSMaP produced a lower value of −0.03, compared to the other two products, CHIRPS and IMERG presentation of this scenario, Figure 10, shows identical simulated moderately higher values of −0.11 and −0.16, performance in forecasting low and high fows. It was noted that GSMaP outerperformed in terms of respectively. IMERG and CHIRPS were rated in the second and third positions for performance evaluation of model parameter optimization and fne-tuning of the IFAS model during the calibration procedure with R , NSE, and PBIAS calibration. In the case of the application and validation of Discharge (m /s) Discharge (m3/s) Discharge (m /s) Discharge (m3/s) Discharge (m3/s) Discharge (m /s) 14 Advances in Meteorology the hydrological model for the high monsoon food of 2017, (2) Statistical evaluation revealed that the efcacy of GSMaP outperformed in each scenario. Overall, the hy- GSMaP has been better, while CHIRPS showed more biases. Te performance of SPPs improved at 10- drological performance of GSMaP was more satisfactory than that of IMERG and CHIRPS, IMERG was ranked daily and monthly time scales than at the daily time second, while the CHIRPS exhibited a lower performance. scale. Higher values of uncertainties (bias and RMSE) were observed during the monsoon season than during the western disturbances. Missed and 4.5. Hydrological Signature-Based Assessment. Te hydro- false alarms were the main errors associated with logical performance of all selected satellite precipitation SPPs due to spatial mobility and the sudden bursting datasets was evaluated through FDC, in which daily ob- of clouds, specifcally during the monsoon season. served stream fows were taken as the baseline and the (3) Te stativity analysis revealed that the aquifer tank variation in simulated fow was determined. Te FDC results parameters were found to be the most sensitive. Te displayed that the selected satellite dataset has relatively base fow coefcient (AGD) was found to be the most inferior performance in capturing extreme fooding con- sensitive parameter in calibrating the IFAS model to ditions. While considering the medium fow, all datasets simulate fows using SPPs. showed excellent performance in the range of 20% to 40% (4) Te model calibration and validation scenarios in- exceedance time, as displayed in Figure 11. dicated that the GSMaP precipitation dataset has Similarly, these precipitation datasets do not yield sat- better capability to calibrate and validate the model isfactory results for the simulation of low fows. For all compared to IMERG and CHIRPS, with the highest precipitation datasets at the catchment outlet stage, an R , NSE and lower PBIAS values. It was also observed exceedance fow analysis was used to estimate the de- that the SPPs have relatively poor performance in pendable fow exceedance of Q5, Q10, Q25, Q50, and Q70. capturing extreme fooding events. While consid- Q5 denotes a fow that exceeds 5% of the analysis time, and so forth. Extreme food events are revealed by 5% and 10% ering the medium fows, in the range of 20%–40% exceedance time, all datasets showed excellent stream fows, while 50% dependability designates the me- dian fow, 70% dependable fow resembles the water performance. availability for agriculture, and higher dependable fows Findings of this study suggested that direct utilizations of correspond to the water availability for domestic supplies. satellite-based precipitation products were not promising at Te performances of SPPs to generate high, medium, and daily scales and bias correction is recommended. For food low fows was analyzed through these dependable fows. It modeling, the hydrological IFAS model should be calibrated was found that all SPPs data sets’ performance was lower, based on peak fow, considering the combination of sta- corresponding to Q5 and Q70. Te SPPs datasets can tistical and error indicators. Further studies may be carried generate medium fow in the range of Q25-Q50 Figure 12. out to assess the efectiveness of the available sensitivity analysis techniques in this study area. 5. Conclusion Acronyms Te present study evaluated three high-resolution multi- satellite precipitation estimation products statistically and SPPs: Satellite precipitation products hydrologically in the Chenab River catchment. Te con- IMERG: Integrated Multi-satellite Retrieval for GPM sistency of rain gauge data observed by PMD was examined GSMaP: Global Satellite Mapping of Precipitation by double mass analysis. Numerous statistical indicators CHIRPS: Climate Hazards Group Infrared Precipitation were applied at daily, monthly, and seasonal scales to detect with Station and quantify errors associated with these products. Tree DMC: Double mass curve diferent calibration scenarios were established for the hy- POD: Probability of detection drological assessment to analyze the satellite precipitation CSI: Critical success index datasets. A sensitivity analysis was performed to study the FAR: False alarm ratio most sensitive parameters of the distributed IFAS model. CC: Correlation coefcient Te hydrological signature was used to assess the potential of RMSE: Root mean square error satellite products to generate high, medium, and low fows. B: Bias Te existence of about 62 percent of the catchment area in IFAS: Integrated Flood Analysis System Indian-held Kashmir and the occurrence of only four FDC: Flow duration curve gauging stations in the rest of the catchment area are the R : Coefcient of determination major limitations of the study towards hydrological and AGD: Base fow coefcient statistical assessment of the satellite products in the study NSE: Nash−Sutclife efciency area, respectively. From the fndings of this study, it was PBIAS: Percentage bias observed as follows: PMD: Pakistan Metrological Department (1) PMD rain gauge-based precipitation data are con- WMO: World Meteorological Organization sistent and can be used for the assessment of satellite- FFD: Flood forecasting division based precipitation datasets. JAXA: Japan Aerospace Exploration Agency Advances in Meteorology 15 cipondoh, tangerang Indonesia,” Journal of Environment and NASA: National Aeronautics and Space Administration Earth Science, vol. 9, no. 1, pp. 52–61, 2019. PMW: Passive microwave [9] M. Aslam, “Flood management current state, challenges and IR: Infrared prospects in Pakistan: a review,” Mehran University Research CPC: Climate prediction center Journal of Engineering and Technology, vol. 37, no. 2, GPM: Global precipitation measurement pp. 297–314, 2018. CCD: Cold cloud duration [10] M. S. Shrestha, M. R. Khan, N. Wagle, Z. Ahmad Babar, DEM: Digital elevation model V. R. Khadgi, and S. Sultan, Review of Hydrometeorological ISCGM: International Steering Committee for Global Monitoring and Forecasting System for Floods in the Indus Mapping Basin in Pakistan, Elsevier Inc, Amsterdam, Netherlands, DSMW: Digital soil map of the world [11] S. Sorooshian, A. Aghakouchak, P. Arkin et al., “Advanced FAO: Food and Agriculture Organization concepts on remote sensing of precipitation at multiple PWRI- Public Works Research Institute distributed scales,” Bulletin of the American Meteorological Society, DHM: hydrological model vol. 92, no. 10, pp. 1353–1357, 2011. IWRM: Integrated water resources management [12] A. U. Rehman, F. Chishtie, W. A. Qazi, S. Ghufar, I. Fatima, MPIs: Model performance indicators. and K. Fatima, “Evaluation of three-hourly TMPA rainfall products using telemetric rain gauge observations at lai nullah Data Availability basin in islamabad, Pakistan,” Remote Sensing, vol. 10, no. 12, p. 2040, 2018. Te data used to support the fndings of this study are [13] Y. Zhang, A. Sun, H. Sun et al., “Error adjustment of TMPA available from the corresponding author upon request. satellite precipitation estimates and assessment of their hy- drological utility in the middle and upper yangtze River Basin, Conflicts of Interest China,” Atmospheric Research, vol. 216, pp. 52–64, 2019. [14] C. Miao, H. Ashouri, K. L. Hsu, S. Sorooshian, and Q. Duan, Te authors declare that they have no conficts of interest. “Evaluation of the PERSIANN-CDR daily rainfall estimates in capturing the behavior of extreme precipitation events over Acknowledgments China,” Journal of Hydrometeorology, vol. 16, no. 3, pp. 1387–1396, 2015. Te authors are grateful to the Center of Excellence in Water [15] B. Yong, Y. Hong, L. L. Ren et al., “Assessment of evolving Resource Engineering for support in conducting this study. TRMM-based multisatellite real-time precipitation estima- Te authors are also thankful to PMD/FFD and satellite tion methods and their impacts on hydrologic prediction in organizations for the data used in the study. a high latitude basin,” Journal of Geophysical Research: At- mospheres, vol. 117, no. 9, 2012. References [16] D. B. Michelson, “Systematic correction of precipitation gauge observations using analyzed meteorological variables,” Jour- [1] M. T. Mendoza and R. Schwarze, “Sequential disaster fo- nal of Hydrology, vol. 290, no. 3-4, pp. 161–177, 2004. rensics: a case study on direct and socio-economic impacts,” [17] P. S. Roy, M. D. Behera, and S. K. Srivastav, “Satellite remote Sustainability, vol. 11, no. 21, pp. 1–20, 2019. sensing: sensors, applications and techniques,” Proceedings of [2] A. Mishra, A. Alnahit, and B. Campbell, “Impact of land uses, the National Academy of Sciences India Section A - Physical drought, food, wildfre, and cascading events on water quality Sciences, vol. 87, no. 4, pp. 465–472, 2017. and microbial communities: a review and analysis,” Journal of [18] K. J. Beven and H. L. Cloke, “Comment on ‘Hyperresolution Hydrology, vol. 596, Article ID 125707, 2021. Global Land Surface Modeling: meeting a Grand Challenge [3] F. Dottori, W. Szewczyk, J. C. Ciscar et al., “Increased human for Monitoring Earth’s Terrestrial Water’ by Eric F. Wood et and economic losses from river fooding with anthropogenic Al,” Water Resources Research, vol. 48, no. 1, pp. 1–10, 2012. warming,” Nature Climate Change, vol. 8, no. 9, pp. 781–786, [19] K. Yan, G. Di Baldassarre, D. P. Solomatine, and G. J. P. Schumann, “A review of low-cost space-borne data for [4] A. Mishra, S. Mukherjee, B. Merz et al., “An overview of food food modelling: topography, food extent and water level,” concepts, challenges, and future directions,” Journal of Hy- Hydrological Processes, vol. 29, no. 15, pp. 3368–3387, 2015. drologic Engineering, vol. 27, no. 6, pp. 1–30, 2022. [20] Z. N. Musa, I. Popescu, and A. Mynett, “A review of appli- [5] S. Janjua, I. Hassan, S. Muhammad, S. Ahmed, and A. Ahmed, cations of satellite SAR, optical, altimetry and DEM data for “Water management in Pakistan’s Indus basin: challenges and surface water modelling, mapping and parameter estimation,” opportunities,” Water Policy, vol. 23, no. 6, pp. 1329–1343, Hydrology and Earth System Sciences, vol. 19, no. 9, pp. 3755–3769, 2015. [6] A. R. Iqbal, “Environmental issues of Indus River Basin: an [21] M. Masood, A. S. Shakir, A. H. Azhar, G. Nabi, and analysis,” Institute for Strategic Studies, Research & Analysis H. U. Rehman, “Assessment of real time, multi-satellite (ISSRA), vol. 5, no. 1, National Defence University, Islamabad, precipitation products under diverse climatic and topo- Pakistan, 2013. graphic conditions,” Asia-Pacifc Journal of Atmospheric [7] Federal Communications Commission, “Government of Sciences, vol. 56, no. 4, pp. 577–591, 2019. pakistan ministry of water resources,” Annual Flood Report [22] J. Shi, F. Yuan, C. Shi et al., “Statistical evaluation of the latest 2017, pp. 1–118, Federal Communications Commission, Washington, DC, USA, 2017. GPM-era IMERG and GSMaP satellite precipitation products in the yellow river source region,” Water (Switzerland), [8] J. Pancawati, S. Nurisjah, and B. Pramudya, “Application of tank model for ungauged reservoir management: a case of situ vol. 12, no. 4, pp. 1006–1023, 2020. 16 Advances in Meteorology [23] N. Saddique, M. Muzammil, I. Jahangir et al., “Hydrological the issen watershed in the souss basin, Morocco,” European evaluation of 14 satellite-based, gauge-based and reanalysis Scientifc Journal, vol. 12, no. 18, p. 347, 2016. [38] A. Sugiura, S. Fujioka, S. Nabesaka, M. Tsuda, and Y. Iwami, precipitation products in a data-scarce mountainous catch- “Development of a food forecasting system on the upper ment,” Hydrological Sciences Journal, vol. 67, no. 3, Indus catchment using IFAS,” Journal of Flood Risk Man- pp. 436–450, 2022. agement, vol. 9, no. 3, pp. 265–277, 2016. [24] G. Villarini, W. F. Krajewski, and J. A. Smith, “New paradigm [39] A. Shahzad, H. F. Gabriel, S. Haider, A. Mubeen, and for statistical validation of satellite precipitation estimates: M. J. Siddiqui, “Development of a food forecasting system application to a large sample of the TMPA 0.25 3-hourly using IFAS: a case study of scarcely gauged Jhelum and estimates over Oklahoma,” Journal of Geophysical Research, Chenab River basins,” Arabian Journal of Geosciences, vol. 11, vol. 114, no. 12, pp. 12106–12112, 2009. no. 14, p. 383, 2018. [25] B. Nijssen and D. P. Lettenmaier, “Efect of precipitation [40] A. Aziz and S. Tanaka, “Regional parameterization and ap- sampling error on simulated hydrological fuxes and states: plicability of integrated food analysis system (IFAS) for food anticipating the global precipitation measurement satellites,” forecasting of upper-middle Indus river classifcation of hy- Journal of Geophysical Research: Atmospheres, vol. 109, no. 2, drological models,” Pakistan Journal of Meterorology, vol. 8, pp. 1–15, 2004. no. 15, pp. 21–38, 2010. [26] J. R. McCollum, R. Witold, R. R. Krajewski, and M. B. Ba, [41] M. Miyamoto, M. Ono, S. Nabesaka, T. Okazumi, and “Evaluation of biases of satellite rainfall estimation algorithms Y. Iwami, “Applicability of a food forecasting method uti- over the continental United States,” Journal of Applied Me- lizing global satellite information to an insufciently-gauged teorology, vol. 41, no. 11, pp. 1065–1080, 2002. River Basin: A case of a River Basin in the Philippines,” in [27] F. L. Conti, K. L. Hsu, L. V. Noto, and S. Sorooshian, Proceedings of the 11th International Conference on Hydro- “Evaluation and comparison of satellite precipitation esti- informatics, New York, NY, USA, August 2014. mates with reference to a local area in the mediterranean sea,” [42] H. Ishak, M. D. Nor, L. M. Sidek et al., “Flood forecasting and Atmospheric Research, vol. 138, pp. 189–204, 2014. early warning system for dungun river basin,” IOP Conference [28] M. Masood, G. Nabi, M. Babur, A. H. Azhar, and M. Kaleem Series: Earth and Environmental Science, vol. 16, no. 1, 2013. Ullah, “Disintegration of uncertainties associated with [43] R. Awan, “Pakistan food Management.Pdf (Application/Pdf real-time multi-satellite precipitation products in diverse object),” Mainstream, vol. 1, pp. 1–4, 2003. topographic and climatic area in Pakistan,” Journal of [44] M. Riaz, A. Aziz, and S. Hussain, “Flood forecasting of an Mountain Science, vol. 18, no. 3, pp. 716–734, 2021. ungauged trans-boundary Chenab River Basin using dis- [29] Y. Tuo, Z. Duan, M. Disse, and G. Chiogna, “Evaluation of tributed hydrological model integrated food analysis system precipitation input for swat modeling in alpine catchment: (IFAS),” Pakistan Journal of Meteorology, vol. 13, no. 26, a case study in the adige River Basin (Italy),” Science of the pp. 51–62, 2017. Total Environment, vol. 573, pp. 66–82, 2016. [45] G. J. Hufman, D. T. Bolvin, D. Braithwaite et al., Algorithm [30] T. M. Sharannya, N. Al-Ansari, S. Deb Barma, A. Mahesha, Teoretical Basis Document (ATBD) Version 5.2 for the NASA and A. Mahesha, “Evaluation of satellite precipitation prod- Global Precipitation Measurement (GPM) Integrated Multi- ucts in simulating streamfow in a humid tropical catchment satellitE Retrievals for GPM (I-MERG), GPM Project, of India using a semi-distributed hydrological model,” Water Greenbelt, MD, USA, 2019. (Switzerland), vol. 12, no. 9, p. 2400, 2020. [46] S. Sharma, Y. Chen, X. Zhou et al., “Evaluation of GPM-era [31] M. Umer, H. F. Gabriel, S. Haider, A. Nusrat, M. Shahid, and satellite precipitation products on the southern slopes of the M. Umer, “Application of precipitation products for food central himalayas against rain gauge data,” Remote Sensing, modeling of transboundary River Basin: a case study of vol. 12, no. 11, p. 1836, 2020. Jhelum basin,” Teoretical and Applied Climatology, vol. 143, [47] C. Funk, P. Peterson, M. Landsfeld et al., “Te climate hazards no. 3-4, pp. 989–1004, 2021. infrared precipitation with stations - a new environmental [32] R. H. McCuen, “Te role of sensitivity analysis in hydrologic record for monitoring extremes,” Scientifc Data, vol. 2, modeling,” Journal of Hydrology, vol. 18, no. 1, pp. 37–53, p. 150066, 2015. [48] M. Masood, A. S. Shakir, and H. U. Rehman, “Assessment of [33] K. K. Yilmaz, H. V. Gupta, and T. Wagener, “A process-based GPM based integrated multi-satellite retrievals under diverse diagnostic approach to model evaluation: application to the climatic and topographic conditions in Pakistan,” vol. 23, NWS distributed hydrologic model,” Water Resources Re- pp. 17–29, 2018. search, vol. 44, no. 9, pp. 1–18, 2008. [49] T. Sugiura, K. Fukami, and H. Inomata, “Development of [34] K. Sawicz, T. Wagener, M. Sivapalan, P. A. Troch, and integrated food analysis system (IFAS) and its applications,” G. Carrillo, “Catchment classifcation: empirical analysis of in Proceedings of the World Environmental and Water Re- hydrologic similarity based on catchment function in the sources Congress, vol. 316, Pittsburgh, PA, USA, May 2008. eastern USA,” Hydrology and Earth System Sciences, vol. 15, [50] S. Sutikno, Y. L. Handayani, M. Fauzi, Fitriani, and A. Kurnia, no. 9, pp. 2895–2911, 2011. “Hydrologic modelling using TRMM-based rainfall products [35] I. K. Westerberg and H. K. McMillan, “Uncertainty in hy- for food analysis,” MATEC Web of Conferences, vol. 101, drological signatures,” Hydrology and Earth System Sciences, pp. 2–6, 2017. vol. 19, no. 9, pp. 3951–3968, 2015. [51] A. Aziz and S. Tanaka, “Regional parameterization and ap- [36] F. Mendonça, R. P. Oliveira, and F. F. Mauad, “Lumped versus plicability of integrated food analysis system (IFAS) for food distributed hydrological modeling of the jacare-guaçu ´ basin, forecasting of upper-middle Indus River classifcation of Brazil,” Journal of Environmental Engineering, vol. 144, no. 8, hydrological models,” Pakistan Journal of Meterorology, vol. 8, Article ID 04018056, 2018. no. 15, pp. 21–38, 2010. [37] H. Brirhet and L. Benaabidate, “Comparison of two hydro- [52] D. N. Moriasi, D. J. G. Arnold, M. W. Van Liew, R. L. Bingner, logical models (lumped and distributed) over A pilot area of R. D. Harmel, and T. L. Veith, “Model evaluation guidelines Advances in Meteorology 17 for systematic quantifcation of accuracy in watershed sim- ulations,” Transactions of the ASABE, vol. 50, no. 3, pp. 885–900, 2007. [53] G. Tang, M. P. Clark, S. M. Papalexiou, Z. Ma, and Y. Hong, “Have satellite precipitation products improved over last two decades? A comprehensive comparison of GPM IMERG with nine satellite and reanalysis datasets,” Remote Sensing of Environment, vol. 240, Article ID 111697, 2020. [54] S. Ning, S. Fan, P. Udmale, J. Jin, B. Raj Tapa, and H. Ishidaira, “Error analysis and evaluation of the latest GSMap and IMERG precipitation products over eastern China,” Advances in Meteorology, vol. 2017, Article ID 1803492, 16 pages, 2017. [55] A. Nusrat, H. F. Gabriel, S. Haider, and M. Siddique, “Sen- sitivity analysis and optimization of land use/cover and aquifer parameters for improved calibration of hydrological model,” Mehran University Research Journal of Engineering and Technology, vol. 41, no. 2, pp. 21–34, 2022. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advances in Meteorology Hindawi Publishing Corporation

Evaluation of Satellite Precipitation Products for Estimation of Floods in Data-Scarce Environment

Loading next page...
 
/lp/hindawi-publishing-corporation/evaluation-of-satellite-precipitation-products-for-estimation-of-ZYzEpUWeCl

References (56)

Publisher
Hindawi Publishing Corporation
ISSN
1687-9309
eISSN
1687-9317
DOI
10.1155/2023/1685720
Publisher site
See Article on Publisher Site

Abstract

Hindawi Advances in Meteorology Volume 2023, Article ID 1685720, 17 pages https://doi.org/10.1155/2023/1685720 Research Article Evaluation of Satellite Precipitation Products for Estimation of Floods in Data-Scarce Environment 1 2 1 1 Muhammad Masood , Muhammad Naveed , Mudassar Iqbal , Ghulam Nabi , 1 1 1 Hafiz Muhammad Kashif , Muhammad Jawad , and Ahmad Mujtaba Centre of Excellence in Water Resources Engineering, University of Engineering & Technology GT-Road, Lahore 54890, Pakistan Department of Civil Engineering, University of Engineering & Technology GT-Road, Lahore 54890, Pakistan Correspondence should be addressed to Muhammad Masood; chmasoud@gmail.com Received 22 November 2022; Revised 22 March 2023; Accepted 11 April 2023; Published 3 May 2023 Academic Editor: Tomeu Rigo Copyright © 2023 Muhammad Masood et al. Tis is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Utilization of satellite precipitation products (SPPs) for reliable food modeling has become a necessity due to the scarcity of conventional gauging systems. Tree high-resolution SPPs, i.e., Integrated Multi-satellite Retrieval for GPM (IMERG), Global Satellite Mapping of Precipitation (GSMaP), and Climate Hazards Group InfraRed Precipitation with Station (CHIRPS), data were assessed statistically and hydrologically in the sparsely gauged Chenab River basin of Pakistan. Te consistency of rain gauge data was assessed by the double mass curve (DMC). Te statistical metrics applied were probability of detection (POD), critical success index (CSI), false alarm ratio (FAR), correlation coefcient (CC), root mean square error (RMSE), and bias (B). Te hydrologic evaluation was conducted with calibration and validation scenarios for the monsoon fooding season using the Integrated Flood Analysis System (IFAS) and fow duration curve (FDC). Sensitivity analysis was conducted using ±20% cal- ibrating parameters. Te rain gauge data have been found to be consistent with the higher coefcient of determination (R ). Te mean skill scores of GSMaP were superior to those of CHIRPS and IMERG. More bias was observed during the monsoon than during western disturbances. Te most sensitive parameter was the base fow coefcient (AGD), with a high mean absolute sensitivity index value. During model calibration, good values of performance indicators, i.e., R , Nash−Sutclife efciency (NSE), and percentage bias (PBIAS), were found for the used SPPs. For validation, GSMaP performed better with comparatively higher values of R and NSE and a lower value of PBIAS. Te FDC exhibited SPPs’ excellent performance during 20% to 40% exceedance time. Pakistan’s history has emphasized the critical need for an 1. Introduction efcacious and steadfast food warning system [7]. Improved A food is a frequently recurring destructive natural hazard food modeling and remote sensing necessitate enhanced that humanity encounters [1]. Te adverse fooding con- food analysis methodology to send timely warnings to ditions have resulted in a signifcant deterioration of envi- populations and better reservoir operations [8]. Te early ronmental sustainability [2] throughout the world, wreaking warning system in Pakistan has limited capabilities. Al- havoc with costly infrastructure, food production, and though substantial improvements in food forecasting have human lives [3]. Globally, climate change and anthropogenic been made by utilizing the weather radar and telemetric activities have increased the frequency and intensity of systems in the warning system, there is still a need for many foods and exacerbated riverine food threats in several world eforts to advance the food forecasting and warning system places [4]. Pakistan is located in a highly vulnerable region to [9, 10]. climate change where foods occur almost every year since Precipitation is an important hydrological parameter the last three decades [5, 6]. Te devastating food in used for watershed management, food forecasting, and 2 Advances in Meteorology necessitated hydrological modeling. In Pakistan, limited climatological assessment [11]. It is also the most complex parameter because of its excessive spatiotemporal variations studies have been conducted to assess satellite precipitation products’ efectiveness, particularly utilizing the distributed that traditional rain gauges and radar networks cannot re- cord due to their sparsity [12]. Accurate and precise pre- IFAS model. In [38], the authors highlighted the scarcity of cipitation data with fne spatiotemporal resolution is hydrological data and the signifcance of upstream fow important for watershed management and food analysis boundary conditions when barrage operation standards are [13]. It necessitates the need for such techniques that sup- unknown in the Indus River. In [31], the authors used plements the rain gauge observations and provides excep- a lump and regional calibration approach to model the tional precipitation data to support hydrological modeling Jhelum river basin. In [39], the authors pointed out the difculty of food modeling at the confuence point of the issues [14–16]. Remote sensing satellites use refected light to detect, Chenab and Jhelum basins. In [38], the authors explored that the performance of IFAS can be improved by utilizing local collect, measure, and record the electromagnetic energy from the earth’s surface [17]. Advances in satellite remote soil texture data in the Indus River. In [38], the authors evaluated the precipitation results from diferent sources sensing have made it an excellent data source, as it can provide metrological data to support hydrological modeling for modeling the Indus River’s middle reach. Aziz [40] issues [18–20]. Moreover, precipitation data obtained from demonstrated that IFAS could be used for hydrological remote sensing have the potential to supplement the tra- modeling of the Kabul River with data scarcity. In [41], the ditional rain gauge system [21]. authors investigated that integrating satellite and gauge Recent studies have shown that the precipitation data rainfall data can enhance food forecasting in the estimated through satellite-based observations contributed Philippines-Cagayan River catchment. In [42], the authors recommended that improved satellite precipitation data be well to detecting rainfall distribution and severity in data- scarce regions [22, 23]. However, the satellite products may used to enhance food prediction in the Dungun River basin, Malaysia. have errors due to indirect estimation, sampling uncertainty, and retrieval algorithms [24–27]. Te properties of these Assessment of SPPs with a fully distributed hydrological model under diferent calibration scenarios is yet to be errors signifcantly vary in contrasting climates, storms, seasons, and altitudes [11, 28]. Terefore, it is essential to evaluated in the study area. In addition, the representation of validate the accuracy of satellite precipitation products and hydrological signatures with diferent rainfall data sets is yet suitability for a broader range of environments. Satellite to be explored. It mandates the investigation of satellite rainfall data can be validated using statistical analysis re- precipitation data sets using a distributed hydrological garding ground-based on gauge data and a proper hydro- model for diferent applications. In this study, three SPP- logical modeling framework [29]. Statistical analysis based datasets, i.e., IMERG, GSMaP, and CHIRPS, have been evaluated statistically and hydrologically in a data-scar determines the accuracy and consistency of satellite pre- cipitation data, while hydrological simulation elucidates the region, i.e., the Chenab River catchment of Pakistan. Te study utilized the IFAS model to generate streamfow for usefulness and application of the same datasets [30]. For reliable hydrological modeling, a proper calibration tech- a sparsely gauged catchment by using satellite precipitation nique, parametric sensitivity, and model capability are of datasets and derived hydrological signatures. primary consideration. A suitable calibration technique is critical because errors in model calibration and input 2. Study Area and Data Description datasets contribute to incorrect outcomes [31]. Similarly, parametric sensitivity produces only important parameters, 2.1. Study Area. Te Chenab River starts in Himachal reduces the analysis time, and contributes to modeling Pradesh, India, at the confuence of the Bhaga and Chandra calibration [32]. streams and fows across Indian-controlled Kashmir to Furthermore, the ability of the hydrological model to Pakistan [43]. Te catchment of the Chenab River covers an simulate water fow can be examined using the FDC, “a key area of about 26,000 km up to the Marala Barrage. It runof variability signature” [33]. Te FDC ofers additional embraces 97% of this catchment area in India, while only 3% details on the basins’ hydrological modeling and underlying in Pakistan up to the Marala barrage [44]. In Pakistan, there processes [34, 35]. Generally, lumped, semidistributed, and are four streamfow gauge stations on the river, i.e., Marala distributed models are used for a watershed’s hydrological barrage, Khanki barrage, Qadirabad barrage, and Trimmu modeling. Lumped models consider spatially uniform wa- barrage. For the present research, the study area ranged from tershed characteristics. Te semidistributed model divides the Marala barrage to the Trimmu barrage (Figure 1). the watershed into subbasins with unique hydrological re- Since the Chenab River and Jhelum River converged at sponses [36]. Conversely, distributed models consider the the Trimmu barrage, hydrological modeling at the con- spatial and temporal variation of physical properties in the vergence point is not possible [39]. So, the study considered watershed. Tese models can interpolate the rainfall data and the assessment of outfow at Qadir Abad barrage with an predict water fow at ungauged locations. However, these assumption of free fow at Khanki barrage during the models require a considerable input data set for fow monsoon fooding season. Te selected catchment is situated ° ° ° ° estimations [37]. between latitudes 72 –78 E and longitudes 32 –34 N, Recently, studies reported that statistical assessment of spanning over ∼16,000 km , with a gradient of 0.4 m/km precipitation data had not yielded reliable results that downstream of plain areas [43]. Advances in Meteorology 3 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 40000 120000 200000 280000 360000 440000 520000 600000 680000 760000 840000 920000 W E 1 cm = 49 km Kilometers 0 30 60 120 180 240 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 40000 120000 200000 280000 360000 440000 520000 600000 680000 760000 840000 920000 Asia Elevations 130 - 920 Pakistan_and_Kashmir_Boundary 930 - 2,200 Study_Area_Boundary 2,300 - 3,400 Headworks 3,500 - 4,500 Upstream Raingages 4,600 - 7,100 Downstream Raingages Streams Figure 1: Location map of the Chenab River Basin showing catchment area, upstream and downstream gauging stations of Marala Barrage. Te crosssectional characteristics of the basin comprise Satellite precipitation datasets at the daily time scale 85 km in length with an average width of 800 m. Te scarcity (Table 1) consisting of GSMaP, IMERGE, and CHIRPS and sparsity of meteorological gauging stations are big con- were downloaded from the Japan Aerospace Exploration cerns in this region. Tere are only four rain gauges organized Agency (JAXA) (https://sharaku.eorc.jaxa.jp/GSMaP/), by the Pakistan Meteorological Department (PMD), which do National Aeronautics and Space Administration (NASA) (https://pmm.nasa.gov/data-access/downloads/gpm), and not meet the requirements of the World Meteorological Or- ganization (WMO) and are inadequate for hydrological the University of California, Santa Barbara’s Climate modeling for watershed management. Te central hydrology of Hazards Group (UC Santa Barbara) (https://chg.geog. the catchment is controlled by the summer monsoon and ucsb.edu/data/chirps/), respectively. GSMaP consists of winter seasons, where the summer monsoon season dominates four types of products; two real-time (GSMaP-NRT, and has triggered signifcant fooding in this region. GSMaP-Gauge, and NRT) and two postreal-time (GSMaP-Gauge, GSMaP-MVK). In the present work, 2.2. Data Description. Te data for the research was collected GSMaP-Gauge NRT (version 6) was used. In order to at a daily scale for the years 2015–2020 and consisted of formulate the GSMaP-Gauge NRT precipitation pre- a gauge rainfall dataset, observed streamfow, and a satellite dictions with a 4 h latency period, the error parameters precipitation dataset. In addition, the topographical data estimated for the postreal-time product of GSMaP-Gauge comprises a digital elevation model, land use, and soil type. are utilized. GSMaP-Gauge also employed a blending of Daily rainfall data for the selected rain gauge stations were passive microwave (PMW) and infrared (IR) data along acquired from the PMD. Streamfow data for stream gauging with a unifed gauge-based analysis of the global daily stations were collected from Pakistan’s Flood Forecasting precipitation dataset from the Climate Prediction Center Division (FFD). (CPC) [22]. 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 3240000 3300000 3360000 3420000 3480000 3540000 3600000 3660000 3720000 3780000 3840000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 000000 3210000 3270000 3330000 3390000 3450000 3510000 3570000 3630000 3690000 3750000 3810000 4 Advances in Meteorology Table 1: Specifcations of the three SPPs used in the study. Satellite precipitation datasets Specifcation GSMaP IMERG CHIRPS Developer JAXA NASA USGS and CHC ° ° ° Spatial resolution 0.1 0.1 0.05 Temporal resolution Daily Daily Daily Coverage 60 N-60 S 90 N-90 S 50 N-50 S Data availability 2014-onwards 2014-onwards 1981-onwards Latency 4 hours 14 hours 48 hours Frequency 1 hour 0.5 hour 24 hours Algorithm PMW and IR GMI and DPR CCD and gauge Version 6 6 2 Similarly, Global Precipitation Measurement (GPM) is precipitation records so that inconsistency does not sig- a multinational satellite project to integrate and enhance nifcantly infuence the average in one of the station’ records. precipitation observations from diferent satellites. IMERG If there are less than 10 stations located in a specifc region, is a GPM-based level 3 multisatellite precipitation algorithm the consistency of each station must be examined. Terefore, that incorporates all passive microwave and infrared-based at all four gauging stations located in the study area, the observations in the constellation. Typically, three products DMC technique was applied to examine the consistency of of IMERG precipitation (early-run, late-run, and fnal-run) annual rainfall data. are mostly considered. Early-run uses only forward morphing with a 4 hour latency period; however, late-run 3.2. Evaluation Statistics. Te efcacy of selected satellite and fnal-run use forward and backward morphing with precipitation products was assessed against four-gauge latency periods of 14 hours and 3.5 months, respectively station records with categorical and continuous metrics [45]. IMERG combines passive microwave, propagated pulse on a daily, ten daily, monthly, and seasonal scales from 2015 width modulation, and infrared radiation-based observa- to 2020. Diferent approaches have been used by comparing tions by the Kalman flter method to obtain precise esti- point precipitation data observed by rain gauges with pixel mation [46]. Tis research work utilized daily IMERG- precipitation data recorded by remote sensing satellites. late-run version 6 for statistical and hydrological assess- Usually, such procedures are based upon the upscale in- ments in the study area. terpolation of point values to grid scale data and the Moreover, CHIRPS is a quasi-global rainfall dataset that downscaling of grid data towards point values. Te rean- consists of three types of temporal data (daily, pentanal alyzed gridded data always vary from the station observa- (5 days), and monthly) with two types of spatial resolution tions and vice versa in several aspects. To avoid inaccuracies ° ° (0.05 , 0.25 ). Te daily data are considered real-time data caused by such upscale interpolation methods and down- with a latency period of 2 days, while pentanal and monthly scaling, a more direct approach has been proposed and used. data are considered post-real-time datasets with a 21 days In this approach, precipitation observed at stations falling latency period. Its algorithm is based on cold cloud duration within a grid cell will be averaged to obtain an estimate for (CCD) and ground gauge observations to approximate the the observed precipitation at the center of that grid cell and rainfall. Tis study employed the daily 0.05 grid CHIRPS then compared to the gridded value [21]. Tis approach has version 2.0 dataset [47]. been used in this study for comparison between rain gauge Furthermore, digital elevation model (DEM) was used to and satellite data sets. For this purpose, satellite precipitation represent the catchment’s topography and delineate the data at the daily time scale was downloaded and then watershed. Te present study collected DEM and Global converted into ten daily, monthly, and seasonal scales. In Map’s land cover data (Version 2) from the International selecting the tile of satellite precipitation data, PMD gauge Steering Committee for Global Mapping (ISCGM). Finally, value recording time (8:00 am daily) was kept in focus. Event the soil type data based on the Digital Soil Map of the World detection capability was evaluated with categorical metrics (DSMW), provided by the Food and Agriculture Organi- that include POD, FAR, and CSI. POD refects the ratio of zation (FAO), was used. accurately identifed rainfall events by the satellite con- cerning gauge rainfall data. FAR demonstrates the fraction of rainfall events in which the satellite predicts precipitation 3. Research Methodology while the rain gauge does not observe it. CSI represents 3.1. Double Mass Curve Analysis. DMC is employed to in- typically the fraction of rainfall occurrence accurately rec- spect the consistency of the hydrologic data and to adjust the ognized by the satellite. Continuous metrics measure the inconsistent precipitation data. In this graphical approach, quantitative diference between observed and predicted the cumulative data of a single station is compared with the precipitation. Tese metrics include bias, CC, and RMSE. pattern composed of cumulative data from other stations in Bias is the mean discrepancy between satellite estimation the area. Likewise, for the used pattern, enough gauging and rain gauge data. Depending on the quality of the rainfall stations must be included while checking the consistency of data, its value could be positive or negative, indicating Advances in Meteorology 5 overestimation and underestimation, respectively. CC esti- 3.4. Model Formulation. For the development of the IFAS mates the degree of agreement between the satellite and rain model, the extent of the target study area was defned by determining the latitude and longitude of the selected gauge precipitation data. RMSE depicts the mean dispersion of predicted precipitation around the known value of gauge catchment. Te IFAS model with a two-layered and three- observations. It is used to evaluate the precision of the tank structure was created by customizing the digital-based rainfall dataset [48]. From historical data, two rainy seasons, land cover, elevation, and soil type data to the appropriate monsoon (June to September) and westerly disturbance grid size. Te shapefle of the study area was imported into (November to February), have been established in the study the basin manager function of IFAS to defne basin and sub area: these were considered for seasonal evaluation at the basin boundaries. Te surface tank parameters were esti- daily scale. Tese statistical metrics are given as follows: mated utilizing the land cover data, while the aquifer tank parameters were tuned according to soil type data. Te POD � , essential aspect of hydrological modeling in a basin is ac- H + M curately estimating runof and water level initial conditions in the river course that afect the parameter optimization of FAR � , the model [31]. Te model was run six months before the H + F calibration of the food event to generate proper initial conditions until hydrological equilibrium was achieved. Te CSI � , principle of equifnality dictates that many combinations of H + M + F parameters are possible that give good agreement with the (1) 􏽐 (Si − Gi) i�1 observed streamfow data. Boundary conditions are essen- BIAS � , tial, especially when hydrological data is scarce and stan- 􏽳������������� dards for barrage operations are unknown. IFAS has an n 2 (Si − Gi). integrated water resources management (IWRM) interface 􏼐􏽐 i�1 RMSE � , that contains various techniques to incorporate barrage operating tasks. Te discharge fle technique was applied using the IWRM function to give the boundary condition in 􏽐 (Gi − Gm)(Si − Sm) I�1 􏽱������������� �􏽱������������� CC � , this study condition. Tis technique employed daily dis- n 2 n 2 􏽐 (Gi − Gm). 􏽐 (Si − Sm). i�1 i�1 charge data in the CSV fle to represent barrage operations. Marala barrage outfows were considered boundary con- where H, M, and F exhibit the number of hit, miss, and false ditions in this research due to data scarcity and unknown alarm events, while Gi and Si denote the gauge and satellite barrage operations upstream of the catchment. precipitation, Gm and Sm represent the mean of gauge and satellite precipitation data, and N indicates the total number of events used for evaluation. 3.5. Calibration Scenarios and Model Performance Indicators. Since the outputs of hydrological models are rarely capable of accurately refecting nature in its completeness, their 3.3. Explication of Hydrological (IFAS) Model. IFAS is a succinct runof analysis toolkit designed for food pre- performance must be evaluated before they can be employed in any decision-making process. Te IFAS is designed for diction in basins with insufcient hydrological and geo- physical information in developing countries. It is food analysis; therefore, it was calibrated and validated categorized as a physically distributed framework that can utilizing precipitation data collected during the monsoon integrate gauge rainfall data, satellite-based precipitation seasons (July to October) of 2015 and 2017, considering data, evaporation data, snowmelt data, and geophysical data medium and high fooding years, respectively (PMD/FFD). to simulate river course fow. It integrates grid-based Te model was calibrated individually using CHIRPS, datasets of topography, geology, and land cover to esti- GSMaP, and IMERG precipitation datasets. All datasets were validated against each calibration scenario. Te cali- mate the parameters of the physical conditions of a basin [49]. Te model can generate the channel network using bration of the model was achieved through a trial-and-error process. Te model performance was evaluated using model topographical data to defne the basin, sub-basin boundaries, 2 2 fow direction, and drainage patterns. It employs the Public performance indicators (MPI), R , NSE, and PBIAS. Te R Works Research Institute Distributed Hydrological Model is a statistical indicator representing the fraction of the (PWRI-DHM), consisting of a two- or three-tank structure dependent variable’s variance predicted by the independent and a routing model for runof simulation. Te three-tank variable. Te ideal value is 1, while a lower value than 1 structure comprises a surface, sub surface, and aquifer, while reveals the variation of model output. Te model perfor- the routing model comprises a kinematic hydraulic river mance with R > 0.5 is acceptable. Likewise, the NSE is the course routing tank. PWRI-DHM uses a nonlinear re- most often used method for determining correlation to test lationship to calculate each cell’s outfow based on the tank the efcacy of hydrological models. Te literature reveals model philosophy, considering Manning’s equation, Darcy’s a variety of acceptable, very good, and excellent value cat- law, and hyperbolic approximations. It uses a kinematic egories for NSE. Te calibration tolerance criteria are very wave equation to calculate the discharge in the river course subjective. Calibration with NSE is generally perceived as tank [38, 50, 51]. good if it is higher than 0.6 and excellent if it is more 6 Advances in Meteorology signifcant than 0.8 in the literature. Te model validation streamfow was taken as the baseline and the variation in criteria are less restrictive than the calibration levels. A value simulated fow was evaluated. Furthermore, the ability of of NSE greater than 0.5 is acceptable for validation, while each precipitation dataset to generate high and medium NSE greater than 0.7 is considered highly excellent. Te fows was examined through dependable fow exceedance, PBIAS examines the average tendency of the simulated fows where the extreme food events were represented in the to be greater or smaller than their observed fows. Te range of Q5–Q25 dependable fows, the medium fow re- perfect value of PBIAS is 0, and lower values represent quired for irrigation was designated by Q50 dependable accurate model reproduction [52]. fows, and Q70 dependable fows correspond to the water Tese MPIS are given as follows: availability for domestic supply. 􏽐 (Oi − Om)(Pi − Pm) I�1 􏽱������������ �􏽱������������ R � , 4. Results and Discussion n n 􏽐 (Oi − Om) 􏽐 (Pi − Pm) i�1 i�1 4.1. Consistency of Gauge Rainfall Data. Double mass n 2 analysis has been used to check the consistency of rainfall (Oi − Pi). i�1 (2) NSE � 1 − , data records at four stations, i.e., Gujrat, Sialkot, Sargodha, n 2 􏽐 (0i − Om). i�1 and Jhang. Te cumulative data of a single station was compared with the cumulative data of other stations in the 􏽐 (Oi − Pi) i�1 respective area. PBIAS � 100 ∗ , 􏽐 (0i) Te straight line shows data consistency, whereas any i�1 change in the straight line manifests a change in the data where Oi and Pi represent the observed and simulated fows, collection method that afects the relationship. For any Om and Pm are denoted by the mean values of the observed station, the rise of the curve from the trend line shows that and simulated fows, and n is the total number of events. there was more annual rainfall than in other stations. For all gauging stations, the R values were 0.98 to 0.99, and the 3.6. Implications of Sensitivity Analysis Technique. Te annual rainfall data were consistent with the DMC technique sensitivity analysis helps to examine the nonlinear variation (Figure 2). where CAR � is the cumulative annual rainfall. of highly uncertain parameters in the complex models. Te IFAS model was investigated to determine the most sensitive 4.2. Statistical Evaluation of SPPs at Diferent Temporal Scales. calibrating parameters severely afecting the calibrating Tis study identifed and quantifed the errors associated hydrograph. In this regard, the values of all calibrated pa- with satellite datasets. Te efcacy of selected satellite rameters were frst increased and then decreased by 20%, one products (CHIRPS, IMERG, and GSMaP) was assessed by one of their calibrated values. C is the 20 percent +20% statistically at daily, 10-daily, monthly, annual, and seasonal change in the calibrated parameter value and is determined scales using precipitation data recorded at PMD stations. using equation (3). So are the simulation results of a cali- brated hydrograph, and S is the change that occurred in +20% the simulation results when one of the calibrating param- 4.2.1. Daily and 10-Daily Scale. Statistical evaluations of eters’ values changed to +20%. Te mean percentage change selected SPPs at daily and 10-daily levels are presented in in the simulation results by increasing or decreasing the Table 2. In the case of categorical metrics, the mean POD of calibrating parameter value to 20% is called the sensitivity GSMaP was better than CHIRPS and IMERG on both index (I) and can be determined using equation (4). Te temporal scales. Te mean POD for CHIRPS and IMERG mean absolute sensitivity index (MASI) can be determined was lower by 50.79% and 22.22% for the daily scale with using equation (5). reference to GSMaP. In terms of mean FAR values, IMERG and GSMaP showed good agreement, and CHIRPS S − S 􏼁 0 ±20 C � 􏼨 􏼩 × 100, (3) underperformed. CSI gives more stable results due to the ±20% characteristics of the blending of POD and FAR. Te per- formance of GSMaP, IMERG, and CHIRPS was improved by C + C + −20 54%, 58%, and 55%, respectively, for CSI at the 10-daily (4) I � , ±20% scale. Remarkably better values of categorical metrics were 􏼌 􏼌 􏼌 􏼌 given by all the used satellite products at the 10 daily time 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 􏼌 􏽮 C + C 􏽯 􏼌 􏼌 􏼌 􏼌 +20 −20 scale as compared to the daily time scale. (5) MASI � . In the case of continuous metrics, the mean BIAS of GSMaP was better than the other two products at daily scale. A slight diference was observed between the mean RMSE 3.7. Flow Assessment Using Hydrological Signature. Te FDC values of the selected products on both temporal scales. is an infuential streamfow variability signature that de- However, it was noted that the used SPPs showed less errors scribes hydrological behavior. It is the graphical represen- (BIAS and RMSE) on a daily scale, compared to 10 daily. tation of fows and the percentage of time that the fows Probably this was due to a reduction in sample size at a larger equal or surpass each other. Satellite precipitation datasets time scale as compared to a smaller time scale. All SPPs did were evaluated through the FDC, in which the observed not show good agreement with rain gauge data at the daily Advances in Meteorology 7 5000 5000 4000 4000 Y = 0.9376X + 69.259 Y = 0.8308X + 85.058 R = 0.9921 R = 0.991 3000 3000 2000 2000 1000 1000 0 0 0 1000 2000 3000 4000 5000 0 1000 2000 3000 4000 5000 CAR Average of 3 stations (mm) CAR Average of 3 stations (mm) (a) (b) 3000 7000 Y = 1.6996 -157.9 Y = 0.628X + 63.17 5000 R = 0.989 2000 R = 0.9854 0 0 0 1000 2000 3000 4000 0 1000 2000 3000 4000 CAR Average of 3 stations (mm) CAR Average of 3 stations (mm) (c) (d) Figure 2: Consistency assessment of rain gauge data in the study area based on DMA. (a) Gujrat (GRT) Gauge Station. (b) Sargodha (SGD) Gauge Station. (c) Jhang (JNG) Gauge Station. (d) Sialkot (SKT) Gauge Station. Table 2: Evaluation of the used SPPs based on mean values of categorical and continuous metrics at daily and 10-daily time scales. Categorical metrics Continuous metrics Assessment scale Satellite product POD FAR CSI BIAS (mm) RMSE (mm) CC GSMaP 0.63 0.63 0.29 0.13 8.32 0.3 Daily scale CHIRPS 0.31 0.74 0.17 0.44 9.24 0.21 IMERG 0.49 0.68 0.24 0.08 9.57 0.25 GSMaP 0.97 0.14 0.83 1.3 23.75 0.71 10-daily scale CHIRPS 0.86 0.18 0.72 0.99 27.55 0.62 IMERG 0.97 0.16 0.82 4.42 23.35 0.74 a similar temporal pattern in monthly estimates from Jan- scale. Te correlation coefcient was low at 0.21–0.30 at the daily scale, while it was high at 0.71–0.74 at the 10-daily scale, uary to December 2015. All SPPs signifcantly overestimated showing better performance of SPPs at a larger time scale, precipitation for July through September 2020. identical with categorical metrics. Overall, the statistical On an annual time scale, a comparison was made be- performance of SPPs was lower on a daily scale and higher tween the rain gauge values and the used SPP values shown on a 10-daily scale. in Figure 4. Te annual average precipitation in the study area, as estimated from observations from 2015 to 2020 from four gauging stations, was 691 mm/year. It has been ob- 4.2.2. Monthly and Annual Scale. Figure 3 shows a monthly served that the selected satellite-based precipitation products overestimated the annual precipitation amounts. IMERG comparison of IMERG, GSMaP, and CHIRPS precipitation observations with the reference data for the entire study and GSMaP showed overestimations of 23.47% and 7.17%, respectively, while CHIRPS showed an overestimation of period (January 2015 to December 2020). Te GSMaP precipitation product represented the best monthly pre- 1.08% with reference to rain gauge values. cipitation temporal pattern. However, both IMERG and Te result showed some diference in estimating pre- CHIRPS were also capable of representing the temporal cipitation magnitudes by the IMERG products over the variability of observed precipitation over the study area, Chenab River basin of Pakistan, but the performance of albeit with notable overestimation. In July and August of CHIRPS and GSMaP encourages the utilization of SPPs in 2016, 2017, and 2018, all precipitation data sources (gauges, the study area at an annual time scale. Several researchers IMERG, GSMaP, and CHIRPS) revealed increased pre- have also reported identical fndings in diferent regions of cipitation magnitude. Almost all data sources exhibited the world. CAR JNG station (mm) CAR GRT station (mm) CAR SKT station (mm) CAR SGD station (mm) 8 Advances in Meteorology Jan-15 May-16 Oct-17 Feb-19 Jun-20 Gauge GSMaP IMERG CHIRPS Figure 3: Comparison among precipitation estimates by the PMD gauges and the three SPPs at the monthly time scale in the study area. varied from 0.21, 27, and 0.29 for CHIRPS, IMERG, and GSMaP, respectively. Intercomparing between SPP revealed that GSMaP depicted higher values of CC. Te statistical performance in terms of BIAS, RMSE, and CC of the selected products revealed that the overall efciency of GSMaP has remained higher than the other two products in the mon- soon season. 4.3.2. Winter Western Disturbance Season. Figure 6 shows the performance of SPPs towards estimation of precipitation during winter due to the western disturbance season (westerly waves) for the entire study period based on daily precipitation data. Te event detection capability revealed that the POD of GSMaP was higher and better than the other Gauge CHIRPS GSMaP IMERG two products. Te CHIRPS underperformed in terms of POD values. In the case of FAR, GSMaP outerperformed Figure 4: Comparison among average annual precipitation esti- than IMERG and CHIRPS. In the case of CSI, all selected mated by the PMD gauges and the three SPPs in the study area. SPPs revealed better performance during western distur- bances. In the case of bias value, IMERG overestimated the 4.3. Evaluation of SPPs at the Seasonal Scale precipitation, while CHIRPS showed excellent performance. While considering the results of RMSE, an agreement was 4.3.1. Summer Monsoon Season. Figure 5 shows the per- observed between the median values of CHIRPS and formance of SPPs through box plots during the monsoon GSMaP. Te box plot results showed the RMSE values season for the study period. Te categorical detection indices ranged from 2.5 to 7.5 mm/day for the selected satellite revealed that POD results were found in the range of 0.26 to products, and higher values were produced by the IMERG. A 0.85. Te precipitation detection capability of GSMaP was strong agreement between CHIRPS and IMERG was ob- better with a POD of 0.75, followed by IMERG and CHIRPS served for the CC results. Intercomparison revealed that with values of 0.56 and 0.34, respectively. Notably, all SPPs SPPs showed comparatively better statistical performance revealed large FAR values during the monsoon season. Te during western disturbance than monsoon season. performance of the selected SPPs was lower in terms of CSI. Conclusively, the statistical performance of GSMaP is Inter comparing the results of CSI revealed the better better than other SPPs, as also reported in [22, 53, 54], in performance of GSMaP for the monsoon period. In the case other regions of the world. of bias, an agreement was observed between CHIRPS and IMERG, while GSMaP revealed better performance. Te results of RMSE indicated that the frst quartile of daily data 4.4. Parametric Sensitivity Assessment. Figure 7 shows the was found in the range of 11 to 20 mm per day for the sensitivity analysis of the IFAS model for surface, aquifer, selected three products. Te values of the second and third and river course tank parameters based on the mean ab- solute sensitivity index (MASI). In the case of surface tank quartiles of RMSE were observed at about 11 to 17 mm/day. For the CC results, it was observed that the median values parameters, the surface tank height (HFMND) and fnal Average Annual Precipitation (mm/year) Precipitation (mm/month) Advances in Meteorology 9 0.85 0.80 0.75 0.75 0.65 0.70 0.55 0.65 0.45 0.60 0.35 0.25 0.55 CHIRPS CHIRPS GSMAP GSMAP POD FAR (a) (b) 0.40 2.00 0.35 1.00 0.30 0.25 0.00 0.20 0.15 -1.00 CHIRPS CHIRPS GSMAP GSMAP CSI BIAS (c) (d) 20.0 0.35 18.0 0.30 16.0 0.25 14.0 0.20 12.0 10.0 0.15 CHIRPS CHIRPS GSMAP GSMAP RMSE CC (e) (f ) Figure 5: Statistical (continuous and categorical) assessment results of the used SPPs at a seasonal scale during the summer monsoon period. (a) POD. (b) FAR. (c) CSI. (d) BIAS. (e) RMSE. (f ) CC. 0.90 0.65 0.85 0.55 0.80 0.45 0.75 0.35 0.70 0.25 0.65 0.60 0.15 CHIRPS CHIRPS GSMAP GSMAP FAR POD (a) (b) 0.60 0.30 0.40 0.20 0.20 0.10 0.00 0.00 CHIRPS CHIRPS GSMAP GSMAP CSI BIAS (c) (d) Figure 6: Continued. CSI POD CSI POD RMSE BIAS CC FAR BIAS FAR 10 Advances in Meteorology 8.0 0.35 6.0 0.25 4.0 0.15 2.0 0.05 CHIRPS CHIRPS GSMAP GSMAP RMSE CC (e) (f ) Figure 6: Statistical (continuous and categorical) assessment results of the used SPPs at seasonal scale during the winter westerly wave period. (a) POD. (b) FAR. (c) CSI. (d) BIAS. (e) RMSE. (f ) CC. infltration capacity of soil (SKF) are the most sensitive and-error method. Another surface tank parameter parameters, with a mean absolute sensitivity index (MASI) (FALFX) was tuned from 0 to 1 to control the subsurface of 8 and 3.5, respectively. It was observed that other surface fow to calibrate the model. Te values of FALFX parameters tank parameters, i.e., SNF, HFOD, and HFID, do not sig- were subsequently decreased to adjust the hydrograph in the calibration process. nifcantly impact the calibration of the model and show a lower mean absolute sensitivity index. Similarly, the Tree diferent calibration scenarios were established to investigate the capacity of selected SPPs to calibrate the IFAS evaluation of aquifer tank parameters indicated that the parameters, i.e., the efect of storage height to generate base model and to examine their efectiveness for diferent ap- plications. In the frst scenario, the model was calibrated fow (AGD) and the initial value used for calculation (HIGD), are the two most sensitive parameters, with MASI utilizing the CHIRPS satellite precipitation data, and then values of 452 and 99, respectively. Te aquifer parameter the validation process was completed using the GSMaP and HIGD depicted a direct relationship with the change in IMERG. For the second scenario, the IFAS model was simulation results of the hydrograph, while HCGD and calibrated utilizing the GSMaP, and the model was validated AUD do not infuence the calibration of the model. by using CHIRPS and IMERG for evaluation. In the third Meanwhile, the analysis of the river course tank pa- scenario, the IMERG precipitation dataset was utilized to calibrate the IFAS model and then validated against GSMaP rameters indicated that the parameters related to the co- efcients of a crosssection of a river, i.e., RLCOF and RBS, and CHIRPS. Calibration and validation of the IFAS model were evaluated using the model performance indicators, i.e., were sensitive to the simulated hydrograph. Te aquifer tank parameters are more sensitive than any other tank param- NSE, R2, and PBIAS. Te model’s performance on each scenario and comparison among the performance of the eters also reported by [55], and river tank parameters played signifcantly less in calibration. Terefore, it is suggested to three scenarios are presented in Table 3. introduce the option of actual groundwater conditions in the For the frst calibration scenario, the statistical perfor- IFAS model for the target area. mance indicators R , NSE, and PBIAS were 0.89, 0.86, and Te IFAS model was calibrated and validated for the −0.16, respectively. Te intercomparison results of model river Chenab at Qadir Abad barrage outlet for the monsoon validation for this scenario revealed the better performance periods of 2015 and 2017, respectively, utilizing the selected of the GSMaP dataset with R , NSE, and PBIAS values of SPPs. In the calibration process, initially the default pa- 0.85, 0.83, and 0.16, respectively. Te IMERG and CHIRPS rameters for surface and aquifer tanks were used to run the datasets showed slightly lower performance during the model. Te surface parameters were based on digital land model validation process. From the graphical presentation cover data, while the aquifer parameters were based on soil of scenario 1 in Figure 8, some variations in simulating low type data for the selected basin. Te parameters were tuned and high fows were observed by the SPPs. and optimized with the trial-and-error technique to bring For the second calibration scenario, the statistical per- them into sound agreement with the observed fow data. Te formance indicators (R , NSE, and PBIAS) were 0.97, 0.96, critical parameters considered for a successful calibration of and −0.03, respectively. According to the calibration criteria, the model are the coefcient of base fow regulation (AGD) this scenario displayed excellent performance, demon- for the aquifer tank, the surface tank height (HFMND), the strating that GSMaP precipitation data resulted in a robust fnal infltration capacity of the soil (SKF), and the initial and trustworthy testing model with utility and accuracy that height of infltration (HFOD) for the surface tank. Since could be used to check and compare the results produced HFOD is a surface parameter, it signifcantly infuences the from the IMERG and CHIRPS precipitation models. adjustment of the peak of the hydrograph. Due to the surface Comparison of model validation results revealed that the tank’s fve distinct feature classes, successful peak calibration GSMaP dataset outperformed the other datasets, with R , requires fne-tuning of the land cover parameter. Land cover NSE, and PBIAS values of 0.9, 0.89, and 0.14, respectively. A classes from the IFAS graphical module were used to cali- strong agreement was observed between IMERG and brate the model, which was then fne-tuned using a trial- CHIRPS-based simulated fows. Te ability of the GSMaP RMSE CC Advances in Meteorology 11 0 0 HFMND SKF HFMXD FALFX SNF HFOD HIFD AGD HIGD HCGD AUD (a) (b) RLCOF RBS RNS RRID RBW RGWD RHW RHS RBH RBET (c) Figure 7: Sensitivity assessments among various parameters of the IFAS model. (a) Surface parameters. (b) Aquifer parameters. (c) River course parameters. Table 3: Performance evaluation comparison among the used calibration and validation scenarios. Scenario 1 Scenario 2 Scenario 3 Year CHIRPS rainfall model GSMaP rainfall model IMERG rainfall model 2 2 2 SPP NSE PBIAS R SPP NSE PBIAS R SPP NSE PBIAS R Calibrated 2015 CHIRPS 0.86 −0.16 0.89 GSMaP 0.96 −0.03 0.97 IMERG 0.91 −0.11 0.92 Validated 2017 CHIRPS 0.78 0.17 0.83 GSMaP 0.80 0.25 0.89 CHIRPS 0.80 0.17 0.85 Validated 2017 GSMaP 0.80 0.16 0.84 IMERG 0.78 0.16 0.84 GSMaP 0.82 0.16 0.86 Validated 2017 IMERG 0.79 0.15 0.82 CHIRPS 0.77 0.26 0.88 IMERG 0.81 0.16 0.85 6000 6000 4000 4000 2000 2000 0 0 25-Jun-15 14-Aug-15 3-Oct-15 22-Nov-15 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME Time Observed-Flow Observed Flow CHIRPS-Flow CHIRPS Flow (a) (b) Figure 8: Continued. MASI Discharge (m3/s) MASI Discharge (m3/s) MASI 12 Advances in Meteorology 6000 6000 4000 4000 2000 2000 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 Time Time Observed Flow Observed Flow GSMaP Flow IMERG Flow (c) (d) Figure 8: Graphical representations of calibration and validation scenario 1. (a) Calibration-CHIRP. (b) Validation CHIRPS. (c) Validation GSMaP. (d) Validation IMERG. 6000 6000 2000 2000 0 0 25-Jun-15 14-Aug-15 3-Oct-15 22-Nov-15 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME TIME Observed Flow Observed Flow GSMaP Flow CHIRPS Flow (a) (b) 6000 6000 4000 4000 2000 2000 0 0 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME TIME Observed Flow Observed Flow IMERG Flow GSMaP Flow (c) (d) Figure 9: Graphical representations of calibration and validation scenario 2. (a) Calibration-GSMaP. (b) Validation CHIRPS. (c) Validation IMERG. (d) Validation GSMaP. dataset to optimize the parameters and calibrate the model For the third calibration scenario, the statistical per- formance indicators R was better when compared with the CHIRPS precipitation , NSE, and PBIAS for model cali- model. A graphical presentation of scenario 2 is shown in bration were observed at 0.92, 0.91, and −0.11, respectively, Figure 9. It depicted a trend identical to scenario 1, but a bit which exhibited excellent performance of this model improved simulation was observed in predicting low and according to the calibration rating described by [52]. Te R high fows. of the GSMaP, IMERG, and CHIRPS datasets were 0.87, Discharge (m3/s) 3 3 Discharge (m /s) Discharge (m /s) 3 3 Discharge (m /s) Discharge (m /s) Discharge (m3/s) Advances in Meteorology 13 6000 6000 4000 4000 2000 2000 0 0 25-Jun-15 14-Aug-15 3-Oct-15 22-Nov-15 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 TIME Time Observed-Flow Observed Flow IMERG-Flow CHIRPS Flow (a) (b) 6000.0 6000.0 4000.0 4000.0 2000.0 2000.0 0.0 0.0 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 25-Jun-17 14-Aug-17 3-Oct-17 22-Nov-17 Time TIME Observed Flow Observed Flow GSMaP Flow IMERG Flow (c) (d) Figure 10: Graphical representations of calibration and validation scenario 3. (a) Calibration-IMERG. (b) Validation CHIRPS. (c) Validation GSMaP. (d) Validation CHIRPS. 0 0 0 10 20 30 40 50 60 70 80 90 100 Exceeded Time (%) Exceedance Time (%) Observed GSMaP Observed GSMaP CHIRPS IMERG CHIRPS IMERG Figure 12: Assessment of the used SPPs in the term Figure 11: Assessment of the used SPPs in terms of hydraulic signature through FDC. dependable fows. values of 0.97, 0.96, and −0.03, respectively. For PBIAS 0.85, and 0.84, respectively, which depicts better GSMaP product performance for the third scenario also. A graphical evaluation, GSMaP produced a lower value of −0.03, compared to the other two products, CHIRPS and IMERG presentation of this scenario, Figure 10, shows identical simulated moderately higher values of −0.11 and −0.16, performance in forecasting low and high fows. It was noted that GSMaP outerperformed in terms of respectively. IMERG and CHIRPS were rated in the second and third positions for performance evaluation of model parameter optimization and fne-tuning of the IFAS model during the calibration procedure with R , NSE, and PBIAS calibration. In the case of the application and validation of Discharge (m /s) Discharge (m3/s) Discharge (m /s) Discharge (m3/s) Discharge (m3/s) Discharge (m /s) 14 Advances in Meteorology the hydrological model for the high monsoon food of 2017, (2) Statistical evaluation revealed that the efcacy of GSMaP outperformed in each scenario. Overall, the hy- GSMaP has been better, while CHIRPS showed more biases. Te performance of SPPs improved at 10- drological performance of GSMaP was more satisfactory than that of IMERG and CHIRPS, IMERG was ranked daily and monthly time scales than at the daily time second, while the CHIRPS exhibited a lower performance. scale. Higher values of uncertainties (bias and RMSE) were observed during the monsoon season than during the western disturbances. Missed and 4.5. Hydrological Signature-Based Assessment. Te hydro- false alarms were the main errors associated with logical performance of all selected satellite precipitation SPPs due to spatial mobility and the sudden bursting datasets was evaluated through FDC, in which daily ob- of clouds, specifcally during the monsoon season. served stream fows were taken as the baseline and the (3) Te stativity analysis revealed that the aquifer tank variation in simulated fow was determined. Te FDC results parameters were found to be the most sensitive. Te displayed that the selected satellite dataset has relatively base fow coefcient (AGD) was found to be the most inferior performance in capturing extreme fooding con- sensitive parameter in calibrating the IFAS model to ditions. While considering the medium fow, all datasets simulate fows using SPPs. showed excellent performance in the range of 20% to 40% (4) Te model calibration and validation scenarios in- exceedance time, as displayed in Figure 11. dicated that the GSMaP precipitation dataset has Similarly, these precipitation datasets do not yield sat- better capability to calibrate and validate the model isfactory results for the simulation of low fows. For all compared to IMERG and CHIRPS, with the highest precipitation datasets at the catchment outlet stage, an R , NSE and lower PBIAS values. It was also observed exceedance fow analysis was used to estimate the de- that the SPPs have relatively poor performance in pendable fow exceedance of Q5, Q10, Q25, Q50, and Q70. capturing extreme fooding events. While consid- Q5 denotes a fow that exceeds 5% of the analysis time, and so forth. Extreme food events are revealed by 5% and 10% ering the medium fows, in the range of 20%–40% exceedance time, all datasets showed excellent stream fows, while 50% dependability designates the me- dian fow, 70% dependable fow resembles the water performance. availability for agriculture, and higher dependable fows Findings of this study suggested that direct utilizations of correspond to the water availability for domestic supplies. satellite-based precipitation products were not promising at Te performances of SPPs to generate high, medium, and daily scales and bias correction is recommended. For food low fows was analyzed through these dependable fows. It modeling, the hydrological IFAS model should be calibrated was found that all SPPs data sets’ performance was lower, based on peak fow, considering the combination of sta- corresponding to Q5 and Q70. Te SPPs datasets can tistical and error indicators. Further studies may be carried generate medium fow in the range of Q25-Q50 Figure 12. out to assess the efectiveness of the available sensitivity analysis techniques in this study area. 5. Conclusion Acronyms Te present study evaluated three high-resolution multi- satellite precipitation estimation products statistically and SPPs: Satellite precipitation products hydrologically in the Chenab River catchment. Te con- IMERG: Integrated Multi-satellite Retrieval for GPM sistency of rain gauge data observed by PMD was examined GSMaP: Global Satellite Mapping of Precipitation by double mass analysis. Numerous statistical indicators CHIRPS: Climate Hazards Group Infrared Precipitation were applied at daily, monthly, and seasonal scales to detect with Station and quantify errors associated with these products. Tree DMC: Double mass curve diferent calibration scenarios were established for the hy- POD: Probability of detection drological assessment to analyze the satellite precipitation CSI: Critical success index datasets. A sensitivity analysis was performed to study the FAR: False alarm ratio most sensitive parameters of the distributed IFAS model. CC: Correlation coefcient Te hydrological signature was used to assess the potential of RMSE: Root mean square error satellite products to generate high, medium, and low fows. B: Bias Te existence of about 62 percent of the catchment area in IFAS: Integrated Flood Analysis System Indian-held Kashmir and the occurrence of only four FDC: Flow duration curve gauging stations in the rest of the catchment area are the R : Coefcient of determination major limitations of the study towards hydrological and AGD: Base fow coefcient statistical assessment of the satellite products in the study NSE: Nash−Sutclife efciency area, respectively. From the fndings of this study, it was PBIAS: Percentage bias observed as follows: PMD: Pakistan Metrological Department (1) PMD rain gauge-based precipitation data are con- WMO: World Meteorological Organization sistent and can be used for the assessment of satellite- FFD: Flood forecasting division based precipitation datasets. JAXA: Japan Aerospace Exploration Agency Advances in Meteorology 15 cipondoh, tangerang Indonesia,” Journal of Environment and NASA: National Aeronautics and Space Administration Earth Science, vol. 9, no. 1, pp. 52–61, 2019. PMW: Passive microwave [9] M. Aslam, “Flood management current state, challenges and IR: Infrared prospects in Pakistan: a review,” Mehran University Research CPC: Climate prediction center Journal of Engineering and Technology, vol. 37, no. 2, GPM: Global precipitation measurement pp. 297–314, 2018. CCD: Cold cloud duration [10] M. S. Shrestha, M. R. Khan, N. Wagle, Z. Ahmad Babar, DEM: Digital elevation model V. R. Khadgi, and S. Sultan, Review of Hydrometeorological ISCGM: International Steering Committee for Global Monitoring and Forecasting System for Floods in the Indus Mapping Basin in Pakistan, Elsevier Inc, Amsterdam, Netherlands, DSMW: Digital soil map of the world [11] S. Sorooshian, A. Aghakouchak, P. Arkin et al., “Advanced FAO: Food and Agriculture Organization concepts on remote sensing of precipitation at multiple PWRI- Public Works Research Institute distributed scales,” Bulletin of the American Meteorological Society, DHM: hydrological model vol. 92, no. 10, pp. 1353–1357, 2011. IWRM: Integrated water resources management [12] A. U. Rehman, F. Chishtie, W. A. Qazi, S. Ghufar, I. Fatima, MPIs: Model performance indicators. and K. Fatima, “Evaluation of three-hourly TMPA rainfall products using telemetric rain gauge observations at lai nullah Data Availability basin in islamabad, Pakistan,” Remote Sensing, vol. 10, no. 12, p. 2040, 2018. Te data used to support the fndings of this study are [13] Y. Zhang, A. Sun, H. Sun et al., “Error adjustment of TMPA available from the corresponding author upon request. satellite precipitation estimates and assessment of their hy- drological utility in the middle and upper yangtze River Basin, Conflicts of Interest China,” Atmospheric Research, vol. 216, pp. 52–64, 2019. [14] C. Miao, H. Ashouri, K. L. Hsu, S. Sorooshian, and Q. Duan, Te authors declare that they have no conficts of interest. “Evaluation of the PERSIANN-CDR daily rainfall estimates in capturing the behavior of extreme precipitation events over Acknowledgments China,” Journal of Hydrometeorology, vol. 16, no. 3, pp. 1387–1396, 2015. Te authors are grateful to the Center of Excellence in Water [15] B. Yong, Y. Hong, L. L. Ren et al., “Assessment of evolving Resource Engineering for support in conducting this study. TRMM-based multisatellite real-time precipitation estima- Te authors are also thankful to PMD/FFD and satellite tion methods and their impacts on hydrologic prediction in organizations for the data used in the study. a high latitude basin,” Journal of Geophysical Research: At- mospheres, vol. 117, no. 9, 2012. References [16] D. B. Michelson, “Systematic correction of precipitation gauge observations using analyzed meteorological variables,” Jour- [1] M. T. Mendoza and R. Schwarze, “Sequential disaster fo- nal of Hydrology, vol. 290, no. 3-4, pp. 161–177, 2004. rensics: a case study on direct and socio-economic impacts,” [17] P. S. Roy, M. D. Behera, and S. K. Srivastav, “Satellite remote Sustainability, vol. 11, no. 21, pp. 1–20, 2019. sensing: sensors, applications and techniques,” Proceedings of [2] A. Mishra, A. Alnahit, and B. Campbell, “Impact of land uses, the National Academy of Sciences India Section A - Physical drought, food, wildfre, and cascading events on water quality Sciences, vol. 87, no. 4, pp. 465–472, 2017. and microbial communities: a review and analysis,” Journal of [18] K. J. Beven and H. L. Cloke, “Comment on ‘Hyperresolution Hydrology, vol. 596, Article ID 125707, 2021. Global Land Surface Modeling: meeting a Grand Challenge [3] F. Dottori, W. Szewczyk, J. C. Ciscar et al., “Increased human for Monitoring Earth’s Terrestrial Water’ by Eric F. Wood et and economic losses from river fooding with anthropogenic Al,” Water Resources Research, vol. 48, no. 1, pp. 1–10, 2012. warming,” Nature Climate Change, vol. 8, no. 9, pp. 781–786, [19] K. Yan, G. Di Baldassarre, D. P. Solomatine, and G. J. P. Schumann, “A review of low-cost space-borne data for [4] A. Mishra, S. Mukherjee, B. Merz et al., “An overview of food food modelling: topography, food extent and water level,” concepts, challenges, and future directions,” Journal of Hy- Hydrological Processes, vol. 29, no. 15, pp. 3368–3387, 2015. drologic Engineering, vol. 27, no. 6, pp. 1–30, 2022. [20] Z. N. Musa, I. Popescu, and A. Mynett, “A review of appli- [5] S. Janjua, I. Hassan, S. Muhammad, S. Ahmed, and A. Ahmed, cations of satellite SAR, optical, altimetry and DEM data for “Water management in Pakistan’s Indus basin: challenges and surface water modelling, mapping and parameter estimation,” opportunities,” Water Policy, vol. 23, no. 6, pp. 1329–1343, Hydrology and Earth System Sciences, vol. 19, no. 9, pp. 3755–3769, 2015. [6] A. R. Iqbal, “Environmental issues of Indus River Basin: an [21] M. Masood, A. S. Shakir, A. H. Azhar, G. Nabi, and analysis,” Institute for Strategic Studies, Research & Analysis H. U. Rehman, “Assessment of real time, multi-satellite (ISSRA), vol. 5, no. 1, National Defence University, Islamabad, precipitation products under diverse climatic and topo- Pakistan, 2013. graphic conditions,” Asia-Pacifc Journal of Atmospheric [7] Federal Communications Commission, “Government of Sciences, vol. 56, no. 4, pp. 577–591, 2019. pakistan ministry of water resources,” Annual Flood Report [22] J. Shi, F. Yuan, C. Shi et al., “Statistical evaluation of the latest 2017, pp. 1–118, Federal Communications Commission, Washington, DC, USA, 2017. GPM-era IMERG and GSMaP satellite precipitation products in the yellow river source region,” Water (Switzerland), [8] J. Pancawati, S. Nurisjah, and B. Pramudya, “Application of tank model for ungauged reservoir management: a case of situ vol. 12, no. 4, pp. 1006–1023, 2020. 16 Advances in Meteorology [23] N. Saddique, M. Muzammil, I. Jahangir et al., “Hydrological the issen watershed in the souss basin, Morocco,” European evaluation of 14 satellite-based, gauge-based and reanalysis Scientifc Journal, vol. 12, no. 18, p. 347, 2016. [38] A. Sugiura, S. Fujioka, S. Nabesaka, M. Tsuda, and Y. Iwami, precipitation products in a data-scarce mountainous catch- “Development of a food forecasting system on the upper ment,” Hydrological Sciences Journal, vol. 67, no. 3, Indus catchment using IFAS,” Journal of Flood Risk Man- pp. 436–450, 2022. agement, vol. 9, no. 3, pp. 265–277, 2016. [24] G. Villarini, W. F. Krajewski, and J. A. Smith, “New paradigm [39] A. Shahzad, H. F. Gabriel, S. Haider, A. Mubeen, and for statistical validation of satellite precipitation estimates: M. J. Siddiqui, “Development of a food forecasting system application to a large sample of the TMPA 0.25 3-hourly using IFAS: a case study of scarcely gauged Jhelum and estimates over Oklahoma,” Journal of Geophysical Research, Chenab River basins,” Arabian Journal of Geosciences, vol. 11, vol. 114, no. 12, pp. 12106–12112, 2009. no. 14, p. 383, 2018. [25] B. Nijssen and D. P. Lettenmaier, “Efect of precipitation [40] A. Aziz and S. Tanaka, “Regional parameterization and ap- sampling error on simulated hydrological fuxes and states: plicability of integrated food analysis system (IFAS) for food anticipating the global precipitation measurement satellites,” forecasting of upper-middle Indus river classifcation of hy- Journal of Geophysical Research: Atmospheres, vol. 109, no. 2, drological models,” Pakistan Journal of Meterorology, vol. 8, pp. 1–15, 2004. no. 15, pp. 21–38, 2010. [26] J. R. McCollum, R. Witold, R. R. Krajewski, and M. B. Ba, [41] M. Miyamoto, M. Ono, S. Nabesaka, T. Okazumi, and “Evaluation of biases of satellite rainfall estimation algorithms Y. Iwami, “Applicability of a food forecasting method uti- over the continental United States,” Journal of Applied Me- lizing global satellite information to an insufciently-gauged teorology, vol. 41, no. 11, pp. 1065–1080, 2002. River Basin: A case of a River Basin in the Philippines,” in [27] F. L. Conti, K. L. Hsu, L. V. Noto, and S. Sorooshian, Proceedings of the 11th International Conference on Hydro- “Evaluation and comparison of satellite precipitation esti- informatics, New York, NY, USA, August 2014. mates with reference to a local area in the mediterranean sea,” [42] H. Ishak, M. D. Nor, L. M. Sidek et al., “Flood forecasting and Atmospheric Research, vol. 138, pp. 189–204, 2014. early warning system for dungun river basin,” IOP Conference [28] M. Masood, G. Nabi, M. Babur, A. H. Azhar, and M. Kaleem Series: Earth and Environmental Science, vol. 16, no. 1, 2013. Ullah, “Disintegration of uncertainties associated with [43] R. Awan, “Pakistan food Management.Pdf (Application/Pdf real-time multi-satellite precipitation products in diverse object),” Mainstream, vol. 1, pp. 1–4, 2003. topographic and climatic area in Pakistan,” Journal of [44] M. Riaz, A. Aziz, and S. Hussain, “Flood forecasting of an Mountain Science, vol. 18, no. 3, pp. 716–734, 2021. ungauged trans-boundary Chenab River Basin using dis- [29] Y. Tuo, Z. Duan, M. Disse, and G. Chiogna, “Evaluation of tributed hydrological model integrated food analysis system precipitation input for swat modeling in alpine catchment: (IFAS),” Pakistan Journal of Meteorology, vol. 13, no. 26, a case study in the adige River Basin (Italy),” Science of the pp. 51–62, 2017. Total Environment, vol. 573, pp. 66–82, 2016. [45] G. J. Hufman, D. T. Bolvin, D. Braithwaite et al., Algorithm [30] T. M. Sharannya, N. Al-Ansari, S. Deb Barma, A. Mahesha, Teoretical Basis Document (ATBD) Version 5.2 for the NASA and A. Mahesha, “Evaluation of satellite precipitation prod- Global Precipitation Measurement (GPM) Integrated Multi- ucts in simulating streamfow in a humid tropical catchment satellitE Retrievals for GPM (I-MERG), GPM Project, of India using a semi-distributed hydrological model,” Water Greenbelt, MD, USA, 2019. (Switzerland), vol. 12, no. 9, p. 2400, 2020. [46] S. Sharma, Y. Chen, X. Zhou et al., “Evaluation of GPM-era [31] M. Umer, H. F. Gabriel, S. Haider, A. Nusrat, M. Shahid, and satellite precipitation products on the southern slopes of the M. Umer, “Application of precipitation products for food central himalayas against rain gauge data,” Remote Sensing, modeling of transboundary River Basin: a case study of vol. 12, no. 11, p. 1836, 2020. Jhelum basin,” Teoretical and Applied Climatology, vol. 143, [47] C. Funk, P. Peterson, M. Landsfeld et al., “Te climate hazards no. 3-4, pp. 989–1004, 2021. infrared precipitation with stations - a new environmental [32] R. H. McCuen, “Te role of sensitivity analysis in hydrologic record for monitoring extremes,” Scientifc Data, vol. 2, modeling,” Journal of Hydrology, vol. 18, no. 1, pp. 37–53, p. 150066, 2015. [48] M. Masood, A. S. Shakir, and H. U. Rehman, “Assessment of [33] K. K. Yilmaz, H. V. Gupta, and T. Wagener, “A process-based GPM based integrated multi-satellite retrievals under diverse diagnostic approach to model evaluation: application to the climatic and topographic conditions in Pakistan,” vol. 23, NWS distributed hydrologic model,” Water Resources Re- pp. 17–29, 2018. search, vol. 44, no. 9, pp. 1–18, 2008. [49] T. Sugiura, K. Fukami, and H. Inomata, “Development of [34] K. Sawicz, T. Wagener, M. Sivapalan, P. A. Troch, and integrated food analysis system (IFAS) and its applications,” G. Carrillo, “Catchment classifcation: empirical analysis of in Proceedings of the World Environmental and Water Re- hydrologic similarity based on catchment function in the sources Congress, vol. 316, Pittsburgh, PA, USA, May 2008. eastern USA,” Hydrology and Earth System Sciences, vol. 15, [50] S. Sutikno, Y. L. Handayani, M. Fauzi, Fitriani, and A. Kurnia, no. 9, pp. 2895–2911, 2011. “Hydrologic modelling using TRMM-based rainfall products [35] I. K. Westerberg and H. K. McMillan, “Uncertainty in hy- for food analysis,” MATEC Web of Conferences, vol. 101, drological signatures,” Hydrology and Earth System Sciences, pp. 2–6, 2017. vol. 19, no. 9, pp. 3951–3968, 2015. [51] A. Aziz and S. Tanaka, “Regional parameterization and ap- [36] F. Mendonça, R. P. Oliveira, and F. F. Mauad, “Lumped versus plicability of integrated food analysis system (IFAS) for food distributed hydrological modeling of the jacare-guaçu ´ basin, forecasting of upper-middle Indus River classifcation of Brazil,” Journal of Environmental Engineering, vol. 144, no. 8, hydrological models,” Pakistan Journal of Meterorology, vol. 8, Article ID 04018056, 2018. no. 15, pp. 21–38, 2010. [37] H. Brirhet and L. Benaabidate, “Comparison of two hydro- [52] D. N. Moriasi, D. J. G. Arnold, M. W. Van Liew, R. L. Bingner, logical models (lumped and distributed) over A pilot area of R. D. Harmel, and T. L. Veith, “Model evaluation guidelines Advances in Meteorology 17 for systematic quantifcation of accuracy in watershed sim- ulations,” Transactions of the ASABE, vol. 50, no. 3, pp. 885–900, 2007. [53] G. Tang, M. P. Clark, S. M. Papalexiou, Z. Ma, and Y. Hong, “Have satellite precipitation products improved over last two decades? A comprehensive comparison of GPM IMERG with nine satellite and reanalysis datasets,” Remote Sensing of Environment, vol. 240, Article ID 111697, 2020. [54] S. Ning, S. Fan, P. Udmale, J. Jin, B. Raj Tapa, and H. Ishidaira, “Error analysis and evaluation of the latest GSMap and IMERG precipitation products over eastern China,” Advances in Meteorology, vol. 2017, Article ID 1803492, 16 pages, 2017. [55] A. Nusrat, H. F. Gabriel, S. Haider, and M. Siddique, “Sen- sitivity analysis and optimization of land use/cover and aquifer parameters for improved calibration of hydrological model,” Mehran University Research Journal of Engineering and Technology, vol. 41, no. 2, pp. 21–34, 2022.

Journal

Advances in MeteorologyHindawi Publishing Corporation

Published: May 3, 2023

There are no references for this article.