Enhancing Rainfall Measurement Using Remote Sensing Data in Sparse Rain Gauge Networks: A Case Study in White Nile State, Sudan ^†

Abstract

Monitoring rainfall is essential to understanding hydrological processes, managing water resources, and mitigating drought and flood risks. Many regions, particularly in developing countries, have sparse rain gauge networks, which limit spatial coverage and result in inaccurate rainfall estimates. By combining remote sensing data with rain gauge measurements, rainfall estimates can be improved, and spatial coverage can be enhanced. Remote sensing techniques provide a valuable resource for supplementing and enhancing rainfall monitoring in such areas. This study leverages Global Precipitation Measurement (GPM) satellite data to enhance rainfall estimation in White Nile State, Sudan, where only two rain gauge stations are operational and the state’s total area is 39.600 km². GPM data, well-known for its high temporal and spatial resolution, offers a promising alternative to mitigate the limitations of sparse ground-based networks. The study integrates GPM satellite data with ground-based measurements through statistical and geostatistical techniques, as well as validation, to improve rainfall accuracy. The results show that, on average, GPM data and rain gauge measurements exhibit a strong correlation of 0.87, with an annual RMSE of 10.23 mm and an AME of 8.25 mm. These findings demonstrate that GPM data effectively complements traditional rain gauge observations by accurately capturing spatial rainfall distributions and extreme precipitation events. The findings underscore the potential of remote sensing to provide reliable rainfall information in data-scarce regions, contributing to better water resource management and disaster risk reduction strategies.

1. Introduction

Rainfall data plays a pivotal role in hydrological modeling, influencing the accuracy and reliability of forecasts to manage water effectively [1]; therefore, the availability of accurate rainfall data is essential for making informed decisions in various industries, including agriculture, urban planning, and disaster management [2,3]. In regions where agriculture is a cornerstone of livelihoods, accurate precipitation data is essential for driving socioeconomic progress and safeguarding water resources [4]. Ground-based precipitation data are traditionally collected using rain gauge networks. However, these networks often suffer from spatial sparsity, data gaps, and measurement uncertainties due to environmental factors like wind, evaporation, and splashing [5]. Rain gauge precipitation data usually contain gaps and inhomogeneities. Meanwhile, remote sensing precipitation has emerged as the most promising potential precipitation product, due to its high spatial and temporal resolution [6].

To overcome these limitations, remote sensing technologies—particularly satellite-based precipitation products—have become essential tools [6]. Among them, the Global Precipitation Measurement (GPM) mission, launched in 2014 as a successor to TRMM, launched in 1997, has significantly improved global precipitation monitoring. With 30 min temporal and 0.1° spatial resolution, GPM’s satellite constellation provides near-global coverage equipped with microwave and radar sensors [7]. Several studies have demonstrated that GPM products such as IMERG are useful for capturing precipitation patterns across various climatic zones, but many biases persist, especially in complex terrain and data-scarce regions [8,9]. Data-scarce regions can be enhanced by satellite-gauge synergy, but expanding gauge networks remains important in refining calibrations and reducing uncertainties [10].

Satellite- and radar-based monitoring systems offer spatial information critical for managing water resources across diverse geographic areas [2]. Rainfall estimates derived from remote sensing data exhibit varying degrees of accuracy when compared to ground-based measurements [11]. Despite these advancements, satellite-derived rainfall still needs to be calibrated and validated against ground-based observations [6]. There is a wide range of spatial–temporal variability in precipitation, and there is a limited number of gauges available in many situations. In fact, only 1.6% of the Earth’s surface is within 10 km of a gauge, while 5.9% is within 25 km [12]. Research has shown that combining satellite estimates with sparse gauge networks can enhance spatial coverage and reduce uncertainty, particularly in arid and semi-arid regions [13].

In Sudan’s White Nile State, where only two operational rain gauges exist, this study evaluates the integration of GPM satellite data with ground-based measurements to improve rainfall estimation. By validating satellite outputs against gauge observations, the study assesses the reliability of remote sensing in a data-scarce environment. The findings reveal strong correlation, affirming the potential of satellite-gauge synergy for early flood warnings, drought preparedness, and agricultural resilience. Furthermore, the study emphasizes the need to expand and optimize rain gauge networks through statistical analysis and spatial modeling to support calibration efforts and enhance regional water management.

2. Study Area

Sudan is positioned in northeastern Africa, and it is the third-largest country on the continent. White Nile State is situated in the southern region of Sudan. It is located between the Latitudes (12° to 15°30′ N), and on the longitudes (31°30′ to 33°9′ E). The state is bordered by five other Sudanese states. After South Sudan’s secession, White Nile State became an international frontier. The White Nile State’s total area is 39.600 km² and falls into nine localities, as shown in Figure 1. The state’s three different ecological zones are present in the White Nile State. Its semi-desert, arid, and semi-arid zones range from north to south [14]. The larger size of the state area is predominated by the semi-arid ecological zones, which are also known as low-rainfall or poor Savannah land. The northern part of the State lies in the semi-desert zone, and a small and limited pocket in the southern part lies within the sub-humid zone. The White Nile state’s annual rainfall ranges from 300 mm to more than 600 mm in the south. Rain quantities and distribution in the White Nile, as well as across Sudan, are important determinants of crop success and LULC change.

3. Materials and Methods

3.1. Data Sources

Meteorological data for this study were obtained from the Computer Center of the Meteorological Authority, part of the Ministry of Irrigation and Water Resources (MIWR) in Sudan. The study focuses on two rain gauge stations located in the White Nile State: Kosti and El Diwaim (see Figure 1). These stations provided monthly rainfall records spanning a 21-year period from 2002 to 2022.

In addition to ground-based observations, the study incorporated satellite-derived precipitation data from the Tropical Rainfall Measurement Mission (TRMM) and its successor, the Global Precipitation Measurement (GPM) mission. Specifically, the Integrated Multi-satellitE Retrievals for GPM (IMERG) Version 07 Final Run was used, which offers high-resolution global precipitation estimates. The data were obtained from NASA through the Giovanni platform (https://giovanni.gsfc.nasa.gov/ (accessed on 15 August 2025)). The IMERG data used in this study consists of monthly rainfall measurements with a spatial resolution of 0.1° × 0.1°, approximately equivalent to 10 km × 10 km.

3.2. Data Integration Framework

3.2.1. Statistical Techniques

The study used several common statistical measurements to assess the correlation and degree of deviation of the IMERG precipitation products quantitatively against ground rain gauges, including Pearson’s correlation (R), root mean square error (RMSE), mean absolute error (MAE), as shown in (Figure 2). In precise terms, R was used to measure the linear correlation between satellite precipitation data and rain gauge stations; RMSE of satellite precipitation products was used to measure the overall error level based on the difference between the sequence of satellite observations and the actual observations; and MAE was used to determine the deviation trend between IMERG products and ground rainfall stations as a whole.

3.2.2. Geostatistical Techniques

The study used Inverse Distance Weighting (IDW), a widely used geostatistical interpolation technique, to map the spatial distribution of rainfall. The objective is to compare rainfall data from the (TRMM/GPM) satellite with ground-based rain gauge measurements spatially. Due to the limited number of rain gauge stations (only two) over the study area see (Figure 1), we implemented a buffer zone of 25 km around each station to improve interpolation accuracy and account for temporal and spatial variability.

4. Results and Discussion

4.1. GPM Data Validation Using Statistical Performance Metrics Against Rain Gauges

The accuracy of the GPM IMERG satellite-derived precipitation data was evaluated by comparing it against ground-based rain gauge observations from the Kosti and El Diwaim stations using key statistical performance metrics (

Table 1. GPM data validation against rain gauges at Kosti and El Diwaim stations.

Statistical Metrics	Kosti vs. GPM	El Diwaim vs. GPM
RMES	11.02	9.27
MAE	9.20	7.30
R	0.87	0.88
RE (%)	34	20

). El Diwaim’s data demonstrates better agreement with GPM estimates, as demonstrated by lower RMSE (9.27 mm), MAE (7.30 mm), and RE (20%) than Kosti’s. However, Kosti exhibits a slightly lower RMSE (11.02 mm) and MAE (9.20 mm), but a comparable correlation coefficient (R = 0.87). These metrics are averaged per year, providing an average yearly comparison of the dataset’s accuracy and consistency with GPM. In this comprehensive evaluation, the GPM data are assessed for their reliability in hydrological and climatological applications.

The scatter plot is used to validate GPM rainfall data against ground-based measurements. The correlation between satellite-based GPM precipitation products and rain gauge observations is notably strong, with a Pearson correlation of R = 0.87 for Kosti and R = 0.88 for El Diwaim (Figure 3 and Figure 4 and Table 1). These values indicate a robust linear relationship and suggest that GPM data can reliably reflect rainfall patterns in the study area.

However, despite the strong correlation, certain limitations persist in the satellite estimates. Discrepancies between GPM and ground observations, where most of the time GPM overestimates the stations (Figure 5), may arise due to factors such as the coarse spatial resolution of satellite data (approximately 10 km²), localized rainfall variability, and regional climatic influences.

The seasonal variation in correlation likely reflects regional climatic influences. During peak rainy months, rainfall tends to be more widespread and uniform, allowing satellite sensors to detect and quantify it more effectively. Contrary to this, during dry or transitional periods, precipitation is sparse and localized, reducing the accuracy of satellite-based measurements (see Figure 6). These findings indicate that satellite sensors struggle to detect light and localized rainfall, especially during dry or transitional periods. These results are in agreement with several studies [13,15].

4.2. Spatial Rainfall Patterns

Rainfall spatial patterns across the study area reveal notable shifts, highlighting zones of increasing and decreasing precipitation. Variability can be seen in the gradient from blue to red as show in (Figure 7). These changes underscore the dynamic nature of rainfall distribution and its implications for water resource planning.

4.3. Enhanced Spatial Coverage

Geostatistical analysis is a valuable tool for comparing rainfall patterns. When using it, the interpolation method used to estimate rainfall in the study area can be assessed for accuracy and reliability. The distribution curve provides a visual representation of how well the model predicts rainfall (Figure 8). Since the stations are limited, we cannot use them as references, but it minimizes the errors mathematically.

5. Conclusions

This study demonstrates the effectiveness of GPM satellite data in enhancing rainfall estimation in regions with limited ground-based observations. By integrating remote sensing with sparse rain gauge networks in White Nile State, Sudan, the analysis provided a more complete understanding of spatial and temporal rainfall patterns. The strong and statistically significant correlations observed at both Kosti (R = 0.87) and El Diwaim (R = 0.88) validate the reliability of GPM data for hydrological applications.

Despite the challenges posed by a sparse gauge network, the use of statistical and geostatistical techniques enabled robust validation of satellite estimates. However, the results also highlight limitations in capturing light and localized rainfall, particularly during dry or transitional seasons. These discrepancies underscore the need to expand ground-based networks and improve satellite calibration to enhance accuracy.

The integration of satellite data into water resource management and disaster risk reduction frameworks offers a promising pathway toward climate resilience. This approach supports more informed decision-making in data-scarce regions, contributing to sustainable water governance and improved preparedness for extreme weather events.