Next Article in Journal
Long-Term Spatiotemporal Analysis of Crop Water Supply–Demand Relationship in Response to Climate Change and Vegetation Greening in Sanjiang Plain, China
Previous Article in Journal
Nighttime Tweek Characteristics in Mid–Low Latitudes: Insights from Long-Term VLF Observations in China
Previous Article in Special Issue
The Performance of GPM IMERG Product Validated on Hourly Observations over Land Areas of Northern Hemisphere
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Remote Sensing
Volume 17
Issue 3
10.3390/rs17030439
remotesensing-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Local Influence of Surface Relative Humidity on Weather Radar Rainfall Observations over an Agricultural Semi-Arid Area

by
Francesc Polls
1,
Joan Bech
1,2,*,
Mireia Udina
1,
Eric Peinó
1 and
Tomeu Rigo
3
1
Department Applied Physics—Meteorology, Universitat de Barcelona, Martí Franquès 1, 08028 Barcelona, Spain
2
Water Research Institute, Universitat de Barcelona, 08028 Barcelona, Spain
3
Equip de Predicció i Vigilància, Servei Meteorològic de Catalunya, Dr. Roux, 80, 08013 Barcelona, Spain
*
Author to whom correspondence should be addressed.
Remote Sens. 2025, 17(3), 439; https://doi.org/10.3390/rs17030439
Submission received: 29 December 2024 / Revised: 17 January 2025 / Accepted: 24 January 2025 / Published: 27 January 2025
(This article belongs to the Special Issue Applications of Remotely Sensed Data in Hydrology and Climatology (Second Edition))
Browse Figures
Versions Notes

Abstract

:
Agricultural areas in semi-arid regions modify low-level atmospheric conditions through changes in heat and moisture surface fluxes and enhanced evapotranspiration. This study aims to investigate the influence of near-ground-level relative humidity (RH) on local precipitation characteristics in a relatively flat, mid-latitude, semi-arid agricultural region, divided into a rainfed and an irrigated area with high evapotranspiration contrast in summer. The region was selected for the Land Surface Interactions with the Atmosphere over the Iberian Semi-Arid Environment (LIAISE) international field campaign in 2021 and here is studied using Automatic Weather Station observations and C-band weather radar data covering six years. Summer RH records show clear contrasts between irrigated and non-irrigated areas, unlike rain gauge and radar-derived rainfall, which do not exhibit substantial differences. A closer analysis indicates that RH differences between irrigated and non-irrigated areas before rainfall tend to diminish for several hours after the rainfall onset. This suggests that the presence of rainfall is temporally more important than whether the terrain is irrigated or not. Examination of radar reflectivity (Z) profiles considered convective and non-convective cases averaged during the first 30 and 180 min from the precipitation onset. Results indicated a dependence on ground-level RH for convective cases, leading to higher Z values with higher RH, clearer for the first 30 min averaged profiles. Finally, a linear relation was found between the lowest 1 km radar Z value and collocated RH for the first 30 min period of convective precipitation, increasing Z with RH. These results point out that, despite no differences in precipitation amounts found over contiguous irrigated and non-irrigated areas, there is a local impact of low-level moisture on convective rainfall.

1. Introduction

Agricultural activity, particularly irrigation in arid and semi-arid regions, is known to significantly modify surface- and low-level atmospheric variables [1,2]. Changes in the partitioning of latent and sensible heat fluxes, an increase in evapotranspiration and water vapour availability, and air cooling near the ground are among the effects of irrigation that have been reported [3,4]. Subsequent effects of irrigation and surface moisture on rainfall have been mainly documented at regional and continental scales [5,6,7,8,9], with some studies identifying impacts downwind of irrigated areas [6,10], but local effects have been more elusive. On the other hand, other studies have addressed variations of weather radar precipitation observations, which vary with low-level atmospheric conditions, such as changes in the radar reflectivity profile pattern over oceanic or continental areas [11], effects of rainfall evaporation below sub-cloud level on radar reflectivity profiles [12,13], or variations in radar reflectivity and collocated rain gauge amounts depending on low-level relative humidity (RH) [14,15]. However, in those studies, a relationship between radar reflectivity and low-level moisture has not been described.
The objective of this study is twofold. First, to assess the possible impact of irrigation on local precipitation characteristics over a semi-arid region with an irrigated and a contiguous rainfed area. Second, to study the possible effect of RH on precipitation characteristics regardless of whether the terrain is irrigated or not. The region was selected for the Land Surface Interactions with the Atmosphere over the Iberian Semi-Arid Environment (LIAISE) international field campaign in 2021 [4,14,16] and here is studied with Automatic Weather Station (AWS) observations and C-band weather radar data covering a six-year period. The analysis considers different precipitation characteristics, such as rainfall, rainfall rates, radar reflectivity profiles, and stratiform and convective regimes in comparison with AWS variables such as temperature and RH. Therefore, results reported here contribute both to a better understanding of irrigation effects on local precipitation processes and also to detail how low-level moisture impacts convective rainfall.
The rest of the article is organized as follows: Section 2 presents the region of study, the datasets, and the methodology used. Section 3 describes the results found and includes an overview of seasonal cycles of the main variables considered, the impact of rainfall onset on selected variables, an analysis of RH and vertical profiles of reflectivity over irrigated and non-irrigated areas, and finally the dependence of reflectivity with RH. Section 4 provides a discussion of the results, and Section 5 presents the conclusion.

2. Materials and Methods

2.1. Region of Study

The region of study is an agricultural area located in the eastern part of the Ebro river basin in Catalonia, in the northeast part of the Iberian Peninsula (Figure 1). This region is characterized by a semi-arid climate, classified as Cold semi-arid (BSK) in the Köppen climate classification [17,18]. Summer months (June, July, and August) are relatively dry, with only 57 mm rainfall and 7 rainy days exceeding 1 mm during the 1981 to 2010 reference period according to the Spanish Meteorological Agency (AEMET) Meteorological Station in Lleida (Figure 1) [19]. Precipitation events in the region are typically influenced by the interaction of mesoscale and synoptic circulations that reach the Ebro Valley either as western and northern flows or southern and eastern flows from the Mediterranean Sea [20,21,22,23].
An artificial water channel, the Urgell channel, which originates from the Segre River (tributary of the Ebro river), divides in two parts the region of study (Figure 1). The western part is irrigated, featuring numerous agricultural fields, while the eastern part is non-irrigated and exhibits intrinsic characteristics of the region, including rainfed fields [24].
The entire study region spans approximately a rectangular area of 20 × 30 km divided by the Urgell channel. While the eastern and western areas differ slightly in size, both can be considered relatively flat, with a gentle slope (lower than 1%) descending from east to west, resulting in low orographic impact on local meteorological conditions. Despite the absence of significant orographic effects, variations in meteorological variables, such as temperature, RH, and water vapour mixing ratio (hereafter mixing ratio) were observed between the two areas, as discussed in Section 3. Part of these differences can be attributed to inherent characteristics of the regional climatology, which exhibits a zonal trend with increased aridity (higher temperature and lower rainfall) from east to west. Other differences, as recently indicated by [25,26], are a consequence of artificial irrigation practices, mostly carried out during summer.

2.2. Datasets

2.2.1. Surface Automatic Weather Stations

We used data provided by automatic weather stations (AWS) of the Meteorological Service of Catalonia, to which different types of quality control procedures are routinely performed, including consistency and homogeneity tests [27,28]. Four AWS in the region of study were examined, two of them in the irrigated part (IR1: Mollerussa, IR2: Castellnou de Seana) and the other two in the non-irrigated area (NI1: Tàrrega, NI2: el Canós)—see blue dots and red dots, respectively, in Figure 1. The data included the mean values of temperature, pressure, RH, and accumulated rainfall, all at 30 min intervals and also 1 min rainfall rates, from 2016 to 2021.
To take into account that the rainfall field may present a high degree of spatial variability and the fact that the distance between the two AWS in each area is slightly different, the maximum value of rainfall of each AWS is taken at each time to characterize the precipitation of each area. This sampling approach is adopted to ensure that the spatial representativity of rainfall is not biased due to the locations of the AWSs.
Finally, another variable used in the study, the mixing ratio w (expressed in g/kg), was calculated from temperature, pressure, and RH AWS observations with the following equation:
w = 0.622 ∗ E s R H 100 P − E s R H 100 ∗ 1000
where RH is expressed in %, and P and Es are the atmospheric pressure and the saturation vapor pressure, both expressed in hPa. The latter is computed using [29]:
E s = 6.112 ∗ e 17.67 ∗ T 243.5 + T
where T is the air temperature in Celsius.

2.2.2. C-Band Weather Radar Data

Radar data were provided by the Meteorological Service of Catalonia C-band weather radar network [30,31]. The dataset comprises a composite of Constant Altitude Plan Position Indicator (CAPPI) [32] radar reflectivity product with 10 vertical levels, 1 km vertical resolution, 2 × 2 km horizontal resolution available every 6 min, discarding radar reflectivity values below −5 dBZ to remove noisy pixels. The reflectivity composite is produced with observations from the four radars shown in Figure 1, which are single-polarization weather radars. The composite product minimizes problems of topographic beam blockage and also the possibility of signal attenuation due to heavy rainfall, which may be relevant in C-band, as illustrated by [33]. The radar data quality is routinely monitored, including both antenna-pointing accuracy and receiver calibration, using solar interferences as external references and other automated procedures [31,33]. On the other hand, over the region of study, the CAPPI 1 km is about 700 m above ground level where the AWSs are located, considering the closest weather radar contributing to the composite product (about 25 km from the region of study).
The radar data used span the same period as the AWS. Each radar pixel was processed to classify the rainfall regime according to the methodology developed by [34]. This approach was originally designed to study rainstorms in a tropical region with an S-band meteorological radar, so some parameters were adapted for our study (see Appendix A Table A1). The methodology, using a single-level CAPPI radar reflectivity data, provides a classification of six different rainfall types: Stratiform Rain, Convective Rain, Mixed, Iso Convective Fringe, Convective Core, and Weak Echo. Here, we simplified these categories into Convective (grouping the original Convective Rain, Iso Convective Fringe, and Convective Core) or Non-convective (the other types).
The rain rate calculated from the radar data were obtained with a power-law Z-R relationship:
Z = a R b
which provides rainfall rate R (mm/h) from radar reflectivity Z (mm6 m−3), considering a = 376 and b = 1.46 according to previous local studies [35].
Given the different nature of weather radar observations (areal measurements) and AWS (point measurements), to make a consistent comparison between them, groups of 3 × 3 radar pixels centred on each AWS have been considered, as in similar previous studies [36,37,38]. In particular, the median value of radar reflectivity was selected for each 3 × 3 pixel group; so, in the following radar reflectivity samples, refer to these median values. Moreover, we checked that the radar estimates and rain gauges had similar characteristics in the irrigated area and non-irrigated areas in terms of radar beam blockage or beam overshooting [39,40].

2.3. Precipitation Events

An event or episode of precipitation for the irrigated and non-irrigated areas is defined here differently for rain gauge and weather radar data considering the different nature of each observation type as explained previously. For rain gauge data, there must be a previous dry period of at least 24 consecutive hours followed by a minimum value of 0.1 mm. For the radar data, no reflectivity observations exceeding 5 dBZ during 24 consecutive hours at the 3 × 3 pixel group must be present, and two consecutive radar reflectivity observations must be equal or greater than 5 dBZ in the lowest three CAPPI levels (1000 m, 2000 m, and 3000 m)—Table 1 summarizes these conditions. Note that the previous conditions imply that two consecutive precipitation events are separated by a Minimum Inter-Event Time (MIT), as in previous studies dealing with rain gauge data analysis [41,42,43,44]. Here, we fixed the MIT to 24 h to include at least one diurnal cycle between events, which was deemed appropriate to ensure that rainfall events can be considered independent.

2.4. Statistical Methods

Statistical analyses included the use of a Welch’s Student t-test for normal distributions and a Wilcoxon test for non-normal distributions and paired data analysis [45]. Additionally, a Wald test was applied for linear regression. Significance in differences was determined based on p-values, considering them significant if they are less than 0.05. Finally, to ensure that events observed with radar data are not duplicated, in case a precipitation event occurs in more than one 3 × 3 pixel group during the same day, only one precipitation pixel is selected randomly.

3. Results

3.1. Seasonal Cycles Overview

An overview of the seasonal cycles of selected variables provides a framework to examine in more detail in the following subsections the link between rainfall and RH. Figure 2 shows the seasonal cycles from 2016 to 2022 (median per season) of the maximum and minimum daily temperature, the RH, the seasonal rainfall, and the mixing ratio (w).
Figure 2 reveals that the behaviour of the variables recorded on AWSs in both areas are relatively similar, with consistently higher temperature ranges in the irrigated area, reflected in warmer (colder) maximum (minimum) temperatures. Differences between the irrigated and non-irrigated areas are dominated by the existence of a zonal gradient that increases semi-arid conditions (drier and warmer) from east to west, as expected from the climatology of the region of study. Similarly, both RH and w consistently show higher values in the irrigated area, with differences becoming more pronounced and significant, especially during the summer months when irrigation is active, the only exception being w in winter, when values are very low in both areas and the median is slightly higher in rainfed area.
On the other hand, there is no significant difference in the accumulated seasonal rainfall between the irrigated and non-irrigated areas. Year to year variability, controlled by predominant synoptic and mesoscale conditions, exhibits a substantial range as indicated by the long boxplot whiskers.
The seasonal 30 min average rainfall from rain gauges and weather radar is shown in Figure 3, where 3 × 3 pixel domains for the weather radar data around each AWS are considered. Despite the fact that there are slight differences between the irrigated and non-irrigated areas, they are not statistically significant. Discrepancies in average rainfall between rain gauge and radar can be attributed to the different measuring principles of each instrument, i.e., rain gauges provide point measurements and radar rainfall rates averaged over a 2 × 2 km2 pixel. Regarding maximum 30 min amounts and convective and non-convective rainfall amounts based on the classification described in Section 2, results are also quite similar between the irrigated and non-irrigated areas, with no significant statistical differences between the two areas.

3.2. RH Distribution

Figure 4 shows the histograms of frequencies of RH for each station during the summer months, first considering all data (Figure 4a, top row) and then only data recorded 2 h before precipitation (Figure 4b, bottom row). The top row displays a higher RH median at the irrigated sites (IR1 and IR2) compared to the non-irrigated sites (NI1 and NI2), with a maximum difference of 8 percentage units. However, it is crucial to highlight that, on several occasions, the non-irrigated sites also register high values of RH.
Figure 4b’s median values are much closer (differences of about 2 percentage units), indicating that the irrigation is no longer relevant in case of rain. Moreover, it clearly indicates that there is variability in both areas and that the precipitation initiation occurs under a wide range of possible values, with instances of both low and high levels of RH, suggesting that the irrigated sites do not always have higher RH compared to the non-irrigated sites.
Moreover, to further evaluate differences in RH within the same precipitation episodes (i.e., when there is precipitation at two stations with a maximum time difference of 24 h), Figure 5 shows the RH differences between two given stations 2 h before the precipitation starts at each one. As expected, the distributions systematically display medians with higher RH at irrigated sites. However, the maximum difference is only 5 percentage units, and in all subpanels, there are cases where the non-irrigated sites have higher RH than the irrigated ones.
Precipitation systems occurring over the region of study usually do not form locally but arrive already formed, which might explain why there are no substantial differences in the precipitation between the irrigated and non-irrigated sites as precipitation systems modify surface conditions, as seen before. The variability in RH observed in both irrigated and non-irrigated areas suggests that, at times, the RH is higher in one area or the other.

3.3. Impact of Precipitation Onset

Previous subsections indicated that, despite the substantial differences in temperature and RH observed between the irrigated and rainfed areas, before the rainfall onset, these differences disappear, and we found similar precipitation characteristics in terms of rainfall amounts and average rain rates over the two areas. To further examine possible differences between precipitation characteristics, we focus here on the evolution of selected variables before and after the rainfall onset, including RH, mixing ratio, and temperature, which are usually correlated with precipitation—see [46,47,48,49] for an example. This analysis was performed over the precipitation events defined in Section 2.3, which means that 24 h before the event, there was no precipitation, so the effect of rainfall onset, not affected by an immediate previous rainfall event, can be examined in detail.
Figure 6 shows the evolution of the selected variables 8 h before and 6 h after rainfall onset at the four AWS considered, plotting the mean value and its standard error to show the temporal variability at each time step. It highlights an abrupt shift in all variables at the onset of rainfall, marked by a significant drop in temperature and an increase in the mixing ratio and RH. This transition occurs within the 30 min period preceding the rain and continues for the first 30 min after the rain onset, when the variability of the selected variables diminishes (Figure 6). Interestingly, afterwards, the variability increases again. In the case of RH, after this transition period, values remain relatively high as a result of the combined effect of temperature and mixing-ratio evolutions.
Appendix B Figure A1 shows a similar plot as Figure 6 but explicitly plotting the average and their standard errors of the selected variables for all four stations considered here instead of the average values of all stations shown above. Systematic differences in the irrigated and non-irrigated areas are shown before the rainfall onset, but they are smoothed afterwards. This result suggests that if potential variations in rainfall are attributed to differences in ground-level RH, they may be negligible shortly after the rain starts. After the first 30 min of rainfall, the RH reaches a constant and high value (approximately between 75% and 80% in the 2 h after rainfall onset) very different from the conditions before the precipitation starts (55% to 60%). Based on this fact and the results described in Section 3.2, it can be stated that despite the fact that the RH is generally clearly influenced by irrigation, under rainfall conditions, this influence temporally vanishes, and what is then really relevant is the occurrence of rainfall and not if an irrigated or non-irrigated site is considered. Additionally, this suggests that the four AWS and 3 × 3 radar pixel areas must be treated independently with no aggregation or averaging between them; otherwise, we could mix distinct characteristics of RH since each site may have a different rainfall onset time. As a result, in the subsequent subsections, this study focuses on the mean radar reflectivity of the 3 × 3 pixels around each AWS for the first 30 or 180 min of precipitation of each rainfall event. This study will exclusively concentrate on the summer period to ensure no radar bright band effects are present in the first levels of the CAPPI product. The radar bright band, characterized by an apparent increase in radar reflectivity on stratiform precipitation, is caused by the melting of precipitation particles close to the freezing level—see [32,50] for more details.

3.4. Vertical Profiles of Reflectivity

The averages of the first 180 and 30 min of precipitation of each summer episode for non-convective and convective profiles are shown in Figure 7, grouped according to ground-level RH quartile of each profile. The reflectivity profiles are rather similar for non-convective cases, with 1 km values not exceeding 10 dBZ and with no apparent differences in the lower levels depending on the RH, neither for the first 180 min nor the 30 min. On the other hand, the convective regime shows higher reflectivity values (exceeding 20 dBZ at 1 km) than the non-convective regime. Moreover, there is an important difference in the first levels of reflectivity between the reflectivity averaged over the first 180 min and 30 min periods. Interestingly, the 30 min average reflectivity values present an increase in the lower levels as the RH increases, revealing a dependence of the reflectivity on RH. On the other hand, if the average is made grouping the first 180 min, this dependence is smoothed and nearly vanishes, showing that the possible differences related to the humidity of the ground are only present during the first minutes of precipitation.

3.5. Z and ΔZ, RH Relation

In this section, the possible existence of a linear relationship between Z and RH is examined for each 3 × 3 pixel group around each site, without distinguishing whether the area is irrigated or not according to results of previous sections. It is assumed here that the AWS RH is representative of ground-level RH of each 3 × 3 radar pixel group. To avoid the possible influence of bright band effects, we focus exclusively on the two lower CAPPI levels (1 and 2 km) during the summer months. Although reflectivity values are plotted in the figures in dBZ, calculations of p-values and r values are performed considering Z units (mm6 m−3) to avoid spurious effects and then converted again to dBZ—see [51] for a discussion on this topic. Moreover, to ensure a robust statistical analysis mitigating duplicated precipitation events described in Section 2.3, a random selection is performed 10,000 times to evaluate the p-value of the regression line and its median slope, if applicable.
To assess the potential dependence of Z with RH, the difference ΔZ between Z at 2 km and Z at 1 km (Z2km–Z1km) is examined in relation to the RH (Figure 8). Additionally, to evaluate the potential increase in Z with RH, the correlation between the lowest level of Z (Z1km) and RH is also explored (Figure 9).
Figure 8 displays the scatter plot and regression line between ΔZ and RH, revealing that all p-values exceed 0.05. The absence of a linear relationship between those variables suggests that no signals of evaporation related to RH exist either for non-convective or convective cases. It is noteworthy that all the 138 episodes present some non-convective cases in one of the four sites of study, and 97 episodes have at least one convective case at one of the sites.
More interestingly, Figure 9 shows that there is a linear dependence of the first 30 min of convective reflectivity in the lowest level and the RH 2 h before rainfall onset. From the 10,000 random sets, more than 95% have a p-value less than 0.05, and all of them present a positive Pearson coefficient. The mean slope presented in Figure 9 has a value of 0.10 dBZ/RH. However, with an r-value of 0.27, the relationship between the two variables can be characterized as weak. This suggests that there is some correlation, but it is not particularly strong.

4. Discussion

The relatively limited impact of irrigation upon local precipitation characteristics reported here, restricted to a linear relation between low-level radar reflectivity and ground-level RH in convective precipitation during the start of summer rainfall events, may not be generalized to other regions or spatial and temporal scales where other factors might play major roles. In particular, the reduced size of the area of study may have hampered the detection of a clear impact of irrigation on local precipitation conditions, despite an effect of RH being detected. Previous related studies found some effects of irrigation on precipitation processes, resulting in positive or negative feedback (i.e., an increase or decrease or rainfall) locally or, mostly, in neighbouring areas.
For example, Ref. [52] found boundary layer convergence lines detected with weather radar observations linked to sharp surface contrasts caused by irrigation on an arid region crossed by the Yellow River in China which, in subsequent studies, could be associated with impacts at synoptic scale [52] or, specifically, with convective initiation [53]. Ref. [54], using a new irrigation parameterization for the WRF model, studied its effect in the Po valley (northern Italy) and found a slight increase in the precipitation over the region consistent with the observations. Similarly, Ref. [26] studied, with the WRF model, a two-week summer period over the same LIAISE region considered in the present study, comparing simulations with and without irrigation parameterization, and the results indicated improvements in ground-level variables, such as temperature and RH, but little impact on local rainfall. The latter and other studies [1,6,10,54] found an increase in CAPE and a decrease in the Lifting Condensation Level caused by the increase in low-level moisture, leading to more intense convection.
This invigoration of convection is consistent with the increase in radar reflectivity with ground-level RH during the onset of summer convective precipitation reported here which, to the best of our knowledge, has not been described before in the literature. However, although the correlation found was statistically significant, its relatively low r value (0.27) suggests that other factors likely influence the relationship between Z and RH in the semi-arid environment studied, such as evaporation of precipitation below cloud base as discussed by [12] or [13]. We are aware that, with the observational datasets used, no specific effect of irrigation has been determined upon local precipitation characteristics, but an effect of RH upon convective precipitation at short temporal scale was found, independently, if irrigated or non-irrigated sites were considered.

5. Conclusions

Results from the examination of the agricultural LIAISE region of study using C-band radar observations and automatic weather station data over the irrigated and non-irrigated areas from 2016 to 2022 can be summarized as follows:
  • Clear seasonal cycles of irrigation influence on RH and mixing ratio (w) during the summer season are found, with both RH and w significantly higher in irrigated areas.
  • In summer, when irrigation is most important, the difference in maximum temperatures is not statistically significant, unlike the rest of the year when the climatological regional zonal thermal trend (with increasing temperatures from east to west) dominates. Therefore, irrigation tends to smooth the climatological thermal difference between the two areas.
  • Differences in precipitation amounts and 30 min average rainfall between the irrigated and non-irrigated areas are not significant in any season, using either rain gauge data or weather radar estimates. The irrigation impact on these precipitation variables is thus negligible.
  • The distribution of RH differences between irrigated and non-irrigated areas is highly variable, and there is not necessarily higher humidity in the irrigated area.
  • The onset of rainfall quickly modifies previous conditions of RH, w, and temperature. Examining only the 2 h period before rainfall and the next 30 min period, the RH increases 26%, w increases 10%, and the temperature decreases 10%.
  • No relationship is found in the radar reflectivity difference ΔZ between the two lowest CAPPI levels (2 km and 1 km) and RH for neither convective nor stratiform rainfall.
  • A linear relationship is detected between the summer convective rainfall reflectivity at the lowest CAPPI level and RH (median slope of 0.10 [dBZ]/[%]) during the first 30 min of rainfall. Note that this result does not imply an effect of irrigation on precipitation but an effect of the low-level moisture.

Author Contributions

Conceptualization, F.P. and J.B.; methodology, F.P. and J.B.; formal analysis, F.P.; investigation, F.P. and J.B.; data curation, F.P.; writing—original draft preparation, F.P. and J.B.; writing—review and editing, E.P., J.B., F.P., M.U. and T.R.; visualization, F.P.; supervision, J.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research was partly funded by the projects WISE-PreP (RTI2018-098693-B-C32), ARTEMIS (PID2021-124253OB-I00 MINECO/FEDER) and the Water Research Institute (IdRA) of the University of Barcelona and AGAUR grant 2022 FISDU 00365.

Data Availability Statement

Data used in this study are available from the Meteorological Service of Catalonia (smc.meteocat@gencat.cat).

Acknowledgments

We thank Francisco-Javier Ruiz (Technical University of Catalonia) for valuable suggestions regarding the statistical analysis and C-Band radar data, and automatic weather station data provided by the Meteorological Service of Catalonia.

Conflicts of Interest

The authors declare no conflicts of interest.

Appendix A

In this Appendix a list of parameters used in the precipitation regime classification is given in the table below. The parameters were obtained after the analysis of the year 2020.
Table A1. List and description of input parameters in the weather radar based precipitation regime classification methodology proposed by [34] showing values used in this study.
Table A1. List and description of input parameters in the weather radar based precipitation regime classification methodology proposed by [34] showing values used in this study.
Parameter [Units]DescriptionValue
Zth [dBZ]Reflectivity threshold at or above which echoes are classified as convective30
Rbg [km]Radius within which background reflectivity is computed5
a [dBZ]Factor for comparing echo to background reflectivity; see Equation (1) in [34] 20
b [dBZ]see Equation (1) in [34]40
Rconv [km]Maximum radius around convective core for possible uncertain classification5
Zconv [dBZ]Minimum dBZ required for Rconv to apply40
Zweak [dBZ]Minimum dBZ for classification as not weak echo5
Zshallow [dBZ]Minimum dBZ for classification as convective for objects with area less than Amed10
Alow [km2]Minimum areal coverage of echo object for classification as convective or stratiform8
Amed [km2]Maximum areal coverage of echo object for allowing Znewth = Zshallow8
Ahigh [km2]Minimum areal coverage of echo object for assigning Znewth = Zth10

Appendix B

In this Appendix, a plot of selected variables is shown to illustrate their evolution regarding the onset of rainfall during the summer season. Here, the plot explicitly shows values of individual stations in the irrigated (IR1 and IR2) and non-irrigated (NI1 and NI2) areas.
Remotesensing 17 00439 g0a1
Figure A1. Impact of rainfall onset during summer season (2016 to 2021) on (a) mixing ratio (w), (b) temperature (T), and (c) relative humidity (RH) for the selected AWS in the irrigated (IR1 and IR2) and non-irrigated (NI1 and NI2) areas. Plots show for each variable the mean value (thick line), standard error (shaded area), rainfall onset (dashed vertical red line), and 2 h before onset and 30 min before and after rainfall onset (dashed vertical black lines).
Figure A1. Impact of rainfall onset during summer season (2016 to 2021) on (a) mixing ratio (w), (b) temperature (T), and (c) relative humidity (RH) for the selected AWS in the irrigated (IR1 and IR2) and non-irrigated (NI1 and NI2) areas. Plots show for each variable the mean value (thick line), standard error (shaded area), rainfall onset (dashed vertical red line), and 2 h before onset and 30 min before and after rainfall onset (dashed vertical black lines).
Remotesensing 17 00439 g0a1

References

  1. Pielke, S. Influence of the Spatial Distribution of Vegetation and Soils on the Prediction of Cumulus Convective Rainfall. Rev. Geophys. 2001, 39, 151–177. [Google Scholar] [CrossRef]
  2. McDermid, S.; Nocco, M.; Lawston-Parker, P.; Keune, J.; Pokhrel, Y.; Jain, M.; Jägermeyr, J.; Brocca, L.; Massari, C.; Jones, A.D.; et al. Irrigation in the Earth System. Nat. Rev. Earth Environ. 2023, 4, 435–453. [Google Scholar] [CrossRef]
  3. Brooke, J.K.; Best, M.J.; Lock, A.P.; Osborne, S.R.; Price, J.; Cuxart, J.; Boone, A.; Canut-Rocafort, G.; Hartogensis, O.K.; Roy, A. Irrigation Contrasts through the Morning Transition. Q. J. R. Meteorol. Soc. 2024, 150, 170–194. [Google Scholar] [CrossRef]
  4. Boone, A.; Bellvert, J.; Best, M.; Brooke, J.K.; Canut-Rocafort, G.; Cuxart, J.; Hartogensis, O.; Le Moigne, P.; Miró, J.R.; Polcher, J.; et al. The Land Surface Interactions with the Atmosphere over the Iberian Semi-Arid Environment (LIAISE) Field Campaign. J. Eur. Meteorol. Soc. 2025, 2, 100007. [Google Scholar] [CrossRef]
  5. Koster, R.D.; Suarez, M.J.; Higgins, R.W.; Van den Dool, H.M. Observational Evidence That Soil Moisture Variations Affect Precipitation. Geophys. Res. Lett. 2003, 30, 1241. [Google Scholar] [CrossRef]
  6. Douglas, E.M.; Beltrán-Przekurat, A.; Niyogi, D.; Pielke, R.A.; Vörösmarty, C.J. The Impact of Agricultural Intensification and Irrigation on Land-Atmosphere Interactions and Indian Monsoon Precipitation—A Mesoscale Modeling Perspective. Glob. Planet. Chang. 2009, 67, 117–128. [Google Scholar] [CrossRef]
  7. Findell, K.L.; Gentine, P.; Lintner, B.R.; Kerr, C. Probability of Afternoon Precipitation in Eastern United States and Mexico Enhanced by High Evaporation. Nat. Geosci. 2011, 4, 434–439. [Google Scholar] [CrossRef]
  8. Taylor, C.M.; De Jeu, R.A.M.; Guichard, F.; Harris, P.P.; Dorigo, W.A. Afternoon Rain More Likely over Drier Soils. Nature 2012, 489, 423–426. [Google Scholar] [CrossRef] [PubMed]
  9. Guillod, B.P.; Orlowsky, B.; Miralles, D.G.; Teuling, A.J.; Seneviratne, S.I. Reconciling Spatial and Temporal Soil Moisture Effects on Afternoon Rainfall. Nat. Commun. 2015, 6, 6443. [Google Scholar] [CrossRef] [PubMed]
  10. Douglas, E.M.; Niyogi, D.; Frolking, S.; Yeluripati, J.B.; Pielke, R.A.; Niyogi, N.; Vörösmarty, C.J.; Mohanty, U.C. Changes in Moisture and Energy Fluxes Due to Agricultural Land Use and Irrigation in the Indian Monsoon Belt. Geophys. Res. Lett. 2006, 33, L14403. [Google Scholar] [CrossRef]
  11. Liu, C.; Zipser, E.J. Why Does Radar Reflectivity Tend to Increase Downward toward the Ocean Surface, but Decrease Downward toward the Land Surface? J. Geophys. Res. Atmos. 2013, 118, 135–148. [Google Scholar] [CrossRef]
  12. Xie, X.; Evaristo, R.; Troemel, S.; Saavedra, P.; Simmer, C.; Ryzhkov, A. Radar Observation of Evaporation and Implications for Quantitative Precipitation and Cooling Rate Estimation. J. Atmos. Ocean. Technol. 2016, 33, 1779–1792. [Google Scholar] [CrossRef]
  13. Kumjian, M.R.; Prat, O.P.; Reimel, K.J.; van Lier-Walqui, M.; Morrison, H.C. Dual-Polarization Radar Fingerprints of Precipitation Physics: A Review. Remote Sens. 2022, 14, 3706. [Google Scholar] [CrossRef]
  14. Song, Y.; Han, D.; Zhang, J. Radar and Rain Gauge Rainfall Discrepancies Driven by Changes in Atmospheric Conditions. Geophys. Res. Lett. 2017, 44, 7303–7309. [Google Scholar] [CrossRef]
  15. Dai, Q.; Yang, Q.; Han, D.; Rico-Ramirez, M.A.; Zhang, S. Adjustment of Radar-Gauge Rainfall Discrepancy Due to Raindrop Drift and Evaporation Using the Weather Research and Forecasting Model and Dual-Polarization Radar. Water Resour. Res. 2019, 55, 9211–9233. [Google Scholar] [CrossRef]
  16. Boone, A.; Best, M.; Cuxart, J.; Polcher, J.; Quintana, P.; Bellvert, J.; Brooke, J.; Canut-Rocafort, G.; Price, J. Land Surface Interactions with the Atmosphere over the Iberian Semi-Arid Environment (LIAISE). GEWEX News 2019, 29, 8–10. [Google Scholar] [CrossRef]
  17. Peel, M.C.; Finlayson, B.L.; McMahon, T.A. Updated World Map of the Köppen-Geiger Climate Classification. Hydrol. Earth Syst. Sci. 2007, 11, 1633–1644. [Google Scholar] [CrossRef]
  18. Peinó, E.; Bech, J.; Polls, F.; Udina, M.; Petracca, M.; Adirosi, E.; Gonzalez, S.; Boudevillain, B. Validation of GPM DPR Rainfall and Drop Size Distributions Using Disdrometer Observations in the Western Mediterranean. Remote Sens. 2024, 16, 2594. [Google Scholar] [CrossRef]
  19. Agència Estatal de Meteorologia—AEMET, Standard Climate Values. Available online: https://www.aemet.es/en/serviciosclimaticos/datosclimatologicos/valoresclimatologicos (accessed on 12 October 2024).
  20. Bech, J.; Arús, J.; Castán, S.; Pineda, N.; Rigo, T.; Montanyà, J.; van der Velde, O. A Study of the 21 March 2012 Tornadic Quasi Linear Convective System in Catalonia. Atmos. Res. 2015, 158–159, 192–209. [Google Scholar] [CrossRef]
  21. Navarro, A.; García-Ortega, E.; Merino, A.; Sánchez, J.L. Extreme Events of Precipitation over Complex Terrain Derived from Satellite Data for Climate Applications: An Evaluation of the Southern Slopes of the Pyrenees. Remote Sens. 2020, 12, 2171. [Google Scholar] [CrossRef]
  22. Gonzalez, S.; Callado, A.; Werner, E.; Escribà, P.; Bech, J. Coastally Trapped Disturbances Caused by the Tramontane Wind on the Northwestern Mediterranean: Numerical Study and Sensitivity to Short-Wave Radiation. Q. J. R. Meteorol. Soc. 2018, 144, 1321–1336. [Google Scholar] [CrossRef]
  23. Lunel, T.; Jimenez, M.A.; Cuxart, J.; Martinez-Villagrasa, D.; Boone, A.; Le Moigne, P. The Marinada Fall Wind in the Eastern Ebro Sub-Basin: Physical Mechanisms and Role of the Sea, Orography and Irrigation. Atmos. Chem. Phys. 2024, 24, 7637–7666. [Google Scholar] [CrossRef]
  24. Dari, J.; Brocca, L.; Modanesi, S.; Massari, C.; Tarpanelli, A.; Barbetta, S.; Quast, R.; Vreugdenhil, M.; Freeman, V.; Barella-Ortiz, A.; et al. Regional Data Sets of High-Resolution (1 and 6 Km) Irrigation Estimates from Space. Earth Syst. Sci. Data 2023, 15, 1555–1575. [Google Scholar] [CrossRef]
  25. Mangan, M.R.; Hartogensis, O.; Boone, A.; Branch, O.; Canut, G.; Cuxart, J.; de Boer, H.J.; Le Page, M.; Martínez-Villagrasa, D.; Miró, J.R.; et al. The Surface-Boundary Layer Connection across Spatial Scales of Irrigation-Driven Thermal Heterogeneity: An Integrated Data and Modeling Study of the LIAISE Field Campaign. Agric. For. Meteorol. 2023, 335, 109452. [Google Scholar] [CrossRef]
  26. Udina, M.; Peinó, E.; Polls, F.; Mercader, J.; Guerrero, I.; Valmassoi, A.; Paci, A.; Bech, J. Irrigation Impact on Boundary Layer and Precipitation Characteristics in Weather Research and Forecasting Model Simulations during LIAISE-2021. Q. J. R. Meteorol. Soc. 2024, 150, 3201–3873. [Google Scholar] [CrossRef]
  27. Llabrés-Brustenga, A.; Rius, A.; Rodríguez-Solà, R.; Casas-Castillo, M.C.; Redaño, A. Quality Control Process of the Daily Rainfall Series Available in Catalonia from 1855 to the Present. Theor. Appl. Climatol. 2019, 137, 2715–2729. [Google Scholar] [CrossRef]
  28. Prohom, M.; Domonkos, P.; Cunillera, J.; Barrera-Escoda, A.; Busto, M.; Herrero-Anaya, M.; Aparicio, A.; Reynés, J. CADTEP: A New Daily Quality-controlled and Homogenized Climate Database for Catalonia (1950–2021). Int. J. Climatol. 2023, 43, 4771–4789. [Google Scholar] [CrossRef]
  29. Petty, G.W. A First Course in Atmospheric Thermodynamics; Sundog Publishing: Madison, WI, USA, 2008; ISBN 978-0-9729033-2-5. [Google Scholar]
  30. Atencia, A.; Mediero, L.; Llasat, M.C.; Garrote, L. Effect of Radar Rainfall Time Resolution on the Predictive Capability of a Distributed Hydrologic Model. Hydrol. Earth Syst. Sci. 2011, 15, 3809–3827. [Google Scholar] [CrossRef]
  31. Altube, P.; Bech, J.; Argemí, O.; Rigo, T.; Pineda, N. Intercomparison and Potential Synergies of Three Methods for Weather Radar Antenna Pointing Assessment. J. Atmos. Ocean. Technol. 2016, 33, 331–343. [Google Scholar] [CrossRef]
  32. Fabry, F. Radar Meteorology: Principles and Practice; Cambridge University Press: Cambridge, UK, 2015; ISBN 978-1-108-46039-2. [Google Scholar]
  33. Rigo, T.; Llasat, M.C.; Esbrí, L. The Results of Applying Different Methodologies to 10 Years of Quantitative Precipitation Estimation in Catalonia Using Weather Radar. Geomatics 2021, 1, 347–368. [Google Scholar] [CrossRef]
  34. Powell, S.W.; Houze, R.A.; Brodzik, S.R. Rainfall-Type Categorization of Radar Echoes Using Polar Coordinate Reflectivity Data. J. Atmos. Ocean. Technol. 2016, 33, 523–538. [Google Scholar] [CrossRef]
  35. Cerro, C.; Codina, B.; Bech, J.; Lorente, J. Modeling Raindrop Size Distribution and Z (R) Relations in the Western Mediterranean Area. J. Appl. Meteorol. 1997, 36, 1470–1479. [Google Scholar] [CrossRef]
  36. Schleiss, M.; Olsson, J.; Berg, P.; Niemi, T.; Kokkonen, T.; Thorndahl, S.; Nielsen, R.; Ellerbæk Nielsen, J.; Bozhinova, D.; Pulkkinen, S. The Accuracy of Weather Radar in Heavy Rain: A Comparative Study for Denmark, the Netherlands, Finland and Sweden. Hydrol. Earth Syst. Sci. 2020, 24, 3157–3188. [Google Scholar] [CrossRef]
  37. Craciun, C.; Catrina, O. An Objective Approach for Comparing Radar Estimated and Rain Gauge Measured Precipitation. Meteorol. Appl. 2016, 23, 683–690. [Google Scholar] [CrossRef]
  38. Di Curzio, D.; Di Giovanni, A.; Lidori, R.; Montopoli, M.; Rusi, S. Comparing Rain Gauge and Weather RaDAR Data in the Estimation of the Pluviometric Inflow from the Apennine Ridge to the Adriatic Coast (Abruzzo Region, Central Italy). Hydrology 2022, 9, 225. [Google Scholar] [CrossRef]
  39. Trapero, L.; Bech, J.; Rigo, T.; Pineda, N.; Forcadell, D. Uncertainty of Precipitation Estimates in Convective Events by the Meteorological Service of Catalonia Radar Network. Atmos. Res. 2009, 93, 408–418. [Google Scholar] [CrossRef]
  40. Crisologo, I.; Warren, R.A.; Mühlbauer, K.; Heistermann, M. Enhancing the Consistency of Spaceborne and Ground-Based Radar Comparisons by Using Beam Blockage Fraction as a Quality Filter. Atmos. Meas. Tech. 2018, 11, 5223–5236. [Google Scholar] [CrossRef]
  41. Dunkerley, D. Identifying Individual Rain Events from Pluviograph Records: A Review with Analysis of Data from an Australian Dryland Site. Hydrol. Process. 2008, 22, 5024–5036. [Google Scholar] [CrossRef]
  42. Pérez Bello, A.; Mailhot, A.; Paquin, D.; Paquin-Ricard, D. Temperature-Precipitation Scaling Rates: A Rainfall Event-Based Perspective. J. Geophys. Res. Atmos. 2022, 127, e2022JD037873. [Google Scholar] [CrossRef]
  43. Grazzini, F.; Craig, G.C.; Keil, C.; Antolini, G.; Pavan, V. Extreme Precipitation Events over Northern Italy. Part I: A Systematic Classification with Machine-Learning Techniques. Q. J. R. Meteorol. Soc. 2020, 146, 69–85. [Google Scholar] [CrossRef]
  44. Kopp, J.; Rivoire, P.; Ali, S.M.; Barton, Y.; Martius, O. A Novel Method to Identify Sub-Seasonal Clustering Episodes of Extreme Precipitation Events and Their Contributions to Large Accumulation Periods. Hydrol. Earth Syst. Sci. 2021, 25, 5153–5174. [Google Scholar] [CrossRef]
  45. Woolson, R.F. Wilcoxon Signed-Rank Test. In Wiley Encyclopedia of Clinical Trials; John Wiley & Sons, Ltd.: Hoboken, NJ, USA, 2008; pp. 1–3. [Google Scholar] [CrossRef]
  46. Singh, M.S.; Warren, R.A.; Jakob, C. A Steady-State Model for the Relationship Between Humidity, Instability, and Precipitation in the Tropics. J. Adv. Model. Earth Syst. 2019, 11, 3973–3994. [Google Scholar] [CrossRef]
  47. Liu, W.; Dong, S.; Zheng, J.; Liu, C.; Wang, C.; Shangguan, W.; Zhang, Y.; Zhang, Y. Quantifying the Rainfall Cooling Effect: The Importance of Relative Humidity in Guangdong, South China. J. Hydrometeorol. 2022, 23, 875–889. [Google Scholar] [CrossRef]
  48. McPherson, R.A. A Review of Vegetation—Atmosphere Interactions and Their Influences on Mesoscale Phenomena. Prog. Phys. Geogr. Earth Environ. 2007, 31, 261–285. [Google Scholar] [CrossRef]
  49. Alfieri, L.; Claps, P.; D’Odorico, P.; Laio, F.; Over, T.M. An Analysis of the Soil Moisture Feedback on Convective and Stratiform Precipitation. J. Hydrometeorol. 2008, 9, 280–291. [Google Scholar] [CrossRef]
  50. Sánchez-Diezma, R.; Zawadzki, I.; Sempere-Torres, D. Identification of the Bright Band through the Analysis of Volumetric Radar Data. J. Geophys. Res. Atmos. 2000, 105, 2225–2236. [Google Scholar] [CrossRef]
  51. Warren, R.A.; Protat, A. Should Interpolation of Radar Reflectivity Be Performed in Z or dBZ? J. Atmos. Ocean. Technol. 2019, 36, 1143–1156. [Google Scholar] [CrossRef]
  52. Huang, Y.; Meng, Z.; Zhang, M. Synoptic Impacts on the Occurrence of Mesoscale Boundaries and Their Associated Convection Over an Area of Sharp Vegetation Contrast. Geophys. Res. Lett. 2022, 49, e2022GL099449. [Google Scholar] [CrossRef]
  53. Liu, H.; Meng, Z.; Zhu, Y.; Huang, Y. Convection Initiation Associated with a Boundary Layer Convergence Line over a Real-World Sharp Vegetation-Contrast Area. Mon. Weather Rev. 2023, 151, 1189–1212. [Google Scholar] [CrossRef]
  54. Valmassoi, A.; Dudhia, J.; Di Sabatino, S.; Pilla, F. Irrigation Impact on Precipitation during a Heatwave Event Using WRF-ARW: The Summer 2015 Po Valley Case. Atmos. Res. 2020, 241, 104951. [Google Scholar] [CrossRef]
Remotesensing 17 00439 g001
Figure 1. Western Europe and northeast of the Iberian Peninsula (Catalonia, green dotted rectangle), showing the C-band weather radars of the Meteorological Service of Catalonia (yellow symbols), the main locations and rivers (light blue lines), and the Urgell channel (dark blue line) dividing in two parts the region of study (magenta rectangle): the western irrigated part (with stations IR1 and IR2, in blue) and the eastern rainfed part (with stations NI1, NI2, in red). Background satellite image from Google Earth illustrates the high contrast between the irrigated and the rainfed parts of the region of study.
Figure 1. Western Europe and northeast of the Iberian Peninsula (Catalonia, green dotted rectangle), showing the C-band weather radars of the Meteorological Service of Catalonia (yellow symbols), the main locations and rivers (light blue lines), and the Urgell channel (dark blue line) dividing in two parts the region of study (magenta rectangle): the western irrigated part (with stations IR1 and IR2, in blue) and the eastern rainfed part (with stations NI1, NI2, in red). Background satellite image from Google Earth illustrates the high contrast between the irrigated and the rainfed parts of the region of study.
Remotesensing 17 00439 g001
Remotesensing 17 00439 g002
Figure 2. Seasonal cycles of selected AWS variables for irrigated (IR, blue) and non-irrigated (NI, red) areas: (a) maximum and minimum temperature, (b) relative humidity (RH), (c) seasonal rainfall, and (d) mixing ratio (w). Black stars indicate seasons with a significant difference between irrigated and non-irrigated areas (p-value < 0.05 based on the Welch’s Student t-test), whiskers show values of the distribution within 1.5 inter-quartile range and outliers are indicated by diamond symbols.
Figure 2. Seasonal cycles of selected AWS variables for irrigated (IR, blue) and non-irrigated (NI, red) areas: (a) maximum and minimum temperature, (b) relative humidity (RH), (c) seasonal rainfall, and (d) mixing ratio (w). Black stars indicate seasons with a significant difference between irrigated and non-irrigated areas (p-value < 0.05 based on the Welch’s Student t-test), whiskers show values of the distribution within 1.5 inter-quartile range and outliers are indicated by diamond symbols.
Remotesensing 17 00439 g002
Remotesensing 17 00439 g003
Figure 3. Similar to Figure 2 but for seasonal box plots of rainfall 30 min average for the irrigated (blue) and non-irrigated (red) areas from 2016 to 2022 observed with: (a) rain gauge and (b) weather radar.
Figure 3. Similar to Figure 2 but for seasonal box plots of rainfall 30 min average for the irrigated (blue) and non-irrigated (red) areas from 2016 to 2022 observed with: (a) rain gauge and (b) weather radar.
Remotesensing 17 00439 g003
Remotesensing 17 00439 g004
Figure 4. Frequency histogram of RH values in the irrigated (IR1 and IR2) and non-irrigated (N1 and NI2) AWS for summer (seasons 2016 to 2021): (a) all data, (b) only 2 h before the rainfall event. Dashed vertical lines indicate median values, explicitly given for each case.
Figure 4. Frequency histogram of RH values in the irrigated (IR1 and IR2) and non-irrigated (N1 and NI2) AWS for summer (seasons 2016 to 2021): (a) all data, (b) only 2 h before the rainfall event. Dashed vertical lines indicate median values, explicitly given for each case.
Remotesensing 17 00439 g004
Remotesensing 17 00439 g005
Figure 5. Frequency histograms of relative humidity differences (∆RH) in summer seasons (2016 to 2021) between the four AWS (IR: irrigated and NI: non-irrigated) for the 2 h period before rainfall. Paired data are considered if there is precipitation on the same day. Dashed lines indicate the median value, explicitly given for each case.
Figure 5. Frequency histograms of relative humidity differences (∆RH) in summer seasons (2016 to 2021) between the four AWS (IR: irrigated and NI: non-irrigated) for the 2 h period before rainfall. Paired data are considered if there is precipitation on the same day. Dashed lines indicate the median value, explicitly given for each case.
Remotesensing 17 00439 g005
Remotesensing 17 00439 g006
Figure 6. Impact of rainfall onset during summer seasons (2016 to 2021) on (a) mixing ratio (w), (b) temperature (T), and (c) relative humidity (RH) for observations recorded by the AWSs showing the mean value (thick magenta line), standard error (shaded area), rainfall onset (dashed vertical red line), 2 h before onset, and 30 min before and after rainfall onset (dashed vertical black lines).
Figure 6. Impact of rainfall onset during summer seasons (2016 to 2021) on (a) mixing ratio (w), (b) temperature (T), and (c) relative humidity (RH) for observations recorded by the AWSs showing the mean value (thick magenta line), standard error (shaded area), rainfall onset (dashed vertical red line), 2 h before onset, and 30 min before and after rainfall onset (dashed vertical black lines).
Remotesensing 17 00439 g006
Remotesensing 17 00439 g007aRemotesensing 17 00439 g007b
Figure 7. Median vertical profiles of reflectivity for summer rainfall episodes considering the first 180 min (top row) for (a) non-convective profiles, (b) convective profiles, and considering the first 30 min (bottom row) for (c) non-convective profiles and (d) convective profiles. Each panel shows reflectivity profiles (dashed lines) coloured in red, green, or blue according to the ground-level RH quartile, where RH_Q1, RH_Q2_Q3 and RH_Q4 mean RH values below the first quartile, between the first and third quartile, and above the third quartile, respectively.
Figure 7. Median vertical profiles of reflectivity for summer rainfall episodes considering the first 180 min (top row) for (a) non-convective profiles, (b) convective profiles, and considering the first 30 min (bottom row) for (c) non-convective profiles and (d) convective profiles. Each panel shows reflectivity profiles (dashed lines) coloured in red, green, or blue according to the ground-level RH quartile, where RH_Q1, RH_Q2_Q3 and RH_Q4 mean RH values below the first quartile, between the first and third quartile, and above the third quartile, respectively.
Remotesensing 17 00439 g007aRemotesensing 17 00439 g007b
Remotesensing 17 00439 g008
Figure 8. Scatter plots and regression lines of ΔZ (RH) for non-convective (a) and convective (b) reflectivity profiles. RH corresponds to two hours before rain onset, and ΔZ is the difference in Z between 2 and 1 km for the first 30 min of rain. The number of events and the p-value from the Wald test are shown.
Figure 8. Scatter plots and regression lines of ΔZ (RH) for non-convective (a) and convective (b) reflectivity profiles. RH corresponds to two hours before rain onset, and ΔZ is the difference in Z between 2 and 1 km for the first 30 min of rain. The number of events and the p-value from the Wald test are shown.
Remotesensing 17 00439 g008
Remotesensing 17 00439 g009
Figure 9. Scatter plots and regression lines of Z (RH). RH two hours before rain onset against the value of Z at 1 km for the first 30 min of rain: (a) only non-convective minutes and (b) only convective minutes. The number of events and the p-value from the Wald test are shown. Additionally, for cases where the p-value is significant, the components of the regression line are also presented.
Figure 9. Scatter plots and regression lines of Z (RH). RH two hours before rain onset against the value of Z at 1 km for the first 30 min of rain: (a) only non-convective minutes and (b) only convective minutes. The number of events and the p-value from the Wald test are shown. Additionally, for cases where the p-value is significant, the components of the regression line are also presented.
Remotesensing 17 00439 g009
Table 1. Conditions defining a rainfall event for rain gauge and weather radar data on one of the two (irrigated or rainfed) areas.
Table 1. Conditions defining a rainfall event for rain gauge and weather radar data on one of the two (irrigated or rainfed) areas.
AWS Rain Gauge Rainfall DataWeather Radar Reflectivity Data
-
Previous 24 h period without precipitation recorded on any of the two AWS of the area considered (irrigated or rainfed)
-
Minimum 0.1 mm rainfall recorded on one of the AWSs
-
Previous 24 h period without reflectivity values exceeding 5 dBZ over any of the 3 × 3 pixel groups at 1 km, 2 km, or 3 km CAPPI located over each AWS
-
Two consecutive radar observations equal or exceeding 5 dBZ over the 3 × 3 pixel group at 1 km, 2 km, or 3 km located over each AWS
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Polls, F.; Bech, J.; Udina, M.; Peinó, E.; Rigo, T. Local Influence of Surface Relative Humidity on Weather Radar Rainfall Observations over an Agricultural Semi-Arid Area. Remote Sens. 2025, 17, 439. https://doi.org/10.3390/rs17030439

AMA Style

Polls F, Bech J, Udina M, Peinó E, Rigo T. Local Influence of Surface Relative Humidity on Weather Radar Rainfall Observations over an Agricultural Semi-Arid Area. Remote Sensing. 2025; 17(3):439. https://doi.org/10.3390/rs17030439

Chicago/Turabian Style

Polls, Francesc, Joan Bech, Mireia Udina, Eric Peinó, and Tomeu Rigo. 2025. "Local Influence of Surface Relative Humidity on Weather Radar Rainfall Observations over an Agricultural Semi-Arid Area" Remote Sensing 17, no. 3: 439. https://doi.org/10.3390/rs17030439

APA Style

Polls, F., Bech, J., Udina, M., Peinó, E., & Rigo, T. (2025). Local Influence of Surface Relative Humidity on Weather Radar Rainfall Observations over an Agricultural Semi-Arid Area. Remote Sensing, 17(3), 439. https://doi.org/10.3390/rs17030439

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop