Study on Change of Small Area Land Water Reserves Based on

GRACE-FO Satellite

Tianhao Fei, Ya Ban, Ling Yu and Xinping Xu

Chongqing Academy of Metrology and Quality Inspection, Chongqing, China

Keywords: Water Resource, Small Area Land Water Reserves, GRACE-FO, Satellite Gravity, Precipitation; Chongqing.

Abstract: In order to monitor the changes of water resources in small regions, taking Chongqing as an example, the

temporal and spatial changes of groundwater reserves in the region are studied by using GRACE-FO gravity

satellite data, and the data are processed by Gaussian smoothing, and analyzed in combination with

precipitation data and terrain. The results show that the land water reserves in Chongqing all show obvious

seasonal changes, showing an upward trend of 1.2mm/mon in summer and autumn, and a downward trend

of - 0.9mm/mon in winter and spring; From 2002 to 2022, the change of equivalent water height of land

water reserves in Chongqing shows an upward trend, with a change trend of 0.8mm/a. There is a negative

correlation between land water reserves and precipitation in Chongqing, with a correlation coefficient of -

0.76. The northeast mountainous area with high altitude in Chongqing has a high frequency of rainfall, and

the monthly precipitation has little difference. The precipitation in other areas is seasonal; The land water

reserves in Chongqing are also related to the topography and population density. The water reserves in the

mountainous areas in the north and southeast are high, and the water reserves in the densely populated areas

in the southwest are low; The retrieved land water reserves of Chongqing are in good agreement with the

county spatial distribution data of groundwater resources in 2015. Based on GRACE-FO satellite, the

changes of land water reserves in small areas can be better retrieved.

1 INTRODUCTION

Water is the origin of all life on the earth and the

basis for human survival. With the rapid economic

development and climate change, water resources

are rapidly consumed. Even though China's

freshwater resources are in the forefront of the

world, due to the large population base, the per

capita water resources are below the world average

(LIU et al.,2017). Therefore, studying the change of

water resources is of great significance to our

survival and sustainable development (Vorosmarty

et al.,2000). Quantitative research on the change of

land water reserves is helpful to better study regional

drought, flood disasters and changes in groundwater

resources (Hu X et al.,2006). For traditional water

resource observation, hydrometeorology requires

multiple monitoring points or monitoring networks,

which is difficult to maintain after construction. Due

to the impact of topography and other conditions, the

observation can only cover local spatial scales, and

cannot provide enough spatio-temporal observation

to accurately describe the changes in land water

storage. In March 2002, the National Aeronautics

and Space Administration of the United States

(NASA) and the German Aerospace Center

launched GRACE gravity satellite, which had been

out of service until its crash in 2017. The successor

satellite, GRACE-FO, was launched in 2018,

providing a new means for retrieving changes in

land water reserves (Tapley B D et al.,2004 and

Wahr J et al.,2004). Many scholars at home and

abroad use GRACE satellite data to recover high-

precision earth gravity field information and the

research shows that the quality change of the earth's

surface and interior is often related to the change of

land water reserves in a short time scale (Chen Kun

et al.,2018). Therefore, the change of land water

reserves can be retrieved from GRACE satellite

gravity data.

In recent years, the work of using

GRACE/GRACE-FO satellite time-varying gravity

observation to study the changes of land water and

groundwater reserves has also been carried out.

Mohamed et al. Luo et al. Hu et al. preliminarily

analyzed the seasonal change of water reserves in

Fei, T., Ban, Y., Yu, L. and Xu, X.

Study on Change of Small Area Land Water Reserves Based on GRACE-FO Satellite.

DOI: 10.5220/0011946400003536

In Proceedings of the 3rd International Symposium on Water, Ecology and Environment (ISWEE 2022), pages 129-135

ISBN: 978-989-758-639-2; ISSN: 2975-9439

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

129

Many relevant studies should added in this

sentence.the Yangtze River basin using the early

GRACE observation data; Yang Yuande and others

calculated the change of global water reserves by

using the time-varying data of GRACE, and further

discussed the change of water reserves in Amazon

River, Yangtze River and other basins; Xiong

Jinghua et al. used GRACE gravity field model to

invert the temporal and spatial variation

characteristics of land water reserve anomaly in the

Pearl River basin from 2002 to 2017 in 2021; Meng

Ying et al. explored the change trend of land water

reserve anomaly (TWSA) in China's ten major

basins and its correlation with temperature,

precipitation and water use in 2021; Ni Shengnan et

al. discussed the seasonal and interannual variation

characteristics of water reserves in the Yangtze

River and Yellow River basins for many years in

2014. At the same time, by monitoring the drought

in the Yellow River basin from 2002 to 2003, they

proved the ability of GRACE satellite in monitoring

drought events.

Due to the problem of resolution of GRACE

satellite data, previous studies have often analyzed

the water reserves in large regions, and few

literatures have analyzed the changes of land water

reserves in small regions. In this paper, the gravity

field model data and GRACE data of GRACE-FO

satellite released recently are used to process the

data with Gaussian smoothing, and the relationship

between groundwater reserves and precipitation,

topography and population density is studied by

combining precipitation data and terrain analysis;

The characteristics of seasonal and interannual

variation are analyzed; Compared with the known

water reserve data, the reliability of inversion results

is verified.

2 OVERVIEW OF THE STUDY

AREA

Chongqing is located in the southwest of China,

between 105 ° 11 '- 110 ° 11' east longitude and 28 °

10 '- 32 ° 13' north latitude. It is the largest economic

center in the upper reaches of the Yangtze River and

an important land and water transportation hub in

southwest China. Its landform is composed of river

valleys, basins, hills and mountains, with high

mountains and deep valleys, crisscross ravines,

mountainous areas accounting for 76%, hills

accounting for 22%, and flat river valleys

accounting for only 2%, with an altitude of

73m~2796m, as shown in Figure 1. With a total area

of 82400 square kilometers and a permanent

population of 32124300, Chongqing is rainy in

summer and autumn and dry in spring and winter.

The climate of Chongqing is subtropical monsoon

humid climate. It is hot in summer and warm in

winter, humid and overcast, with high temperature,

long rainy season, less frost and snow, more

overcast days and high humidity. The northeast of

Chongqing is Daba Mountain, the southeast is

Wushan Mountain, Dalou Mountain and other

mountains, the middle is mainly low mountains and

hills, and the main rivers are Yangtze River, Jialing

River, Wujiang River, Fujiang River, Qijiang River,

Daning River, Apeng River, Youshui River, etc. The

main stream of the Yangtze River runs across the

whole territory from west to east, with a length of

665 kilometers, crossing the three anticlines of

Wushan Mountain, forming the famous Qutang

Gorge, Wuxia Gorge and Xiling Gorge in Hubei,

namely the world-famous Three Gorges of the

Yangtze River.

Figure 1: Overview of the study area.

3 DATA AND METHODS

3.1 Data Sources

The satellite gravity data in this paper adopts the

RL06 gravity field model (CSR GRACE/GRACE-

FO RL06 Mason Solutions (version 02)) of

GRACE-FO satellite Level 2 released by the Space

Research Center (CSR) of the University of Texas.

The highest order is 30, and the resolution of the

Mason RL06 data is 0.25 ° × 0.25 °, once a month.

ISWEE 2022 - International Symposium on Water, Ecology and Environment

130

As of June 14, 2022, Mascon RL06 results based on

GRACE-FO data have been updated to March 2022.

From April 2002 to March 2020, there are 206

periods of data in total (some data are missing).

The historical precipitation data of Chongqing

comes from the monthly precipitation data of

Chongqing from 1951 to 2021 given by NOAA

(National Oceanic and Atmospheric Administration

of the United States).

The data of groundwater resources in Chongqing

in 2015 comes from the national science and

technology basic conditions platform - national earth

system science data sharing platform - southwest

mountain

data resource point （

http://www.geodata.cn).

The population data of all districts and counties

in Chongqing by the end of 2020 are from

Chongqing Statistical Yearbook issued by

Chongqing Municipal Bureau of Statistics in 2021.

3.2 Research Method

3.2.1 Mason Method

In this paper, Mason method is used to divide the

study area into regular grids. Based on the same

quality change of each grid data, the functional

relationship between point quality change and

observation value in the grid area is established, and

the impact of quality change in different regions on

gravity field change is separated, so that the surface

quality change can be inversely obtained. For the

elastic earth model, Wahr et al. proposed in 1998

that the change of spherical harmonic coefficient of

gravitational potential and the change of load on the

earth's surface (Wahr J et al.,1998), as shown in

equation (1),

[]

(,)

cos sin (cos )

ave

lm lm lm

CmSmP

σθφ

φφ

Δ= ×

Δ+Δ



(1)

Where,

is the average radius of the earth,

ave

is the average density of the earth,

(cos )

is a

fully normalized associated Legendre polynomial,

lm、

is respectively the order and number of

spherical harmonic expansion,

is the maximum

order of spherical harmonic coefficient expansion,

is the Leff number of the Lth order load,

、

respectively the complementary latitude and

longitude,

CΔ

and

SΔ

is the normalized

variation of gravitational potential coefficient, which

is often used to express the surface quality anomaly

in the form of equivalent water height (EWT).

Therefore, GRACE gravity satellite can be used to

retrieve the changes of land water reserves.

3.2.2 Gaussian Smoothing Method

Gaussian smoothing is an image smoothing filter,

which uses normal distribution to calculate the

transformation of each pixel in the image. (Guo J Y

et al.,2010). The definition on the 2D plane is

22 2

()/(2)

(,)

Guv e

πσ

−+

(2)

Where,

(,)uv

is the coordinate of the relative

center point and

is the standard deviation. The

template coefficient of the Gaussian filter can be

obtained by discretizing the Gaussian function. In

this paper, because the Mason data of GRACE-FO

has a resolution of 0.25 °×0.25 ° grid, Gaussian

smoothing can better refine the equivalent water

height distribution of land water reserves, especially

for the junction area of the grid.

4 RESULTS AND ANALYSIS

4.1 Analysis of the Temporal and

Spatial Characteristics of Land

Water Reserves in Chongqing

In this paper, the data of the RL06 gravity field

model (CSR GRACE/GRACE-FO RL06 Mason

Solutions (version 02)) of GRACE-FO satellite

Level 2 are processed by MATLAB. Based on the

land water reserve data of the past three years, the

equivalent water height of Chongqing's monthly

average land water reserves is obtained by month

average, as shown in Figure 2. After fitting, it can be

seen that the land water reserves reached the peak in

December, increased gradually in July December,

and decreased gradually in January June. Then the

land water reserves from July to June of the next

year follow the normal distribution, and its axis of

symmetry is in December. Through linear fitting, the

change trend of equivalent water height of

Chongqing's land water reserves from July to

December is 1.2mm/mon, and from January to June

is -0.9mm/mon. Figure 3 shows the monthly

variation of equivalent water height of land water

reserves in Chongqing from 2002 to 2022. After

linear fitting, it is found that the overall trend is

upward, with a variation trend of 0.8mm/a.

Study on Change of Small Area Land Water Reserves Based on GRACE-FO Satellite

131

Figure 2: Equivalent water height of monthly average land

water reserves in Chongqing

Figure 3: Equivalent water height of land water reserves in

Chongqing in recent 20 years.

4.2 Analysis of the Temporal and

Spatial Characteristics of Land

Water Reserves in Chongqing

Based on the monthly precipitation data of

Chongqing from 1951 to 2021 given by NOAA

(National Oceanic and Atmospheric Administration

of the United States), the distribution of precipitation

in Chongqing in the past 70 years is plotted as shown

in Figure 4. It can be seen that the precipitation

distribution is normal, the peak appears in July, and

the precipitation is concentrated in June August. In

order to analyze the relationship between

Chongqing's land water reserves and precipitation, it

is obvious that, compared with the precipitation data,

There is a negative correlation between land water

reserves and precipitation. Based on the correlation

analysis between the monthly equivalent water height

retrieved from GRACE land water reserves in Figure

3 and the monthly precipitation data in recent two

years, the correlation coefficient is -0.76, and the

precipitation in August is significantly lower than

that in July and September, which is synchronous

with the increase of land water reserves. After the

water on the earth is irradiated by the sunlight, it

evaporates into water vapor into the air. When the

water vapor encounters cold air at high altitude, it

condenses into small water droplets and falls from

the air, forming precipitation. The reason is that there

is a lot of precipitation in summer, indicating that

there is a lot of water on the surface evaporating into

the air, resulting in less land water reserves. In the

dry and rainy winter, due to the temperature, the

water evaporation is low, and there is less water

vapor in the air, so the land water reserves are large.

Figure 4: Precipitation statistics of Chongqing in recent 70

years.

Figure 5 shows the spatial distribution map of

land water reserves in Chongqing in 2021 by month.

From the perspective of space, the seasonal variation

of land water reserves in the northeast of Chongqing

is not very different, while the land water reserves in

other regions of Chongqing are quite different. The

water reserves are low in March July and high in

October December. From the negative correlation

between precipitation and land water reserves, it is

known that the precipitation frequency in the

northeast of Chongqing is high, and the monthly

precipitation is not different, The precipitation in

other regions is seasonal, while the northeast is

dominated by mountains, indicating that the rainfall

frequency is higher in mountain areas with higher

altitude, which is consistent with the conclusion that

Fang Dexian et al. obtained higher rainfall frequency

in mountain areas with high altitude using the

rainfall data of the National Meteorological

Information Center from 2008 to 2016.

4.3 Relationship Between Land Water

Reserves and Population Density

After averaging the equivalent water height data of

land

water reserves retrieved from the GRACE-FO

123456789101112

Equivalent water height of average land water

reserves in recent three months（mm）

ISWEE 2022 - International Symposium on Water, Ecology and Environment

132

Figure 5: Land water reserves of Chongqing in 2021

gravity satellite data of Chongqing in the last two

years, the distribution of land water reserves in

Chongqing is obtained. Since the resolution of

GRACE-FO gravity satellite data is 0.25 ° × 0.25 °

grid distribution. After smoothing the grid data with

the Gaussian smoothing method, the distribution of

land water reserves in Chongqing is shown in Figure

Figure 6: Average equivalent water height of land water

reserves in Chongqing.

The population density distribution map of

Chongqing is shown in Figure 7. It can be seen from

Figure 1 and Figure 7 that the southeast and

northeast regions are mainly mountainous regions

with high altitude, and the middle regions are mainly

hilly regions, The terrain is relatively low, and the

population distribution is related to this. It is

concentrated in the middle of the lower terrain and

distributed along the Yangtze River basin.

Chongqing has more land water reserves in the

northeast and southeast, while the land water

reserves in the central and western regions are less.

Therefore, it can be seen that due to the impact of

human activities, the land water reserves have had a

certain impact, and the densely populated areas have

less water reserves.

4.4 Verification of Land Water

Reserves Retrieved by GRACE-FO

Satellite

Figure 8 shows the amount of groundwater resources

in Chongqing in 2015. The data comes from the

National Science and Technology Infrastructure

Platform - National Earth System Science Data

Sharing Platform - Southwest Mountain Data

Resource

Point. As shown in Figure 6, Chongqing's

Study on Change of Small Area Land Water Reserves Based on GRACE-FO Satellite

133

Figure 7: Population density distribution map of

Chongqing in 2020.

Figure 8: Population density distribution map of

Chongqing in 2020.

land water reserves in recent two years are more in

the northeast and southeast, while the land water

reserves in the central and western regions are less,

consistent with the data of known conclusions, It

shows that GRACE-FO gravity satellite can well

reverse the land water reserve data, and has certain

reliability.

5 CONCLUSIONS

In this paper, Chongqing is taken as the research

area. Based on GRACE-FO gravity satellite data, the

land water reserve changes of Chongqing from 2002

to 2022 are retrieved. The temporal and spatial

characteristics of land water reserve changes in

Chongqing are analyzed. The relationship between

land water reserves and precipitation is analyzed by

combining Chongqing precipitation data, and the

relationship between land water reserves and

population density is analyzed by combining

Chongqing population data. The following

conclusions are obtained:

Using the RL06 gravity field model data of

GRACE-FO satellite Level 2 to retrieve the land

water reserves data of Chongqing in recent three

years, the land water reserves show obvious seasonal

changes, reaching the peak in December, increasing

gradually from July to December, and decreasing

gradually from January to June. Through linear

fitting, the change trend of equivalent water height

of Chongqing's land water reserves from July to

December is 1.2mm/mon, and from January to June

is -0.9mm/mon. The monthly variation of equivalent

water height of land water reserves in Chongqing

from 2002 to 2022 shows a rising trend after linear

fitting, with a variation trend of 0.8mm/a.

There is a negative correlation between land

water reserves and precipitation in Chongqing, with

a correlation coefficient of -0.76. The precipitation

frequency in the northeast of Chongqing is high,

with little difference in monthly rainfall. The

precipitation in other regions is seasonal, while the

northeast is dominated by mountains, indicating that

the rainfall frequency in mountainous areas with

higher altitude is high.

The land water reserves in Chongqing are also

related to the terrain and population density. The

water reserves in the north and southeast mountain

areas are high. Due to the impact of human activities,

the water reserves in densely populated areas in the

southwest are low.

The land water reserves of Chongqing are more

in the northeast and southeast, while the land water

reserves in the central and western regions are less.

The comparison with the county spatial distribution

data of groundwater resources in 2015 shows that

the GRACE-FO gravity satellite can better reflect

the land water reserves data of small regions, with

certain reliability and wider use value.

It is suggested to strengthen the research on the

refined algorithm of gravity data model in small

areas in the follow-up work

ISWEE 2022 - International Symposium on Water, Ecology and Environment

134

ACKNOWLEDGEMENTS

This paper is Sponsored by Natural Science

Foundation of Chongqing, China (cstc2019jcyj-

msxmX0701, cstc2019jcyj-msxmX0783,

cstc2019jcyj-msxmX0383 and cstc2019jcyj-

msxmX0800).

REFERENCES

Liu Junguo, YANG Hong，GOSLING Simon N，et al．

（2017）. Water scarcity assessments in the past，

present，and future．Earth's future， 5(6):545-559．

Vorosmarty, Charles, I, et al.(2000). Global Water

Resources: Vulnerability from Climate Change and

Population Growth. Science, 289(5477):284-284.

Hu X，Chen J，Zhou Y，et al．(2006) Seasonal water

storage change of the Yangtze River basin detected by

GRACE． Sci China Ser D， 49( 5) :483-491．

Tapley B D, Bettadpur S, Ries J C, et al.(2004) GRACE

Measurements of Mass Variability in the Earth System.

Science, 305(5683):503-505.

Wahr J, Swenson S, Zlotnicki V, et al.(2004). Time-

variable gravity from GRACE: First results.

Geophysical Research Letters ,31(11): L11501.

Mohamed A, Sultan M, Ahmed M, et al.(2017). Aquifer

recharge, depletion, and connectivity: Inferences from

GRACE, land surface models, and geochemical and

geophysical data. Geological Society of America

Bulletin: B31460.1.

Luo Z C，Yao C L，Li Q，et al．(2016). Terrestrial

water storage changes over the Pearl River Basin from

GRACE and connections with pacific climate

variability．Geodesy and Geodynamics，7(3) :39-49．

171-179．

Wahr J ， Molenaar M ， Bryan F ． (1998). Time

variability of the Earth＇s gravity field: hydrological

and oceanic effects and their possible detection using

GRACE.J Geophys Res，103( B12) : 30205-30229．

Guo J Y

，Duan X J，Shum C K．(2010)Non-isotropic

Gaussian smoothing and leakage reduction for

determining mass changes over land and ocean using

GRACE data． Geophysical Journal International，

181(1) : 290-302．

Chen Kun, Jiang Weiguo, He Fuhong, et al. （2018）

Analysis on Change Characteristics of China's Water

Reserves Based on GRACE Data. Journal of Natural

Resources, 33 (2): 275-286

Study on Change of Small Area Land Water Reserves Based on GRACE-FO Satellite

135