Spatial and Temporal Evolution of Carbon Sink in the Three North

Forest Region Based on Remote Sensing Date

Chengxuan Jiang

College of Resources and Environment, Henan Polytechnic University, Jiaozuo 545000, China

Keywords: Remote Sensing, Casa Model, Carbon Sink, Net Primary Productivity Conversion, The Three North Region.

Abstract: The investigation is to estimate the region carbon sink in the three North areas in different time periods by

integrating forest data maps, and to further evaluate the outstanding contribution of the Three North

Shelterbelt project to forest carbon sink in recent years. The casa model is the foundation of research, using

primary quality productivity (NPP), which has a clear quantified index of forest carbon sink, to evaluate the

relationship between modern terrestrial sink for carbon and forest vegetation cover in research region.

Multiple sources of historical and existing land use, including publicly available high-resolution satellite

images and remote sensing products, statistical data from various organizations we concordance and use.

Based on the 16dMODIS data, we obtained the spatio-temporal date set and distribution of carbon sink in the

three northern shelterbelts in 2006, the NPP of the three north region range from 0 to 877.184 gC/ (m2·a),

and the vegetation coverage rate range from 0 to 80%.

1 INTRODUCTION

As the one of the most important country on earth and

the developing country with the largest carbon

emission, China has announced the climate target of

reaching the peak carbon emssions by 2030 and

achieving carbon neutrality before 2060 (Zeng et al.,

2022). The "dual carbon" action is undoubtedly a

powerful and significant measure to curb global

warming. We must act from the two directions of

carbon source and carbon sink at the same time, to

accelerate the pace of carbon neutrality. The

seemingly simple step of photosynthesis, in which

carbon is fixed and stored, plays a significant role in

ecosystems. The amount of carbon fixed by

vegetation varies widely depending on climate,

location and other reasons. Vegetation in different

regions has different characteristics, for example,

Forest is an ecosystem with a huge amount of biomass.

As a high-quality carbon sink, it can not only fix a

large amount of greenhouse gas co2, but also resist

wind sand and reduce soil erosion. An accurate

assessment of such a typical terrestrial ecosystem -

forest helps us to study the process of carbon flow and

fixation throughout the Earth system. Absorbing

about 45% of co2 emissions, the world's second

largest carbon reservoir, terrestrial ecosystems are not

only a single vegetation, but also a large number of

other types of land, such as wetlands, cultivated land

and so on. The Three-North Shelterbelt project, which

began to be built in the last century, has nearly half of

Chinese land area, and has made an indelible help and

contribution to the implementation of China's carbon

neutral plan. Due to its large area and long depth, it is

extremely difficult to assess and early warning

through field investigation and vegetation monitoring.

In 2010, China's terrestrial carbon sink was

estimated to be 198±54TgC−1. From 2019 to 2021,

China's terrestrial ecosystem carbon sink is about −

0.44 Pg C/ year. Various of ways serve as actual tool

to quantify carbon sink. Due to its fast and

convenient characteristics, Remote sensing

technology has long been widely and profoundly

applied in the assessment, monitoring and early

warning of afforestation management; Estimation of

vegetation biomass at the global scale; exploration

and research of the driving mechanism of carbon

sinks and the spatio-temporal differences in \regional

ecosystems.（Zheng et al 2017；Jiarui Dong et al

2003；Bowen Pang 2024）The casa model is based

on the integration of satellite images and remote

sensing data and vegetation productivity is often used

in the research of various regions, such as the spatial

Jiang, C.

Spatial and Temporal Evolution of Carbon Sink in the Three North Forest Region Based on Remote Sensing Date.

DOI: 10.5220/0013042700004601

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Innovations in Applied Mathematics, Physics and Astronomy (IAMPA 2024), pages 223-227

ISBN: 978-989-758-722-1

223

and temporal dynamic changes of the NPP in

grassland, Gannan; the NPP distribution of vegetation

in northern Hebei Province, and so on. casa model can

help us to carry out simple research on it

What’s more, the research has provided positive

economic and policy with influence. this including,

but not limited to assessing and evaluating forest

carbon sink and their value, and organizing the

construction and maintenance of man-made forests,

so as to achieve the co-development of man and

nature.

2 METHODOLOGY

2.1 Study Area

The Three North Shelterbelt Forest Program

construction area is located in 725 counties for 13

provinces ， beginning with Bin County in

Heilongjiang Province in the east, and extending to

northwest China, North China, Northeast China, and

the Xinjiang Production and Construction Corps,

west to Uzberi Pass in Xinjiang, north to the national

boundary, south along Tianjin, Fenhe River, Wei

River, Taohe River downstream, Burhanbuda

Mountain, Karakoram Mountain, 38◦ 430-53 ◦ 330N,

118◦ 020-135 ◦ 530E.(www.forestdata.cn)

Its length is 4480 kilometers from east to west,

and its width is 560 – 1460 kilometers in both

directions. Covering a land area totaling 4.358 million

square kilometers, which represents 45.2% of the

overall land area. The construction area of the project

from south to north across the zones of warm

temperate, temperate, and cold temperate, belongs to

the humid and semi-humid continental monsoon

climate, for the vegetation type, the construction of

farmland shelter forest as the basic framework,

multiple forest species, multiple tree species, mesh

belt, the combination of trees, agriculture, forestry

and animal husbandry inlay with each other, county

and county adjacent, mutual integration of regional

shelter forest system.

The image can help us determine the scope of the

research.

Figure1: the location of the three north region (Picture

credit: Original; Date from: Shao, Q. (2022). Three-North

Shelterbelt construction project [Data set]. National Data

Center for Ecological Sciences).

2.2 Dates

2.2.1 Data Sources

One of the primary instruments aboard the Terra

spacecraft and Aqua spacecraft is MODIS, which

stands for Moderate Resolution Imaging

Spectroradiometer. MODIS vegetation indices are

generated at various spatial resolutions at 16-day

intervals, the date on 2006s July was used in

calculatiton.

The process of data preprocessing includes

geometric correction like geolocation, image

registration, mosaic and cropping, orthographic

correction, picture fusion, cloud removal and shadow

processing, and atmospheric correction.

2.2.2 Landuse

These date comes from China's first annual land cover

product (CLCD) obtained from Landsat data. this

date is divided into several types inclouding farmland,

woodland, shrubbery, grassland, water, ice, barren,

impermeable, wetland. It selected the required date

and processed it.

These data provide an abundance of research-

useful information, allowing a thorough knowledge

of the effects of land-use change on ecosystems and

the environment. By revealing the mechanisms by

which natural and human causes influence changes in

land cover, data analysis can provide a scientific basis

for the formulation of sustainable land management

policies. Meteorological data can be used to

investigate the effects of climatic change on land use

simultaneously, and socio-economic data can be used

IAMPA 2024 - International Conference on Innovations in Applied Mathematics, Physics and Astronomy

224

to study the effects of urbanization and economic

development on land cover. These multi-perspective

analyses help us understand the complexity of land

use change more thoroughly.

2.2.3 Vegetation Carbon Sink

Aimed to estimate carbon sink，we used the date

from governmental agencies ， dividing the three

north shelter forest region.

NPP refers to the remaining part of the vegetation

after photosynthesis and respiration cancel each other.

This part is very important in the ecosystem. It

indicates that the vegetation converts light energy into

chemical energy through its own reaction, and it

relate to the health and quality of the vegetation. By

quantifying NPP, we can get a better understanding

of vegetation productivity and the entire ecosystem.

According to the formula:

PP=GPP-Ra (1)

EP=NPP-Rh=GPP-Ra-Rh (2)

The terms "Ra" and "Rh" stand for autotrophic

and heterotrophic respiration, respectively. “GPP”

which means the amount of photosynthates or total

organic carbon fixed by organisms. “NEP” is net

ecosystem productivity.

There are significant differences in NEP, which

are influenced by different land cover types. From the

perspective of types of vegetation, the annual average

NEP of forest, grassland and agricultural land is

higher than that of the latter.

2.2.4 Vegetation Data Indicators

The NDVI was caculated using the equation,and The

EVI ， an improvement of NDVI, accounts for

atmospheric scattering and land surface reflection,

resulting in increased precision in regions with

substantial vegetation coverage.where "Red" denotes

reflectance in the red portion of the spectrum and

"NIR" denotes spectral in the near-infrared

wavelength range 、 ‘Blue’ on behalf of the

reflectivity of the blue band in the spectrum.In view

of different environment factor，the other elements

are coefficients introduced to reflect different

environmental conditions. The computation of EVI is

outlined as follows:

aNDVI = （NIR-Red）/（NIR+Red）

(3)

EVI=G*（NIR-Red）/（NIR+K1*Red-

K2*Blue+L）

(4)

With the help of this formula, EVI can more

successfully reduce the effects of surface reflection

and atmospheric scattering, producing data on

vegetation cover that is more accurate. This

development is essential for researching the dynamics

of vegetation, evaluating the health of ecosystems,

and monitoring the environment. Beyond the

monitoring of a particular vegetation cover, EVI is

widely applied and is essential to research on global

climate change. Long-term EVI data series analysis

provides a scientific basis for managing ecosystems

and developing environmental protection policies by

highlighting regional and global trends and patterns

in changes in plant cover.

Furthermore, EVI has a great deal of application

value in forestry, land resource management, and

agriculture. It helps managers and farmers better

comprehend crop growth status, forest health, and

changes in land use, which improves resource use

efficiency and management decisions. Additionally,

as an enhanced vegetation index, EVI proves to be

very accurate and dependable across a range of

environmental circumstances, making it a crucial

instrument in the study of ecosystems and vegetation

monitoring. Its easy-to-use computation technique

and wide range of applications provide essential data

support and a scientific basis for comprehending and

safeguarding the Earth's ecology.

2.2.5 Driving Factor Analysis

Formerly known as Carnegie – Ames – Stanford

Approach model，the CASA model， is used for

analyzing the driving factors behind the spatio-

temporal evolution of carbon sinks in the three north

forest regions. This model is widely recognized for its

robust framework in estimating net primary

productivity (NPP) by incorporating and utilizing

remote sensing data and ecological principles. Our

study builds upon the CASA-NPP transformation

method.

In many previous studies, we have shown that the

reflection features of vegetation to infrared and near-

infrared light can effectively help us estimate

photosynthetically active radiation, and as a driving

mechanism of plant photosynthesis, it is affected by

many factors, such as the amount of biomass in the

area, the total amount of solar radiation in the field.

Moreover, the photosynthetically active radiation

also showed a good linear relationship with the

normalized vegetation difference index in a certain

range, which also proved the feasibility of the method

This is a relatively classical NPP estimation

method in the CASA model proposed by Zhu.

Spatial and Temporal Evolution of Carbon Sink in the Three North Forest Region Based on Remote Sensing Date

225

Through the measurement and calculation of NPP

data, we can effectively obtain the increment of NPP

in a large spatiotemporal range, so as to effectively

analyze the status of forest carbon sink and evaluate

the quality of the construction project of the Three-

North protective fence.

PP(

)=APAR(

)×ε(

,m) (5)

APAR(

)=SOL(

,m)×FPAR(

,m)×

(6)

FPAR(p,m)=(NDVI(p,m)-

NDVIi,min)/(NDVIi,max-

NDVIi,min)×(FPARmax-

FPARmin)+FPARmin

(7)

APAR is photosynthetically active radiation, ε is

solar energy utilization efficiency, FPAR is

photosynthesis radiation absorption ratio. “i” is the

sequence of vegetation species, “p” represents the

pixels, and “m” represents the months in the time

series.

3 RESULT

Initially ， We find the sequence occupied by

vegetation in the land use type, confirm some basic

attributes of them, deploy their static parameter files,

and import pre-treated vegetation types, ndvi time

series, climate factors, environmental factors, etc.

into the casa calculation module to obtain the original

data and import it into gis for analysing.

Then, comprehensive reclassification is carried

out on the npp results and the vegetation coverage

images at the same time, and the results are divided

into several categories according to equal proportions,

which are represented by different legends in the

images. In the vegetation coverage rate, the dark

green color indicates the high vegetation content in

the region. Similarly, in the npp results, the color can

quantitatively represent the exact npp value of the

pixel. They have a trend from fewer to more. And the

region showed an increasing from east to west.

The vegetation remote sensing image of NDVI

can reflect the change of vegetation to a certain extent,

but the image obtained by the model in this study can

more truly and effectively reflect the npp distribution

in the three northern regions combined with multi-

breadth and multi-depth influence factors

Through the comparison of forest cover and

primary quality productivity in the same year, we can

conclude that forest cover and carbon sink capacity

are highly positive correlation. That is, areas with

high forest cover usually have higher primary

productivity, which absorbs more carbon dioxide and

forms a stronger carbon sink capacity. This is a

positive and benign cycle, in which forests fix

atmospheric carbon dioxide into organic carbon

through photosynthesis. The higher the forest

coverage rate, the more obvious this carbon

fixation effect is, thus improving the regional carbon

sink capacity and making the effect of carbon

neutrality more obvious.

Figure 2: The three region vegetation coverage in 2006

(Picture credit: Original).

There are a little yellow areas or on the map,

which means there is little or no vegetation in this

region .it explains why the initial values of vegetation

coverage and NPP start at 0.

Figure 3: The Net primary productivity conversion

result

(Picture credit: Original).

At the same time, we obtained the temperature as

the most suitable growth temperature for vegetation

to fix carbon according to different time series in a

year. Finally, we lock the result between month and

July, and finally we can know the raster result.

The NPP of the three north region range from 0 to

877.184 gC/ (m2·a), and the vegetation coverage rate

range from 0 to 80%.

IAMPA 2024 - International Conference on Innovations in Applied Mathematics, Physics and Astronomy

226

4 CONCLUSION

Because of the importance of the Three Northern

regions to Chinese "dual carbon" action, it is

suggested to strengthen the continuous monitoring

and evaluation of the Three Northern shelterbelts, and

use the combination of remote sensing technology

and ground observation to obtain more detailed and

long time series data, which can also make up for the

shortcomings of the small spatiotemporal range and

fracture of the experiment, and the incomplete

accuracy of the experimental results. To study the

temporal and spatial changes of carbon sink capacity,

and actively estimate the potential of forest carbon

sink. At the same time, in forest management, the

impact of climate change should be considered, and

management action need be taken adjusting to local

natural conditions, such as disaster prevention and

mitigation, selecting drought-tolerant and water-

tolerant tree species, and avoiding planting a large

number of single tree species, so as to improve the

forest resistance and carbon sink capacity, so as to

achieve the goal of carbon neutrality as soon as

possible.

Certainly, human intervention and protective

strategies for forestry are of paramount importance.

The scenario of curbing global warming is escalating

in severity and urgency， yet behavior of remnants

of deforestation and the subsequent situation of

reduction of forest carbon sinks persist. Initiatives

such as reverting cropland to forested land and

implementing stringent control measures can people

pay attention to, coupled with penalties for unchecked

deforestation, comprise various tactics aimed at

safeguarding the ecological resources within this

expansive carbon reserve. In conclusion, extensive

research underscores the indispensable and

substantial role that the Three North regions will play

in China's, and indeed the world's, future. This role is

intrinsically linked to the sustainability of human

society, economic progression, and ultimate

prosperity.

In the whole research process, the primary

productivity and vegetation coverage of the three

northern regions were analyzed from a large spatio-

temporal range by remote sensing. Compared with

the field survey, the whole result was faster and more

intuitive. However, due to the difference in data

quality and other reasons, the limitations of the

research results were also shown. The spatio-

temporal span of the results failed to clearly show the

whole change process. Secondly, only several land

cover types were used in the study, which will have

an impact on the results. Therefore, future studies can

explore the carbon sink in the three North regions in

a finer spatio-temporal range.

REFERENCES

Liu, J., Meng, B., Ge, J., Gao, J., Yin, J., Hou, M., Feng, Q.,

& Liang, T. (2019). Spatiotemporal dynamic changes

of NPP in Gannan grassland based on CASA model.

Journal of Pratacultural Science, 28(6), 19-32.

Zhu, W.-Q., Pan, Y.-Z., & Zhang, J.-S. (2007). Estimation

of net primary productivity of Chinese terrestrial

vegetation based on remote sensing. Chinese Journal of

Plant Ecology, 31(3), 413-424.

Z., N., J., K., & Han, P. et al. (2022). Carbon neutrality in

China: Challenges and prospects. Advanced

Atmospheric Science, 39, 1229 – 1238.

https://doi.org/10.1007/s00376-022-2151-0

Chen, B., Liu, Y., Li, Z., Xu, C., & Hou, W. (2024).

Spatiotemporal carbon sequestration by forests among

counties and grids in China. Journal of Cleaner

Production, 468, 142971.

https://doi.org/10.1016/j.jclepro.2024.142971

Xu, X. (2018). Spatial distribution data set of China

monthly Vegetation Index (NDVI). Resource and

Environmental Science Data Registration and

Publication System. http://www.resdc.cn/DOI

Yang, J., & Huang, X. (2023). The 30 m annual land cover

datasets and its dynamics in China from 1985 to 2022

[Data set]. Earth System Science Data, 13(1), 3907–

3925. Zenodo. https://doi.org/10.5281/zenodo.3925

Feng, F., & Wang, K. (2021). Merging high-resolution

satellite surface radiation data with meteorological

sunshine duration observations over China from 1983

to 2017. Remote Sensing, 13(602).

https://doi.org/10.3390/rs13040602

Yuan, J., Niu, Z., & Wang, C. (2006). Vegetation NPP

distribution based on MODIS data and CASA model—

A case study of northern Hebei Province. Chinese

Geographical Science, 16, 334 – 341.

https://doi.org/10.1007/s11769-006-0334-5

Zhu, W., Zhang, J., & Pan, Y. (2007). Estimation of net

primary productivity of terrestrial vegetation in China

by remote sensing. Journal of Plant Ecology, 31(3),

413-424.

An, X., Jin, W., Long, X., Chen, S., Qi, S., & Zhang, M.

(2022). Spatiotemporal evolution of carbon storage in

Dongting Lake wetland based on remote sensing data.

Geocarto International, 37(27), 14983 – 15009.

https://doi.org/10.1080/10106049.2021.1934893

Shao, Q. (2022). Three-North Shelterbelt construction

project [Data set]. National Data Center for Ecological

Sciences.

Spatial and Temporal Evolution of Carbon Sink in the Three North Forest Region Based on Remote Sensing Date

227