Aboveground carbon biomass estimate with Physics-informed deep network Juan Nathaniel

2025-04-30 0 0 5.55MB 6 页 10玖币

侵权投诉

Aboveground carbon biomass estimate with

Physics-informed deep network

Juan Nathaniel∗

Columbia University

NY, USA

Levente J. Klein

IBM Research

Yorktown Heights, NY, USA

Campbell D. Watson

IBM Research

Yorktown Heights, NY, USA

Gabrielle Nyirjesy

Columbia University

NY, USA

Conrad M. Albrecht

German Aerospace Center

Weßling, Germany

Abstract

The global carbon cycle is a key process to understand how our climate is changing.

However, monitoring the dynamics is difﬁcult because a high-resolution robust

measurement of key state parameters including the aboveground carbon biomass

(AGB) is required. Here, we use deep neural network to generate a wall-to-wall map

of AGB within the Continental USA (CONUS) with 30-meter spatial resolution

for the year 2021. We combine radar and optical hyperspectral imagery, with a

physical climate parameter of SIF-based GPP. Validation results show that a masked

variation of UNet has the lowest validation RMSE of 37.93

1.36 Mg C/ha, as

compared to 52.30

0.03 Mg C/ha for random forest algorithm. Furthermore,

models that learn from SIF-based GPP in addition to radar and optical imagery

reduce validation RMSE by almost

10%

and the standard deviation by

40%

. Finally,

we apply our model to measure losses in AGB from the recent 2021 Caldor wildﬁre

in California, and validate our analysis with Sentinel-based burn index.

1 Introduction

Aboveground carbon biomass (AGB) is an important component to monitor carbon cycle on the

local [

] and global scale [

]. A recent state-of-the-art light detection and ranging (LiDAR)

mission from NASA’s Global Ecosystem Dynamics Investigation (GEDI) generates a global yet

non-continuous sparse measurements of vegetation parameters, including AGB estimates with a

60-meter along-track and 600-meter across-track gaps between footprints [

]. Here, we apply a

Physics-informed deep network to generate a 30-meter resolution, wall-to-wall continuous AGB

estimate in CONUS for the summertime (June-August) 2021 period. The model is trained on more

than one million GEDI footprints using a combination of radar and optical imagery [6]. In addition,

we incorporate a measure of photosynthetic intensity as captured by solar-induced ﬂuorescence

(SIF)-based gross primary production (GPP). SIF-based GPP is one of the key physical parameters

regulating AGB [

]. We validate our results using ﬁeld-based AGB observations and evaluate their

consistency across climate zones. Finally, we use our high resolution AGB map to assess the impact

of wildﬁre, in terms of how much carbon biomass had been lost to the environment.

∗Corresponding author (jn2808@columbia.edu)

Tackling Climate Change with Machine Learning: workshop at NeurIPS 2022.

arXiv:2210.13752v1 [cs.LG] 25 Oct 2022

2 Methodology

This section provides a detailed description of the data, models, and evaluation metrics used, and are

summarized in Figure 1.

Sentinel-1 data Bilinear

resampling

Sentinel-2 data Near-cloud-

free ﬁltering

SIF-based GPP

Summertime

averaging

GEDI/Field-

based AGB

Gridcell

matching

Data processing

Training

datacube

Validation

datacube

Datacube

Hyperparameter

optimization

Training

Trained model

Validation

Modeling

Figure 1: Schematic overview that summarizes our data-processing steps at the leftmost panel and

datacube generation during training and validation steps. The modeling process is iterated across

different data setups (SIF/S1/S2, S1/S2, and S2-only) in an ablation study.

2.1 Data Processing

GEDI is one of the current state-of-the-art spaceborne LiDAR missions to capture a detailed measure

of ecosystem structure [

]. However, the data generated is spatially sparse. Therefore, we attempt

to produce a dense 30-meter resolution AGB estimate using multiple input features including radar

and optical hyperspectral imagery from Sentinel-1 and Sentinel-2, as well as SIF-based GPP from

the OCO-2 mission [

]. We use the vertical co-polarization (VV) and vertical/horizontal cross-

polarization (VH) band of Sentinel-1, and the entire 12 bands of Sentinel-2. Furthermore, we produce

near-cloud-free optical imagery by taking the median value of non-cloudy pixel as deﬁned by the

scene classiﬁcation layer (SCL) of Sentinel-2. Finally, we perform grid cell geospatial matching to

combine the dataset together. The validation sites are situated in the Northwestern America [

] and

New England [11].

2.2 Model and Experimental Setup

We benchmark our deep network model against linear regressor (LR), random forests (RF) [

], and

extreme gradient boosting (XGBoost) [

] algorithms due to their reported robustness in climate-

related tasks [

]. Speciﬁcally, we implement a masked variation of UNet [

] due to the sparsity

of our target AGB variable [

]. The models are optimized using a randomized grid search and a

k-fold cross validation (k=5) approaches. We split the data into a 90-10 training-testing set, where set

here refers to pixels for ML-based models and tiles for deep networks. We use Adam optimizer [

]

with a learning rate of 0.01 and root-mean-squared error (RMSE) as our loss function. For the deep

network, we use a size 32 batching and implement a collection of image augmentations including

random horizontal and vertical ﬂipping, as well as cropping to a 512x512 image size. The full deep

network modeling workﬂow is illustrated in Figure 2.

3 Results and Discussion

This section summarizes and validates our results, showcasing the application of our AGB 30-meter

resolution (Figure 3) to assess how much carbon has been released to the environment from a major

wildﬁre event.

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

AbovegroundcarbonbiomassestimatewithPhysics-informeddeepnetworkJuanNathanielColumbiaUniversityNY,USALeventeJ.KleinIBMResearchYorktownHeights,NY,USACampbellD.WatsonIBMResearchYorktownHeights,NY,USAGabrielleNyirjesyColumbiaUniversityNY,USAConradM.AlbrechtGermanAerospaceCenterWeßling,GermanyAbstractTh...

展开>> 收起<<

Aboveground carbon biomass estimate with Physics-informed deep network Juan Nathaniel.pdf

共6页,预览2页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Aboveground carbon biomass estimate with Physics-informed deep network Juan Nathaniel

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: