Deep Learning for Monitoring Agricultural Drought in South Asia Using Remote Sensing Data

Drought, a climate-related disaster impacting a variety of sectors, poses challenges for millions of people in South Asia. Accurate and complete drought information with a proper monitoring system is very important in revealing the complex nature of drought and its associated factors. In this regard, deep learning is a very promising approach for delineating the non-linear characteristics of drought factors. Therefore, this study aims to monitor drought by employing a deep learning approach with remote sensing data over South Asia from 2001–2016. We considered the precipitation, vegetation, and soil factors for the deep forwarded neural network (DFNN) as model input parameters. The study evaluated agricultural drought using the soil moisture deficit index (SMDI) as a response variable during three crop phenology stages. For a better comparison of deep learning model performance, we adopted two machine learning models, distributed random forest (DRF) and gradient boosting machine (GBM). Results show that the DFNN model outperformed the other two models for SMDI prediction. Furthermore, the results indicated that DFNN captured the drought pattern with high spatial variability across three penology stages. Additionally, the DFNN model showed good stability with its cross-validated data in the training phase, and the estimated SMDI had high correlation coefficient R2 ranges from 0.57~0.90, 0.52~0.94, and 0.49~0.82 during the start of the season (SOS), length of the season (LOS), and end of the season (EOS) respectively. The comparison between inter-annual variability of estimated SMDI and in-situ SPEI (standardized precipitation evapotranspiration index) showed that the estimated SMDI was almost similar to in-situ SPEI. The DFNN model provides comprehensive drought information by producing a consistent spatial distribution of SMDI which establishes the applicability of the DFNN model for drought monitoring.

Download Full-text

Data assimilation of in situ and satellite remote sensing data to 3D hydrodynamic lake models: a case study using Delft3D-FLOW v4.03 and OpenDA v2.4

Geoscientific Model Development ◽

10.5194/gmd-13-1267-2020 ◽

2020 ◽

Vol 13 (3) ◽

pp. 1267-1284 ◽

Cited By ~ 2

Author(s):

Theo Baracchini ◽

Philip Y. Chu ◽

Jonas Šukys ◽

Gian Lieberherr ◽

Stefan Wunderle ◽

...

Keyword(s):

Remote Sensing ◽

Data Assimilation ◽

Satellite Remote Sensing ◽

Model Performance ◽

Remote Sensing Data ◽

Lake Ecosystem ◽

Sensing Data ◽

Localization Scheme ◽

Spatiotemporal Scales

Abstract. The understanding of physical dynamics is crucial to provide scientifically credible information on lake ecosystem management. We show how the combination of in situ observations, remote sensing data, and three-dimensional hydrodynamic (3D) numerical simulations is capable of resolving various spatiotemporal scales involved in lake dynamics. This combination is achieved through data assimilation (DA) and uncertainty quantification. In this study, we develop a flexible framework by incorporating DA into 3D hydrodynamic lake models. Using an ensemble Kalman filter, our approach accounts for model and observational uncertainties. We demonstrate the framework by assimilating in situ and satellite remote sensing temperature data into a 3D hydrodynamic model of Lake Geneva. Results show that DA effectively improves model performance over a broad range of spatiotemporal scales and physical processes. Overall, temperature errors have been reduced by 54 %. With a localization scheme, an ensemble size of 20 members is found to be sufficient to derive covariance matrices leading to satisfactory results. The entire framework has been developed with the goal of near-real-time operational systems (e.g., integration into meteolakes.ch).

Download Full-text

Assessment of meteorological and agricultural droughts using in-situ observations and remote sensing data

10.5194/egusphere-egu2020-21408 ◽

2020 ◽

Author(s):

Depeng Zuo ◽

Siyang Cai ◽

Zongxue Xu ◽

Hong Yang

Keyword(s):

Remote Sensing ◽

Land Surface Temperature ◽

Land Surface ◽

Northeast China ◽

Remote Sensing Data ◽

Agricultural Drought ◽

Drought Assessment ◽

Sensing Data ◽

In Situ Observations

Most research on drought assessment adopted historical in-situ observations, however, there has been increased data availability from remote sensing during the recent years. This study utilizes the two sources of data in drought assessment. Using the historical in-situ observations, the spatiotemporal variations of meteorological drought were firstly investigated by calculating the standardized precipitation index (SPI) and standardized precipitation evapotranspiration index (SPEI) at 1, 3, 6-month time scales in Northeast China. Using remote sensing data, the combined deficit index (CDI) for agricultural drought assessment was computed based on tri-monthly sum of deficit in antecedent rainfall and deficit in monthly NDVI at land cover type and sub-type levels in the same region. In the end, the agricultural drought calculated by the CDI was evaluated against the deficit in crop yield, as well as deficit in Land Surface Temperature (LST) and Evapotranspiration (ET), in order to verify the applicability of the CDI for agricultural drought assessment in the study region. The results showed that the CDI has better correlations with the SPEI (R2=0.48) than the SPI (R2=0.05) at 3-month scales with weight factor a=0.5 in dry farming areas. The spatial pattern of the CDI showed that the area of agricultural drought increased from July to October. In addition, a significant linear correlation was found between the CDI and anomaly in annual agricultural yield (R2=0.55), and anomaly in monthly land surface temperature (R2=0.42). The results prove that the CDI calculated by remote sensing data is not only a reliable indicator for agricultural drought assessment in Northeast China, but also provides useful information for agricultural drought disaster prevention and mitigation, and water management improvement.

Download Full-text

Establishment of a Comprehensive Drought Monitoring Index Based on Multisource Remote Sensing Data and Agricultural Drought Monitoring

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing ◽

10.1109/jstars.2021.3052194 ◽

2021 ◽

Vol 14 ◽

pp. 2113-2126

Author(s):

Zhaoxu Zhang ◽

Wei Xu ◽

Zhenwei Shi ◽

Qiming Qin

Keyword(s):

Remote Sensing ◽

Remote Sensing Data ◽

Agricultural Drought ◽

Drought Monitoring ◽

Sensing Data ◽

Monitoring Index

Download Full-text

Deep Learning Estimation of Daily Ground‐Level NO 2 Concentrations from Remote Sensing Data

Journal of Geophysical Research Atmospheres ◽

10.1029/2021jd034925 ◽

2021 ◽

Author(s):

Masoud Ghahremanloo ◽

Yannic Lops ◽

Yunsoo Choi ◽

Bijan Yeganeh

Keyword(s):

Remote Sensing ◽

Deep Learning ◽

Remote Sensing Data ◽

Ground Level ◽

Sensing Data

Download Full-text

STUDY ON REMOTE SENSING MONITORING MODEL OF AGRICULTURAL DROUGHT BASED ON RANDOM FOREST DEVIATION CORRECTION

INMATEH Agricultural Engineering ◽

10.35633/inmateh-64-41 ◽

2021 ◽

pp. 413-422

Author(s):

Shao Li ◽

Xia Xu

Keyword(s):

Remote Sensing ◽

Random Forest ◽

Remote Sensing Data ◽

Agricultural Drought ◽

Data Sources ◽

Drought Monitoring ◽

Monitoring Methods ◽

Large Area ◽

Sensing Data ◽

Monitoring Model

Using remote sensing data to monitor large area drought is one of the important methods of drought monitoring at present. However, the traditional remote sensing drought monitoring methods mainly focus on monitoring single drought response factors such as soil moisture or vegetation status, and the research on comprehensive multi-factor drought monitoring is limited. In order to improve the ability to resist drought events, this paper takes Henan Province of China as an example, takes multi-source remote sensing data as data sources, considers various disaster-causing factors, adopts random forest method to model, and explores the method of regional remote sensing comprehensive drought monitoring using various remote sensing data sources. Compared with neural network, classification regression tree and linear regression, the performance of random forest is more stable and tolerant to noise and outliers. In order to provide a new method for comprehensive assessment of regional drought, a comprehensive drought monitoring model was established based on multi-source remote sensing data, which comprehensively considered the drought factors such as soil water stress, vegetation growth status and meteorological precipitation profit and loss in the process of drought occurrence and development.

Download Full-text

ESTIMATION OF SURFACE SNOW WETNESS USING SENTINEL-2 MULTISPECTRAL DATA

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-iv-5-223-2018 ◽

2018 ◽

Vol IV-5 ◽

pp. 223-228 ◽

Cited By ~ 7

Author(s):

D. Varade ◽

O. Dikshit

Keyword(s):

Remote Sensing ◽

Water Resource Management ◽

Remote Sensing Data ◽

Microwave Remote Sensing ◽

In Situ Measurements ◽

Physical Relationship ◽

Sensing Data ◽

Surface Snow ◽

Sentinel 2

Abstract. Snow cover characterization and estimation of snow geophysical parameters is a significant area of research in water resource management and surface hydrological processes. With advances in spaceborne remote sensing, much progress has been achieved in the qualitative and quantitative characterization of snow geophysical parameters. However, most of the methods available in the literature are based on the microwave backscatter response of snow. These methods are mostly based on the remote sensing data available from active microwave sensors. Moreover, in alpine terrains, such as in the Himalayas, due to the geometrical distortions, the missing data is significant in the active microwave remote sensing data. In this paper, we present a methodology utilizing the multispectral observations of Sentinel-2 satellite for the estimation of surface snow wetness. The proposed approach is based on the popular triangle method which is significantly utilized for the assessment of soil moisture. In this case, we develop a triangular feature space using the near infrared (NIR) reflectance and the normalized differenced snow index (NDSI). Based on the assumption that the NIR reflectance is linearly related to the liquid water content in the snow, we derive a physical relationship for the estimation of snow wetness. The modeled estimates of snow wetness from the proposed approach were compared with in-situ measurements of surface snow wetness. A high correlation determined by the coefficient of determination of 0.94 and an error of 0.535 was observed between the proposed estimates of snow wetness and in-situ measurements.

Download Full-text

Cloud-based Research Data Infrastructures Integrating In-Situ and Remote Sensing Data

10.5194/egusphere-egu21-14441 ◽

2021 ◽

Author(s):

Simon Jirka ◽

Benedikt Gräler ◽

Matthes Rieke ◽

Christian Autermann

Keyword(s):

Remote Sensing ◽

Remote Sensing Data ◽

Research Data ◽

Monitoring Networks ◽

Sensing Data ◽

Data Infrastructures ◽

Practical Experiences ◽

The One ◽

Cloud Infrastructures

For many scientific domains such as hydrology, ocean sciences, geophysics and social sciences, geospatial observations are an important source of information. Scientists conduct extensive measurement campaigns or operate comprehensive monitoring networks to collect data that helps to understand and to model current and past states of complex environment. The variety of data underpinning research stretches from in-situ observations to remote sensing data (e.g., from the European Copernicus programme) and contributes to rapidly increasing large volumes of geospatial data.However, with the growing amount of available data, new challenges arise. Within our contribution, we will focus on two specific aspects: On the one hand, we will discuss the specific challenges which result from the large volumes of remote sensing data that have become available for answering scientific questions. For this purpose, we will share practical experiences with the use of cloud infrastructures such as the German platform CODE-DE and will discuss concepts that enable data processing close to the data stores. On the other hand, we will look into the question of interoperability in order to facilitate the integration and collaborative use of data from different sources. For this aspect, we will give special consideration to the currently emerging new generation of standards of the Open Geospatial Consortium (OGC) and will discuss how specifications such as the OGC API for Processes can help to provide flexible processing capabilities directly within Cloud-based research data infrastructures.

Download Full-text

Remote Sensing Approaches and Related Techniques to Map and Study Landslides

Landslides - Investigation and Monitoring ◽

10.5772/intechopen.93681 ◽

2020 ◽

Author(s):

Ram L. Ray ◽

Maurizio Lazzari ◽

Tolulope Olutimehin

Keyword(s):

Remote Sensing ◽

Hazard Assessment ◽

Remote Sensing Data ◽

Remotely Sensed ◽

Landslide Monitoring ◽

In Situ Observation ◽

Remotely Sensed Data ◽

Sensing Data ◽

Stability Model

Landslide is one of the costliest and fatal geological hazards, threatening and influencing the socioeconomic conditions in many countries globally. Remote sensing approaches are widely used in landslide studies. Landslide threats can also be investigated through slope stability model, susceptibility mapping, hazard assessment, risk analysis, and other methods. Although it is possible to conduct landslide studies using in-situ observation, it is time-consuming, expensive, and sometimes challenging to collect data at inaccessible terrains. Remote sensing data can be used in landslide monitoring, mapping, hazard prediction and assessment, and other investigations. The primary goal of this chapter is to review the existing remote sensing approaches and techniques used to study landslides and explore the possibilities of potential remote sensing tools that can effectively be used in landslide studies in the future. This chapter also provides critical and comprehensive reviews of landslide studies focus¬ing on the role played by remote sensing data and approaches in landslide hazard assessment. Further, the reviews discuss the application of remotely sensed products for landslide detection, mapping, prediction, and evaluation around the world. This systematic review may contribute to better understanding the extensive use of remotely sensed data and spatial analysis techniques to conduct landslide studies at a range of scales.

Download Full-text