remotesensing-logo

Journal Browser

Journal Browser

Geostatistics and Spatial Data Mining for Ecological Climatology

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Ecological Remote Sensing".

Deadline for manuscript submissions: closed (19 January 2024) | Viewed by 12648

Special Issue Editors

Centre for Oceans, Rivers, Atmosphere and Land Sciences (CORAL) & School of Water Resources, Indian Institute of Technology (IIT), Khargpur 721302, West Bengal, India
Interests: ecological climatology; biophysical variables; spatial biodiversity; forest cover dynamics
Special Issues, Collections and Topics in MDPI journals
Department of Remote Sensing, Birla Institute of Technology (BIT), Ranchi, India
Interests: geostatistics; remote sensing based vegetation and environmental analysis; modelling space-time vegetation dynamics; land surface phenology; vegetation and climate; landscape metrics and modelling; land cover change modelling; downscaling of satellite derived vegetation data; multi-criteria decision modelling; fuzzy logic and software development
Department of Information Technology, National Institute of Technology Karnataka (NITK) Surathkal, NH 66, Srinivasnagar, Surathkal, Mangalore, Karnataka 575025, India
Interests: geoscience and spatial data mining; spatial graph analytics

Special Issue Information

Dear Colleagues,

Scientists practicing ecological climatology now recognize that the patterns and processes of plant communities and ecosystems not only respond to weather, climate, and atmospheric compositions, but also feedback through a variety of physical, chemical, and biological processes to influence the atmosphere. The geoscientific understanding of planet Earth has given way to a new paradigm of biogeosciences. Geostatistical techniques have enabled geoscientists to commonly include spatial support (the size, geometry, scale, and orientation of the space on which observations are made) and the principled handling of different spatial concepts.

With the emergence of the era of big data science, data availability from remote sensing platforms has been preferred over model-based approaches (including geostatistics) to handle big datasets. This drive toward data-oriented approaches, including spatial data mining and machine learning, has undoubtedly brought tremendous innovation in the field of ecological climatology as well.

This Special Issue covers the development and/or evaluation of different methods and techniques for studying the responses and feedback of different physical, chemical, and biological processes and their interactions using Earth observation data. We invite you to contribute a research article to this Special Issue related to, among others, climate change, global warming, ecosystem degradation, physiological response, ecosystem resilience, ecosystem services, lidar remote sensing, spatial data mining, machine learning, and big data analytics.

Dr. Mukunda Dev Behera
Prof. Jeganathan Chockalingam
Prof. Peter M. Atkinson
Dr. Shrutilipi Bhattacharjee
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (4 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

17 pages, 5878 KiB  
Article
Predicting the Forest Canopy Height from LiDAR and Multi-Sensor Data Using Machine Learning over India
by Sujit M. Ghosh, Mukunda D. Behera, Subham Kumar, Pulakesh Das, Ambadipudi J. Prakash, Prasad K. Bhaskaran, Parth S. Roy, Saroj K. Barik, Chockalingam Jeganathan, Prashant K. Srivastava and Soumit K. Behera
Remote Sens. 2022, 14(23), 5968; https://doi.org/10.3390/rs14235968 - 25 Nov 2022
Cited by 9 | Viewed by 3697
Abstract
Forest canopy height estimates, at a regional scale, help understand the forest carbon storage, ecosystem processes, the development of forest management and the restoration policies to mitigate global climate change, etc. The recent availability of the NASA’s Global Ecosystem Dynamics Investigation (GEDI) LiDAR [...] Read more.
Forest canopy height estimates, at a regional scale, help understand the forest carbon storage, ecosystem processes, the development of forest management and the restoration policies to mitigate global climate change, etc. The recent availability of the NASA’s Global Ecosystem Dynamics Investigation (GEDI) LiDAR data has opened up new avenues to assess the plant canopy height at a footprint level. Here, we present a novel approach using the random forest (RF) for the wall-to-wall canopy height estimation over India’s forests (i.e., evergreen forest, deciduous forest, mixed forest, plantation, and shrubland) by employing the high-resolution top-of-the-atmosphere (TOA) reflectance and vegetation indices, the synthetic aperture radar (SAR) backscatters, the topography and tree canopy density, as the proxy variables. The variable importance plot indicated that the SAR backscatters, tree canopy density and the topography are the most influential height predictors. 33.15% of India’s forest cover demonstrated the canopy height <10 m, while 44.51% accounted for 10–20 m and 22.34% of forests demonstrated a higher canopy height (>20 m). This study advocates the importance and use of GEDI data for estimating the canopy height, preferably in data-deficit mountainous regions, where most of India’s natural forest vegetation exists. Full article
(This article belongs to the Special Issue Geostatistics and Spatial Data Mining for Ecological Climatology)
Show Figures

Graphical abstract

22 pages, 28292 KiB  
Article
The Time Lag Effect Improves Prediction of the Effects of Climate Change on Vegetation Growth in Southwest China
by Meng Wang, Zhengfeng An and Shouyan Wang
Remote Sens. 2022, 14(21), 5580; https://doi.org/10.3390/rs14215580 - 04 Nov 2022
Cited by 7 | Viewed by 1791
Abstract
Climate change is known to significantly affect vegetation development in the terrestrial system. Because Southwest China (SW) is affected by westerly winds and the South and East Asian monsoon, its climates are complex and changeable, and the time lag effect of the vegetation’s [...] Read more.
Climate change is known to significantly affect vegetation development in the terrestrial system. Because Southwest China (SW) is affected by westerly winds and the South and East Asian monsoon, its climates are complex and changeable, and the time lag effect of the vegetation’s response to the climate has been rarely considered, making it difficult to establish a link between the SW region’s climate variables and changes in vegetation growth rate. This study revealed the characteristics of the time lag reaction and the phased changes in the response of vegetation to climate change across the entire SW and the typical climate type core area (CA) using the moving average method and multiple linear model based on the climatic information of CRU TS v. 4.02 from 1982 to 2017 together with the annual maximum (P100), upper quarter quantile (P75), median (P50), lower quarter quantile (P25), minimum (P5), and mean (Mean) from GIMMS NDVI. Generally, under the single and combined effects of temperature and precipitation, taking the time lag effect (annual and interannual delay effect) into account significantly improved the average prediction rates of temperature and precipitation, which increased by 18.48% and 25.32%, respectively. The optimal time delay was 0–4 months when the annual delay was taken into consideration, but it differed when considering the interannual delay, and the delaying effect of precipitation was more significant than that of temperature. Additionally, the response intensity of vegetation to temperature, precipitation, and their interaction was significantly more robust when the annual delay was taken into account than when it was not (p < 0.05), with corresponding multiple correlation coefficients of 0.87 and 0.91, respectively. However, the degree of response to the combined effect of individual effects and climate factors tended to decrease regardless of whether time delay effects were taken into account. A more comprehensive analysis of the effects of climate change on vegetation development dynamics suggested that the best period for synthesizing NDVI annual values might be the P25 period. Our study could provide a new theoretical framework for analyzing, predicting, and evaluating the dynamic response of vegetation growth to climate change. Full article
(This article belongs to the Special Issue Geostatistics and Spatial Data Mining for Ecological Climatology)
Show Figures

Graphical abstract

27 pages, 9439 KiB  
Article
Geostatistical Resampling of LiDAR-Derived DEM in Wide Resolution Range for Modelling in SWAT: A Case Study of Zgłowiączka River (Poland)
by Damian Śliwiński, Anita Konieczna and Kamil Roman
Remote Sens. 2022, 14(5), 1281; https://doi.org/10.3390/rs14051281 - 05 Mar 2022
Cited by 13 | Viewed by 2025
Abstract
A digital elevation model (DEM) is an essential element of input data in the model research of watersheds. Recently, progress in measurement techniques has led to the availability of such data with high spatial resolution. Therefore, simplification of DEMs to shorten the time [...] Read more.
A digital elevation model (DEM) is an essential element of input data in the model research of watersheds. Recently, progress in measurement techniques has led to the availability of such data with high spatial resolution. Therefore, simplification of DEMs to shorten the time of their processing is a significant, but insufficiently investigated issue. This study, gradually and with various methods, carried out a great simplification of a detailed LiDAR-derived DEM. Then, the impact of that treatment on the precision of the selected elements for modeling a watershed was assessed. The simplification comprised a reduction in resolution, with the use of statistical resampling methods, namely giving an average, modal, median, minimum, maximum, or the closest value to the pixels. This process was carried out in a wide range of pixel sizes, increasing by 50% each time (from 1 m to 1.5, 2.3, 3.4, 5.1, 7.6, 11, 17, 26, 38, 58, and 86 m, respectively). The precision of the obtained DEMs and the precision of the delineation of boundaries of the watershed and watercourses were assessed. With the systematic reduction in the resolution of a DEM, its precision systematically decreased. The changes in the precision of determining the watercourses and boundaries of a watershed were irregular, ranging from being very small, to mild, to significant. A method of giving the minimum value, that was simple with regard to computing, was singled out. In the determination of both the watercourses and the boundaries of a watershed, this method produced one of the best results for the higher resolution and for the lower resolution—considerably better than the other methods tested. The research was conducted on a flat agricultural catchment, and it can be assumed that the obtained conclusions can be considered for similar cases. For catchments with different characteristics, further research is advisable. Full article
(This article belongs to the Special Issue Geostatistics and Spatial Data Mining for Ecological Climatology)
Show Figures

Figure 1

23 pages, 8513 KiB  
Article
Resilience of the Central Indian Forest Ecosystem to Rainfall Variability in the Context of a Changing Climate
by Beependra Singh, Chockalingam Jeganathan, Virendra Singh Rathore, Mukunda Dev Behera, Chandra Prakash Singh, Parth Sarathi Roy and Peter M. Atkinson
Remote Sens. 2021, 13(21), 4474; https://doi.org/10.3390/rs13214474 - 08 Nov 2021
Cited by 7 | Viewed by 3472
Abstract
Understanding the spatio-temporal pattern of natural vegetation helps decoding the responses to climate change and interpretation on forest resilience. Satellite remote sensing based data products, by virtue of their synoptic and repetitive coverage, offer to study the correlation and lag effects of rainfall [...] Read more.
Understanding the spatio-temporal pattern of natural vegetation helps decoding the responses to climate change and interpretation on forest resilience. Satellite remote sensing based data products, by virtue of their synoptic and repetitive coverage, offer to study the correlation and lag effects of rainfall on forest growth in a relatively longer time scale. We selected central India as the study site. It accommodates tropical natural vegetation of varied forest types such as moist and dry deciduous and evergreen and semi-evergreen forests that largely depend on the southwest monsoon. We used the MODIS derived NDVI and CHIRPS based rainfall datasets from 2001 to 2018 in order to analyze NDVI and rainfall trend by using Sen’s slope and standard anomalies. The study observed a decreasing rainfall trend over 41% of the forests, while the rest of the forest area (59%) demonstrated an increase in rainfall. Furthermore, the study estimated drought conditions during 2002, 2004, 2009, 2014 and 2015 for 98.2%, 92.8%, 89.6%, 90.1% and 95.8% of the forest area, respectively; and surplus rainfall during 2003, 2005, 2007, 2011, 2013 and 2016 for 69.5%, 63.9%, 71.97%, 70.35%, 94.79% and 69.86% of the forest area, respectively. Hence, in the extreme dry year (2002), 93% of the forest area showed a negative anomaly, while in the extreme wet year (2013), 89% of forest cover demonstrated a positive anomaly in central India. The long-term vegetation trend analysis revealed that most of the forested area (>80%) has a greening trend in central India. When we considered annual mean NDVI, the greening and browning trends were observed over at 88.65% and 11.35% of the forested area at 250 m resolution and over 93.01% and 6.99% of the area at 5 km resolution. When we considered the peak-growth period mean NDVI, the greening and browning trends were as follows: 81.97% and 18.03% at 250 m and 88.90% and 11.10% at 5 km, respectively. The relative variability in rainfall and vegetation growth at five yearly epochs revealed that the first epoch (2001–2005) was the driest, while the third epoch (2011–2015) was the wettest, corresponding to the lowest vegetation vigour in the first epoch and the highest in the third epoch during the past two decades. The study reaffirms that rainfall is the key climate variable in the tropics regulating the growth of natural vegetation, and the central Indian forests are dominantly resilient to rainfall variation. Full article
(This article belongs to the Special Issue Geostatistics and Spatial Data Mining for Ecological Climatology)
Show Figures

Graphical abstract

Back to TopTop