Kelp dynamics and environmental drivers in the southern Salish Sea, British Columbia, Canada

Frontiers in Marine Science, Journal Year: 2024, Volume and Issue: 11

Published: Jan. 30, 2024

The impacts of local-scale temperatures and winds on bull kelp ( Nereocystis luetkeana ) vary along a coastal gradient, while also being influenced by corresponding global-scale oceanic conditions. Around Vancouver Island the Gulf Islands, BC, Canada, floating canopies were mapped using high-resolution imagery from 2005 to 2022, whereas largest bed area was with medium-resolution spanning 1972 2022. In order understand spatial patterns resilience, abiotic characteristics used organize coastline into four clusters, ranging coldest most exposed coast more sheltered warmer location. Additionally, sea surface temperatures, winds, marine heatwaves categorized temporal conditions defined positive/negative oscillations Oceanic Niño Index (ONI) Pacific Decadal Oscillation (PDO). Comparing categories, we observed that years positive ONI PDO, in particular 2014–2019 period, concentrated spring temperature peaks. However, there are some indications an underlying long-term trend. During period 2020–2022, when PDO negative, summer kept increasing wind displayed higher frequency extreme events. Mapped showed different trends these stressors: constant presence during entire even dating back 1972. Warmer semi-sheltered coasts increased percentage cover after ONI+PDO 2014–2019, facing Strait Georgia lower than other clusters. summary, resilient study area, but for reasons: colder had favorable kelp, beds may have benefited wind-wave forcing.

Language: Английский

---

Local and regional variation in kelp loss and stability across coastal British Columbia

Marine Ecology Progress Series, Journal Year: 2024, Volume and Issue: 733, P. 1 - 26

Published: Feb. 13, 2024

Kelp forests are among the most abundant coastal marine habitats but vulnerable to climate change. The Northeast Pacific has experienced recent large-scale changes in kelp abundance and distribution, little is known about north of British Columbia (BC)-Washington border. Here, we assessed whether how floating canopy ( Macrocystis pyrifera, Nereocystis luetkean a ) distributions have changed decades along extensive coast BC. We assembled analysed available distributional data, comparing snapshots linear extent from 1.5-3 ago (1994-2007) recently collected data (2017-2021) across 11 different subregions spanning province. then leveraged timeseries, where (n = 7 sets), contextualise patterns In aggregate, suggest that declined considerably some parts province, with variable change warmest areas (southern BC), persistence was negatively correlated mean summer sea surface temperatures, which at times exceeded thermal tolerances. contrast, northern subregions, top-down control by urchins otters appeared modulate dynamics, declines occurring 2 despite cool ocean temperatures. Timeseries many occurred around 2014-2016 heatwave, an event associated sustained warming altered trophic dynamics. Our results BC’s places decades, regional local-scale factors influence their responses environmental

Language: Английский

---

Validating Landsat Analysis Ready Data for Nearshore Sea Surface Temperature Monitoring in the Northeast Pacific

Remote Sensing, Journal Year: 2024, Volume and Issue: 16(5), P. 920 - 920

Published: March 6, 2024

In the face of global ocean warming, monitoring essential climate variables from space is necessary for understanding regional trends in dynamics and their subsequent impacts on ecosystem health. Analysis Ready Data (ARD), being preprocessed satellite-derived products such as Sea Surface Temperature (SST), allow easy synoptic analysis temperature conditions given consideration biases within a dynamic range. This especially true SST retrieval thermally complex coastal zones. this study, we assessed accuracy 30 m resolution Landsat ARD to measure nearshore SST, derived 8 TIRS, 7 ETM+, 5 TM thermal bands over 37-year period (1984–2021). We used situ lighthouse buoy matchup data provided by Fisheries Oceans Canada (DFO). Excellent agreement (R2 0.94) was found between spring/summer at farshore site (>10 km coast), with mean bias (root square error) 0.12 °C (0.95 °C) general pattern underestimation −0.28 (0.96 overestimation 0.65 (0.98 °C). Spring/summer matchups revealed best −0.57 (1.75 90–180 coast temperatures 25 °C. Overall, image sampling distance recommended manuscript seeks capture close possible margin—and critical habitats interest—while minimizing pixel mixing adjacent land emissivity SST.

Language: Английский

---

Research on Object Detection and Recognition Method for UAV Aerial Images Based on Improved YOLOv5

Drones, Journal Year: 2023, Volume and Issue: 7(6), P. 402 - 402

Published: June 17, 2023

In this paper, an object detection and recognition method based on improved YOLOv5 is proposed for application unmanned aerial vehicle (UAV) images. Firstly, we the traditional Gabor function to obtain convolutional kernels with better edge enhancement properties. We used eight enhance edges from directions, enhanced image has obvious features, thus providing best area subsequent deep feature extraction work. Secondly, added a coordinate attention (CA) mechanism backbone of YOLOv5. The plug-and-play lightweight CA considers information both spatial location channel features can accurately capture long-range dependencies positions. like eyes YOLOv5, making it easier network find region interest (ROI). Once again, replaced Path Aggregation Network (PANet) Bidirectional Feature Pyramid (BiFPN) at neck BiFPN performs weighting operations different input layers, which helps balance contribution each layer. addition, adds horizontally connected branches across nodes bidirectional fusion structure fuse more in-depth information. Finally, trained overall model our integrated dataset LSDUVD compared other models multiple datasets. results show that convergence effect mAP value, demonstrates unique advantages in processing tasks UAV

Language: Английский

---

Comparing Object-Based and Pixel-Based Machine Learning Models for Tree-Cutting Detection with PlanetScope Satellite Images: Exploring Model Generalization

International Journal of Applied Earth Observation and Geoinformation, Journal Year: 2023, Volume and Issue: 125, P. 103555 - 103555

Published: Nov. 8, 2023

Despite utilizing various remote sensing datasets, precise tree-cutting detection remains challenging due to spatial and spectral resolution constraints in satellite imagery, complex landscapes, data integration issues, the need for accurate multi-temporal reference datasets. This study investigates utilization of PlanetScope (PS) images, along with pixel-based (PBIA) object-based (OBIA) image analysis, mapping forest cover tree cuttings. Detailed datasets were collected based on airborne laser scanning (ALS)-derived canopy height models (CHM) very high-resolution (VHR) aerial orthomosaics. Reference used train three machine learning (ML) models: random (RF), support vector (SVM), feed-forward neural network (Nnet) two districts located Western Northern Poland. The also assessed generalization capabilities best model both local temporal contexts. Regarding mapping, OBIA RF classifier outperformed all other an overall accuracy (OA) 99.27 % Kappa 98.18 %, while PBIA SVM showed lowest (OA = 97.18 94.35 %). testing model's confirmed performance model, Dice Coefficient ranging from 95.86 96.74 %. methodology's effectiveness was demonstrated, rate 96.20 99.39 total number cuttings, 99.45 99.86 volume. In conclusion, PS spectral-textural features, generalized ML proves be effective detection.

Language: Английский

---

Back to the past: long-term persistence of bull kelp forests in the Strait of Georgia, Salish Sea, Canada

Frontiers in Marine Science, Journal Year: 2024, Volume and Issue: 11

Published: Sept. 17, 2024

The Salish Sea, a dynamic system of straits, fjords, and channels in southwestern British Columbia, is home to ecologically culturally important bull kelp ( Nereocystis luetkeana ) forests. Yet the long-term fluctuations area persistence this pivotal coastal marine habitat are unknown. Using very high-resolution satellite imagery map forests over two decades, we present spatial changes forest within before (2002 2013) after (2014 2022) ‘Blob,’ an anomalously warm period Northeast Pacific. This analysis was spatially constrained by local environmental conditions. Based on nearshore sea surface temperatures (SSTs) from four decades (1984–2022), found periods distinct increases SST, one starting 2000 another 2014. Further, highest SST anomalies occurred warmer coastlines enclosed inlets Strait Georgia, while smaller were colder near Juan de Fuca Discovery Passage. total 2014 2022 has decreased compared 2002 2013, particularly northern sector Sea. satellite-derived data, also with historical distribution depicted Admiralty Nautical Charts 1858 1956. shows that warm, sheltered areas experienced considerable decrease beds when modern kelp, confirming century-scale loss. In particular, presence Georgia warmest coasts considerably century, likely due warming temperatures. While coldest south have maintained their centennial persistence, Sea requires further research understand its current dynamics. contributes wider understanding temporal factors for regional perspective

Language: Английский

---

Enhancing kelp forest detection in remote sensing images using crowdsourced labels with Mixed Vision Transformers and ConvNeXt segmentation models

International Journal of Remote Sensing, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 19

Published: Jan. 14, 2025

Kelp forests, as foundation species, are vital to marine ecosystems, providing essential food and habitat for numerous organisms. This study explores the integration of crowdsourced labels with advanced artificial intelligence models develop a fast accurate kelp canopy detection pipeline using Landsat images. Building on success machine learning competition, where this approach ranked third performed consistently well both local validation public private leaderboards, research highlights effectiveness combining Mixed Vision Transformers (MIT) ConvNeXt models. Training these various image sizes significantly enhanced accuracy ensemble results. U-Net emerged best segmentation architecture, UpperNet also contributing final ensemble. Key bands, such ShortWave InfraRed (SWIR1) Near-InfraRed (NIR), were crucial while altitude data was used in postprocessing eliminate false positives land. The methodology achieved high rate, accurately identifying about three out four pixels containing keeping low. Despite medium resolution satellites, their extensive historical coverage makes them effective studying forests. work underscores potential scalable environmental monitoring. All running code training all inference can be found at https://github.com/IoannisNasios/Kelp_Forests.

Language: Английский

---

Assessment of the application of each multibeam echosounder data product for monitoring of Laminaria digitata in the UK

Frontiers in Remote Sensing, Journal Year: 2025, Volume and Issue: 6

Published: Feb. 26, 2025

Amid warming seas, high rates of pollution and declining fish stocks observed around the UK, vital role kelp as ecosystem mediators on our coastlines is increasingly significant; currently estimated at £500 billion. Extensive research rapid decline forests its potential consequences has prompted initiation numerous conservation efforts. This set out to determine applicability efficiency a less invasive, remote sensing technique for monitoring kelp. A resolution multibeam echosounder (MBES) survey was performed acquire depths, backscatter water column data in an area known have An evaluation different combinations MBES products forest carried out. image-based processing methodology using random algorithm used generate classification models, which were trained tested ground truth samples obtained through video imagery. study reports climbing model accuracy scores from 62.2% (±11%, 1σ) 90% (±10%, consecutive input products, indicating effective tool with respect other technologies. When considering practical difficulties associated simultaneous record all against their individual value, this suggests that bathymetry deliver greatest value distinction small form kelp, while angular response analysis lesser but are required optimised accuracy.

Language: Английский

---

Canopy-forming kelp forests persist in the dynamic subregion of the Broughton Archipelago, British Columbia, Canada

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 12

Published: March 13, 2025

Canopy-forming kelp forests act as foundation species that provide a wide range of ecosystem services along temperate coastlines. With climate change, these ecosystems are experiencing changing environmental and biotic conditions; however, the distribution drivers change in British Columbia remain largely unexplored. This research aimed to use satellite imagery data investigate spatiotemporal persistence resilience dynamic subregion cool ocean temperatures high abundance Broughton Archipelago, Columbia. The specific objectives were identify: 1) long-term (1984 2023) short-term (2016 responses changes; 2) spatial patterns persistence. time series was divided into three periods: 1984 1998, 1999 2014, 2014 2023. first transition between periods represented shift cooler regional sea-surface negative Pacific Decadal Oscillation 1999. second warmer (with more marine heatwaves El Niño conditions) after 2014. In 2023), which covered site with Macrocystis pyrifera beds, area increased slightly start period For focused on eight sites Nereocystis luetkeana most either did not significantly or expanded area. suggests areas remained persistent across despite showing interannual variability. Thus, Archipelago may be refuge for kelps, likely due water below both species’ upper thermal limits. Spatially, bed level, center but subregion, had than , suggesting life history and/or other factors impacting beds differently. These findings demonstrate informing management forest by First Nations local communities.

Language: Английский

---

From archives to satellites: uncovering loss and resilience in the kelp forests of Haida Gwaii

Frontiers in Marine Science, Journal Year: 2025, Volume and Issue: 12

Published: April 4, 2025

Coastal foundation species such as kelps, corals, and seagrasses play vital roles in supporting marine biodiversity ecosystem services globally, but are increasingly threatened by climate change. In particular, kelp forests highly dynamic ecosystems experiencing natural fluctuations across seasons cycles, e.g., El Niño Southern Oscillation, Pacific Decadal Oscillation. As change increases variability these cycles extreme events heatwaves become more frequent, long term data essential to understand deviations from the norm better estimate trends of This study uses a century-long dataset examine forest responses regional drivers Haida Gwaii, British Columbia, combining remote sensing 1973-2021 with snapshot distribution derived historical records 1867-1945. We reveal complex patterns change, losses resilience varying at different spatial scales. Kelp that had likely persisted for over century exhibited an overall declining trend 5 ± 2% per decade starting 1970s. Throughout time series area was driven multi-year impacts Nino sea surface temperature anomalies heatwaves, 1998 2014-2016 heatwave known ‘Blob’. warmest areas, completely disappeared during 1977 Oscillation shift. Cooler areas showed greater resilience, buffering loss region wide scale, highlighting importance local gradients understanding vulnerable Lastly, situ surveys lack urchin barrens, presence turf algae region, further hypothesis temperature, not herbivory, drove this region.

Language: Английский

---