Monitoring vegetation changes in Pasvik (Norway) and Pechenga in Kola Peninsula (Russia) using multitemporal Landsat MSS/TM data

doi:10.1016/S0034-4257(03)00014-2

Remote Sensing of Environment

Volume 85, Issue 3, 30 May 2003, Pages 370-388

https://doi.org/10.1016/S0034-4257(03)00014-2 Get rights and content

Abstract

During the period 1973–1999, the effects of air pollution on the terrestrial ecosystems in the border areas of Norway and Russia have been investigated by the use of satellite remote sensing. Landsat Multispectral Scanner (MSS) and Thematic Mapper (TM) data, in combination with field work data collected in the period 1970–2000, were applied to produce land cover maps for seven different years, with an overall accuracy of 75–83%. On the basis of this monitoring, we can conclude that the main effect of air pollution was that the areas of lichen (Cladina spp.)-dominated forests and mountain heaths were reduced from 37% in 1973 to 10% in 1994, followed by a slight increase to 12% in 1999. The lichen-dominated vegetation types were changed into barrens, partly damaged vegetation entities, and dwarf shrub (e.g., Vaccinium myrtillus)-dominated vegetation. We found a significant negative relationship (r=−0.94, p=0.001) between the extent of the area of mixed forests with lichen content and the emissions of SO₂ during the period. The area of the category industrial barrens had a significant negative relationship (r=−0.95, p=0.001) to the emissions of SO₂ during the same period.

Section snippets

Vegetation damage in the study area caused by SO₂ emissions

Over the last decade, the large smelting and copper (Cu)–nickel (Ni)-producing factories (companies) at Nikel and Monchegorsk in the western part of Kola Peninsula, Northwest Russia (Fig. 1), have attracted increasing media coverage, criticism, and pressure from environmentalists, particularly in Norway and Finland. The impact on the arctic environment of wind-borne pollutants from the Cu–Ni smelters is manifested in several forms, not least in regard to human health. The arctic ecosystems are

Study area

The study area is located in the northernmost part of Norway close to the Russian border. The area includes parts of the Enare district in Finland and parts of the Pechenga district in Russia. The study area is outlined in Fig. 1 and covers approximately 7100 km². Within this area, the Pechenganikel in Nikel was established in 1933. Emissions reached a level of 100 thousand tons of sulphur dioxide (SO₂) at the highest during the first 30 years of production. The emissions from this processing

Air pollution data and SO₂ dispersion maps

The Norwegian Air Research Institute Bekkestad et al., 1995, Hagen et al., 2000 provided different maps showing the concentrations of SO₂ in the ground air layer that is used as a background material in the analysis of the data. The estimated average SO₂ concentrations in the ground air layer (μg/m³) in summer seasons of 1975 and 1998 are illustrated in Fig. 3 Bekkestad et al., 1995, Hagen et al., 2000. Wind measurements from the summer (April to September) of 1994 show that the prevailing wind

Vegetation and land cover maps of 1973–1999 based on MSS/TM data

Approximately 7100 km² of the Pasvik–Nikel area is covered by the vegetation and land cover maps based on MSS and TM data. The results from the field work, classification, and interpretation of Landsat MSS and TM data are presented in Table 2, Table 3, Table 4. The vegetation and land cover maps are shown in Fig. 4a–c. The accuracy assessments of the Landsat MSS and TM data-based maps in relation to the ground truth were performed by in situ controls of four transects within the study area, as

Assessment of results and methods

Pitblado and Amiro (1982) showed that use of vegetation indices gave the most satisfactory results concerning mapping and monitoring of vegetation damage around nickel smelters. Rees and Williams (1997) stated that remote sensing is best to monitor changes in species composition rather than changing biomass values. Our results show that the use of the hybrid classification gave satisfactory results concerning vegetation cover changes, including lichen-dominated vegetation formations. The cover

Conclusions

The main change observed during the period 1973–1999 was that the area with lichen-dominated vegetation (both heaths and forests) had decreased from 37% of the investigated area in 1973 to 12% in 1999. The lichen heaths and the lichen forests, especially, situated in the higher altitudes and in areas exposed towards Nikel and Zapolyarnij suffered during the period 1973–1999. The lichen cover in these areas/types is often removed, and bare soils, gravels, and bare rocks have been exposed. Lichen

Acknowledgements

The work on this article has benefited from financial support from a 3-year fellowship funded by the Norwegian Research Council, and by funding from The Joint Norwegian–Russian Environmental Commission and Barents E4-INTERREG11 Program.

References (62)

R. Congalton
A review of assessing the accuracy of classifications of remotely sensed data
Remote Sensing of Environment
(1991)
O. Hagner et al.
Detection of forest decline in Monchegorsk area
Remote Sensing of Environment
(1998)
D.E. Petzold et al.
Reflectance spectra of subarctic lichens
Remote Sensing of Environment
(1988)
H. Tømmervik et al.
Monitoring the effects of air pollution on terrestrial ecosystems in Varanger (Norway) and Nikel–Pechenga (Russia) using remote sensing
Science of Total Environment
(1995)
C. Uhlig et al.
Airborne heavy metal pollution and its effects on foliar elemental composition of Empetrum hermaphroditum and Vaccinium myrtillus in Sør-Varanger, northern Norway
Environmental Pollution
(2001)
J.L. van Genderen et al.
Remote sensing: statistical testing for thematic map accuracy
Remote Sensing of Environment
(1978)
Aamlid, D. (2002). Air pollution impact on the state of terrestrial and aquatic ecosystems in the Norwegian–Russian...
D. Aamlid et al.
Determination of exceedance of critical levels in the border area between Norway and Russia
D. Aamlid et al.
Methods of monitoring the effects of the air pollution on forest and vegetation of eastern Finnmark, Norway
Norwegian Journal of Agricultural Sciences
(1993)
Anonymous (1988). Statview SE+ Graphics Version 68000. Berkeley, CA, USA: Abacus...

Anonymous (1993). Convention on long-range transboundary air pollution. Draft manual on critical levels and loads for...

Bekkestad, T., Knudsen, S., Johnsrud, M., & Larsen, M. (1995). Modellberegninger av SO2 og metallavsetning i...

D.J. Booth et al.

A comparison of classification algorithms in terms of speed and accuracy after application of a post-classification modal filter

International Journal of Remote Sensing

(1989)

A. Colpaert et al.

Remote sensing, a tool for reindeer range land management

Polar Record

(1995)

J.J. Craighead et al.

A definitive system for analysis of grizzly bear habitat and other wilderness resources. Monograph no. 1

(1982)

J.J. Craighead et al.

Mapping arctic vegetation in northwest Alaska using Landsat MSS imagery

National Geographic Research

(1988)

N.M. Deyeva et al.

The state of the bilberry in polluted and unpolluted forests of the Kola Peninsula

R.O. Duda et al.

Pattern classification and scene analysis

(1973)

R. Elven

Johannes Lid and Dagny Tande Lid: Norsk Flora

(1994)

B. Freedman et al.

Long-term effects of smelter pollution at Sudbury, Ontario, on forest community composition

Canadian Journal of Botany

(1980)

E. Fremstad

Vegetasjonstyper i Norge

E. Gaare et al.

Hagen, L. O. (1995). NILU—Norwegian Air Research Institute, personal...

Hagen, L. O., Sivertsen, B., & Arnesen, K. (2000). Overvåking av luft-og nedbørkvalitet i grenseområdene i Norge og...

O. Hagner et al.

Detection of forest decline in Monchegorsk area

T. Hallingbäck et al.

D.L. Hawksworth et al.

Lichen re-colonisation in London (UK) under conditions of rapidly falling sulphur dioxide levels and the concept of zone skipping

Botanical Journal of Linnean Society

(1989)

Helle, T., & Kojola, I. (1992). The variation in the growth rate of Cladonia rangiferina in eastern Fennoscandia. In:...

L. Janssen et al.

Accuracy assessment of satellite derived land cover data, a review

Photogrammetric Engineering and Remote Sensing

(1994)

Kalabin, G. (1991). Ecological situation in the Pechenga region. Report. INEP. Apatity, Russia: Kola Science...

R. Kalliola

Pflanzensoziologische untersuchungen in der alpinen stufe Finnish Lapplands

Annales Botanici Societatis Zoologicae-Botanicae Fennicae Vanamo

(1939)

Cited by (106)

Investigating anthropogenically transformed landscapes with remote sensing
2021, Remote Sensing Applications: Society and Environment
The assessment of the current state and spatio-temporal changes in natural landscapes is crucial, among others, for the agro-industrial and agricultural sectors of every country, including Ukraine. The main objective of this paper is to estimate the state and spatio-temporal changes in landscapes with land-cover identification using remote sensing data processing and geographic information technologies. The generalized scheme of the present study consists in the analysis of natural and anthropogenic changes of Ukraine's capital city of Kiev landscapes using satellite data. The processing involved step-by-step decipherment of Kiev's landscapes using cluster analysis, namely supervised and unsupervised classification based on a series of satellite images from Landsat missions 5, 7, 8 (1985–2020) and Sentinel-2 MSI (2015–2020). Spectral signatures were determined, and their quality was estimated based on the classification. The Normalized Difference Vegetation Index (NDVI) was calculated for vegetated landscapes. The following landscapes were identified: water bodies, vegetation, urban areas, and bare soil. Two land cover maps were compiled from Sentinel-2 MSI images for 2015 and 2020 with an overall accuracy of 89.76% and 82.05%, respectively. Two maps of anthropogenic transformations of Kiev's landscapes were also compiled. According to Landsat data (spatial resolution 30 m), the area of urban landscapes increased by about 41% from 1985 to 2020, and according to Sentinel-2 MSI data (spatial resolution 10 m), it grew around 5% from 2015 to 2020.
Assessment of Landsat-based terricolous macrolichen cover retrieval and change analysis over caribou ranges in northern Canada and Alaska
2020, Remote Sensing of Environment
Terricolous macrolichens are an important food source for caribou (Rangifer tarandus) and can greatly influence their movement, distribution and demography over time. Mapping the spatial distribution and cover of macrolichens with remote sensing can serve as an important approach for assessing the impact of disturbances (e.g. fire, grazing, trampling) on lichen cover at the landscape scale and for monitoring post-disturbance rates of recovery. Previous remote sensing-based efforts to retrieve the distribution and abundance of lichen have been restricted to particular regions and thus are not indicative of the potential for large extent mapping and monitoring. In this study, we assessed the effectiveness of machine learning methods for retrieving lichen cover and change across different regions in northern Canada and Alaska using Landsat-5 images, topographic and climate data. Global and regional-scale models were evaluated to assess whether regionally specific analyses would improve performance. Of the models tested, the deep neural network was the most accurate for predicting lichen cover (model efficiency (ME) = 0.58, mean absolute error (MAE) < 7%). For the regional analysis, the performance was the best in north-central Canada (ME = 0.56, MAE = 8%) and the worst in north-eastern Canada (ME = 0.22, MAE < 4%) due to lower lichen cover, more exposed ground, and reduced sample quality and distribution. Analysis of trend-based change detection from 1984 to 2011 in the three regional test areas showed the expected directional response with declining lichen cover in north-western Canada in response to climate-induced shrub expansion, slow recovery to wildfire in north-central Canada, and declining lichen cover in north-eastern Canada related to caribou foraging/trampling and shrub expansion.
Modelling occurrence and status of mat-forming lichens in boreal forests to assess the past and current quality of reindeer winter pastures
2019, Ecological Indicators
Lichens play an essential role in northern ecosystems as important contributors to the water, nutrient and carbon cycles, as well as the main winter food resource for reindeer (Rangifer tarandus, also called caribou in North America), the most abundant herbivores in arctic and subarctic regions. Today, climate change and several types of land use are rapidly transforming northern ecosystems and challenging lichen growth. Since lichens are important indicators of ecosystem health and habitat suitability for reindeer, large-scale assessments are needed to estimate their past, present and future status. In our study, we aimed to develop models and equations that can be used by stakeholders to identify the occurrence of lichen-dominated boreal forests and to determine lichen conditions in those forests. Data were collected in Sweden and most input data are publicly available. We focused on mat-forming lichens belonging to the genera Cladonia and Cetraria, which are dominant species in the reindeer and caribou winter diet. Our models described lichen-dominated forests as being dominated by Scots pine (Pinus sylvestris), having low basal area and thin canopy cover, and being located in south- and west-facing areas with high summer precipitation, low winter precipitation and temperature, and on gentle slopes. Within those forests, lichen biomass was positively related to tree canopy cover and summer precipitation, while negatively and exponentially related to intensity of use of the area by reindeer. Forest, meteorological, topographic and soil data can be used as input in our models to determine lichen conditions without having to estimate lichen biomass through demanding and expensive fieldwork.
Using NDVI time series to diagnose vegetation recovery after major earthquake based on dynamic time warping and lower bound distance
2018, Ecological Indicators
Citation Excerpt :
More efforts should be paid to develop new methods for revealing the dynamic process of vegetation recovery and conduct case studies for cross-comparison of various methods. Remote sensing data can effectively diagnose vegetation recovery of mountainous regions (Tømmervik et al., 2003, Turner et al., 2007, Ni et al., 2009, Wang et al., 2009, Lu et al., 2012), although researchers always face challenges from unsatisfactory data quality (e.g., cloud cover) and natural variability (e.g., changes in phenology) (Lu et al., 2012). The vegetation recovery rate (VRR) (Lin et al., 2006) method compares vegetation changes at a few cross-sectional data before and after a major disturbance.
Major earthquake occurred in mountainous areas usually cause large number of landslides that lead to severe impact to local vegetation cover and growth. The negative influence to vegetation may last for many year and vegetation recovery may experience dynamic fluctuation. Existing methods for vegetation recovery diagnosis face difficulty in capturing the dynamic behaviours both within and between years that makes the interannual comparison impossible. This paper proposes a new method to diagnose regional vegetation recovery after a major earthquake by defining a difference measurement index (DMI) using MODIS NDVI time series at 8-day interval. This differs from many existing methods in its quantification of the difference between the studied time series and historical samples, by using a proposed algorithm consisting of lower bound distance and dynamic time warping. This algorithm can better differentiate vegetation disturbance from its natural fluctuation. Second, the method investigates relatively regional vegetation recovery via a dynamic index, the DMI. Vegetation conditions in different years can be compared with a historical benchmark and measured by DMI. This makes it possible to diagnose dynamic vegetation recovery and generate a series of interannual spatial distributions of regional vegetation state.
An inverse method for watershed change detection using hybrid conceptual and artificial intelligence approaches
2018, Journal of Hydrology
Land Use and Land Cover (LULC) changes intrinsically lead to various hydrological impacts especially on the outflow rate of the watersheds. This research investigated LULC changes and its effect on outlet runoff by detecting LULC changes location and severity via an inverse method for the Little River Watershed, USA. For this purpose, the Clark’s conceptual Rainfall-Runoff model was applied to the delineated sub-watersheds of the basin to generate different outflows by altering the Storage Coefficient (SC) of the model for each sub-watershed as the representative of LULC. Then, the relation between SC values and outflow time series was simulated by Artificial Intelligence (AI) based models of artificial neural network and least square support vector machine. In this way, in order to ignore redundant information and reduce the dimension of input vector, Wavelet-Entropy (WE) values of the outflow sub-series were computed and used as the inputs of the AI model to compute SCs of the sub-watersheds as the outputs. The trained multi-output AI model as an inverse method could be then used to predict the SC values (as representative of LULC) of the sub-watersheds using the observed time series of runoff at the outlet. The obtained results showed that the proposed inverse method could reliably detect not only the location but also the severity of LULC changes by prediction of SC values in the coming future years. For validation of the method, a comparison was also performed between the obtained results and recorded changes via normalized difference vegetation index (NDVI) and land use classification, extracted from Landsat images. The comparison approved the ability of the proposed method for LULC change detection in a way that deforestation and cropland increasing of the sub-watersheds from 1990 to 2013 were aligned with the SC reduction e.g., 26% decrease of SC for downstream sub-watershed versus 53% decrease and 21% increase of forest and crop lands, respectively.
Carbon based secondary compounds do not provide protection against heavy metal road pollutants in epiphytic macrolichens
2016, Science of the Total Environment
Citation Excerpt :
Lichens often produce fairly species-specific carbon-based secondary compounds (CBSCs) in high concentration; up to 20–30% of dry matter (Huneck, 1973, Solhaug et al., 2009). They are among the more abundant organic molecules in lichen-rich habitats with the potential to change the albedo (Bernier et al., 2011), be seen from outer space (Tømmervik et al., 2003), and kill grazing animals (Cook et al., 2007). Some are considered to influence a lichen species' heavy metal tolerance and uptake (Purvis et al., 1987, 1990, Purvis and Pawlik-Skowronska, 2008, Hauck et al., 2009, 2013).
Lichens are useful monitoring organisms for heavy metal pollution. They are high in carbon based secondary compounds (CBSCs) among which some may chelate heavy metals and thus increase metal accumulation. This study quantifies CBSCs in four epiphytic lichens transplanted for 6 months on stands along transects from a highway in southern Norway to search for relationships between concentrations of heavy metals and CBSCs along a gradient in heavy metal pollutants. Viability parameters and concentrations of 21 elements including nutrients and heavy metals in these lichen samples were reported in a separate paper. Medullary CBSCs in fruticose lichens (Ramalina farinacea, Usnea dasypoga) were reduced in the most polluted sites, but not in foliose ones (Parmelia sulcata, Lobaria pulmonaria), whereas cortical CBSC did not change with distance from the road in any species. Strong positive correlations only occurred between the major medullary compound stictic acid present in L. pulmonaria and most heavy metals, consistent with a chelating role of stictic acid, but not of other studied CBSCs or in other species. However, heavy metal chelating did not protect L. pulmonaria against damage because this species experienced the strongest reduction in viability in the polluted sites. CBSCs with an accumulation potential for heavy metals should be quantified in lichen biomonitoring studies of heavy metals because they, like stictic acid, could overshadow pollutant inputs in some species rendering biomonitoring data less useful. In the two fruticose lichen species, CBSCs decreased with increasing heavy metal concentration, probably because heavy metal exposure impaired secondary metabolism. Thus, we found no support for a heavy metal protection role of any CBSCs in studied epiphytic lichens. No intraspecific relationships occurred between CBSCs versus N or C/N-ratio. Interspecifically, medullary CBSCs decreased and cortical CBSCs increased with increasing C/N-ratio.

View all citing articles on Scopus

View full text

Monitoring vegetation changes in Pasvik (Norway) and Pechenga in Kola Peninsula (Russia) using multitemporal Landsat MSS/TM data

Abstract

Section snippets

Vegetation damage in the study area caused by SO2 emissions

Study area

Air pollution data and SO2 dispersion maps

Vegetation and land cover maps of 1973–1999 based on MSS/TM data

Assessment of results and methods

Conclusions

Acknowledgements

Remote Sensing of Environment

Remote Sensing of Environment

Remote Sensing of Environment

Science of Total Environment

Environmental Pollution

Remote Sensing of Environment

Determination of exceedance of critical levels in the border area between Norway and Russia

Methods of monitoring the effects of the air pollution on forest and vegetation of eastern Finnmark, Norway

Norwegian Journal of Agricultural Sciences

A comparison of classification algorithms in terms of speed and accuracy after application of a post-classification modal filter

International Journal of Remote Sensing

Remote sensing, a tool for reindeer range land management

Polar Record

A definitive system for analysis of grizzly bear habitat and other wilderness resources. Monograph no. 1

Mapping arctic vegetation in northwest Alaska using Landsat MSS imagery

National Geographic Research

The state of the bilberry in polluted and unpolluted forests of the Kola Peninsula

Pattern classification and scene analysis

Johannes Lid and Dagny Tande Lid: Norsk Flora

Long-term effects of smelter pollution at Sudbury, Ontario, on forest community composition

Canadian Journal of Botany

Vegetasjonstyper i Norge

Detection of forest decline in Monchegorsk area

Lichen re-colonisation in London (UK) under conditions of rapidly falling sulphur dioxide levels and the concept of zone skipping

Botanical Journal of Linnean Society

Accuracy assessment of satellite derived land cover data, a review

Photogrammetric Engineering and Remote Sensing

Pflanzensoziologische untersuchungen in der alpinen stufe Finnish Lapplands

Annales Botanici Societatis Zoologicae-Botanicae Fennicae Vanamo

Vegetation damage in the study area caused by SO₂ emissions

Air pollution data and SO₂ dispersion maps