Have you ever heard the phrase "All the rivers run into the sea"? In most cases, this statement holds, with one exception: rivers that end up in lakes. If you imagine mountain ranges as the walls of a bathtub, the ocean is like the bottom of the bathtub, collecting all the water from the bathtub. No matter where you live, you inhabit a land area where all the water, above and below ground, converges into a common body of water (Figure 1). We call this area a watershed. Watersheds vary in size. The watershed of a small mountain stream may be only a few square meters; some watersheds are large, often containing many smaller watersheds (Government of Canada, n.d.). For example, the Amazon Basin, which we are familiar with, covers one-third of the South American continent (Routley, 2021).

Figure 1. The main components of the watershed (Agency of Natural Resources, 2023). The white arrows indicate the movement or transfer of water in the watershed.
Figure 1. The main components of the watershed (Agency of Natural Resources, 2023). The white arrows indicate the movement or transfer of water in the watershed.


The influence of forests on watersheds

Forests cover about 30% of the Earth's surface (FAO and UNEP, 2020). As shown in Figure 2, they can be found in arid areas, wetlands, and tidal regions, from the cold Arctic Circle to the hot tropics. Forests are hydrological regulators and can store, release, and purify water through the interaction of hydrological processes (Zhang and Wei, 2021). In addition, they can slow down the speed of water flow passing through the Earth's surface, reduce runoff and enhance soil infiltration(Netzer et al. 2019). In some cases, they even increase groundwater recharge (Netzer et al. 2019).

Figure 2. Map of forest distribution by global ecological zone (FAO and UNEP, 2020).
Figure 2. Map of forest distribution by global ecological zone (FAO and UNEP, 2020).


Forests within watersheds play a significant role in regulating downstream water supply and related ecosystem services (Zhang and Wei, 2021). As shown in Figure 3, forests, especially those upstream, can regulate streamflow through canopy interception, evapotranspiration, and soil infiltration. Changes in forests affect river flow, and influence downstream precipitation and water supply (Creed and Noordwijk, 2018).

Figure 3. The influence of forests on water supply in the watershed (Zhang and Wei, 2021).
Figure 3. The influence of forests on water supply in the watershed (Zhang and Wei, 2021).


Despite the numerous positive effects of forests on watersheds, many forests are being destroyed. The global forest area has been declining overall since 1990, with deforestation and forest degradation continuing at alarming rates (FAO and UNEP, 2020). According to FAO and UNEP (2020), it is estimated that 420 million hectares of forest area have been lost through conversion to other land use since 1990. Agricultural expansion remains a primary driver of deforestation and forest degradation (Chakravarty et al., 2012). In recent years, more frequent and large-scale forest fires (especially in the Brazilian Amazon, Australia, the western United States, etc.) and forest pest outbreaks (such as bark beetle infestations in western North America) due to climate change have also caused significant damage to forest ecosystems (Zhang and Wei, 2021).

The consequences of this destruction are significant. Chakravarty et al. (2012) and Zhang and Wei (2021) highlight the effects on regional water cycles as listed below

  1. Loss of water storage capacity: Trees absorb water from the soil through transpiration and release it into the atmosphere, helping to maintain soil moisture and facilitate rainfall infiltration. Deforestation reduces the number of trees, weakening the forest's ability to retain and release water, making the soil more prone to drying out and hindering rainfall infiltration, resulting in water wastage and depletion of water resources.
  2. Disruption of the water cycle: Forests help regulate atmospheric water content and influence cloud formation through transpiration and rainfall interception. Deforestation disrupts this balance, potentially leading to changes in climate patterns, including more frequent droughts or extreme rainfall events, a reduction of evaporation and precipitation.
  3. Increased risk of floods and droughts: After deforestation, soil erosion and accelerated water flow, increase the risk of floods. On the other hand, deforested areas may be more prone to drought due to a lack of sufficient vegetation to retain moisture.
  4. Deterioration of water quality: Forest vegetation can filter and purify water sources, preventing sediment and pollutants from entering water bodies. Deforestation leads to soil erosion, increasing the sediment and pollutant content in water bodies, and posing greater challenges in obtaining clean water sources.

Space technology-based applications to monitor forest changes in watersheds

Space technology has revolutionized our ability to observe the Earth's surface. The advanced remote sensing instruments on the satellite provide valuable data for forest monitoring. Thanks to long-term consistent archives, remote sensing can be used for long-term continuous monitoring, such as forest area change and deforestation, as well as spatial mapping of forest ecosystems (Archard et al. 2017).

In general, medium resolution (30m) satellite imagery is the minimum requirement for monitoring changes in forest cover, with a Minimum Mapping Unit (MMU) area ranging from 1 to 5 ha (Archard et al. 2017). Currently, there are two main sources of global medium resolution (30m to 10m) remote sensing images for open access:

  • Optical medium resolution data. Due to its low cost, long-time series and data records, and unrestricted licensing, Landsat satellite sensors (30m) have always been the main tool for monitoring deforestation (Archard et al. 2017). Sentinel data (10m) is a good choice. due to the finer resolution, which leads to better results in monitoring changes in coverage (ESA, n.d.).
  • Other types of sensors, such as radar and LiDAR, also have limited applications. Radar can provide all-weather, day-and-night observation, while LiDAR has advantages in elevation measurement and vertical structure capture (Joshi et al. 2016; Schumacher and Christiansen, 2020).

More details of the characteristics of commonly used sensors are shown in Table 1.

Table 1. Characteristics of some commonly used sensors (Earth Data. n.d.a; Earth Data. n.d.b; Earth Online, n.d.a; Earth Online, n.d.b; EROS Center, 2018; eoPortal, 2012a; eoPortal, 2012b; ESA, n.d.; NASA, n.d.; Neuenschwander et al. 2023)
Type of Sensor Satellite and Sensor Resolution (m) Swath Width Revisit Frequency (days) Data Portal


Landsat-8 OLI




USGS Earth Explorer: https://earthexplorer.usgs.gov/

Sentinel-2 MSI


290 km


Copernicus Open Access Hub: https://scihub.copernicus.eu/



26 m in range (across track)

Between 6 m and 30 m in azimuth (along track)



ESA Earth Online Portal: https://earth.esa.int/eogateway



175 km


JAXA G-Portal: https://gportal.jaxa.jp/gpr/





ESA Earth Online Portal: https://earth.esa.int/eogateway


3 - 100 (depends on mode)

70 - 350 km (depends on mode)

46 (ALOS-1)

14 (ALOS-2)

JAXA G-Portal: https://gportal.jaxa.jp/gpr/


ICESat-2 L3A Land and Vegetation Height

Not Specified



National Snow and Ice Data Center: https://nsidc.org/home


To monitor forest change in the watershed, it is important to classify the Land Use and Cover Change (LUCC). Spectral vegetation indices (VIs) have been wildly applied because they can highlight specific vegetation features, such as leaf area index (LAI), the percentage of green cover, chlorophyll content, green biomass and absorbed photosynthetically active radiation (da Silva et al. 2020). Table 2 shows commonly used VIs, especially in scenarios with high vegetation coverage such as that of forests.

In practical applications, VIs are typically customized according to specific application requirements and supplemented with appropriate validation tools and methods. For example, the Ratio Vegetation Index (RVI) is highly sensitive to vegetation and has a good correlation with plant biomass. For vegetation with a larger leaf area index or higher canopy closure, the value is higher, with healthy vegetation typically falling within the range of 2 to 8. Conversely, for soil, moisture, and non-vegetative elements, the value is lower. However, it becomes sensitive to atmospheric effects and weakens its representation of biomass when vegetation cover is low (less than 50%) (Xue and Su, 2017). The Enhanced Vegetation Index (EVI) is sensitive to changes in the background canopy and does not saturate at high biomass levels (Aljahdali et al. 2021). The Normalized Difference Vegetation Index (NDVI) measures the active biomass of photosynthesis in plants and is one of the most suitable indices for monitoring crop growth dynamics. However, it is highly sensitive to soil brightness and atmospheric effects (Xue and Su, 2017). The Soil Adjusted Vegetation Index (SAVI) was modified by NDVI to take into account background soil components that may interfere with vegetation signals (Sashikkumar et al. 2017). The Normalized Difference Moisture Index (NDMI) and Normalized Difference Water Index (NDWI) are sensitive to vegetation moisture content and can distinguish it from soil moisture content (Aljahdali et al. 2021).

Table 2. Commonly used vegetation indices (VIs)



Main Application


Ratio Vegetation Index (RVI)

RVI = ρNIR / ρred

Green biomass estimations and monitoring, specifically, at high-density vegetation coverage

(Xue and Su, 2017)


Vegetation Index (EVI)

EVI = 2.5 (ρNIR - ρred) / (ρNIR + 6ρred – 7.5ρblue + 1)

Areas with rich chlorophyll (such as tropical rainforests) and low topographic impact

(Aljahdali et al. 2021)

Normalized Difference Vegetation Index (NDVI)

NDVI = (ρNIR - ρred) / (ρNIR + ρred)

Regional and global vegetation assessments

(Xue and Su, 2017)

Soil Adjusted Vegetation Index (SAVI)

SAVI = (1 + L) * (ρNIR - ρred) / (ρNIR + ρred + L)

where L is the soil conditioning index

Monitoring of soil-exposed areas

(Sashikkumar et al. 2017)

Normalized Difference Moisture Index (NDMI)


Forest disturbance and recovery

(Aljahdali et al. 2021)

Normalized Difference Water Index (NDWI)

NDWI = (ρNIR - ρ1240nm) / (ρNIR + ρ1240nm)

Vegetation water content

(Aljahdali et al. 2021)


Example Applications: Monitoring the impact of deforestation on the Lower Mekong river basin

To provide an example application, research by Netzer et al. (2019) assented tp the impact of deforestation on the Lower Mekong River Basin (LMB). Using Landsat satellite data as the main data source, the study mapped LMB's land use/land cover change. Land cover maps for 2001 and 2013 were then run using the Soil and Water Assessment Tool (SWAT) hydrological model to assess the impact of forest loss on waterborne ecosystems.

For land cover change due to forest loss, Figure 4 shows the general steps to monitor it in the LMB with the highest deforestation rates globally. In simple terms, after creating the forest cover benchmark map, non-forest classes are established on the map to determine the change of land use over time. The results show that the LMB lost 13% of its forest area between 2001 and 2013, with most of the forest loss (82%) being converted to cropland (Figure 5).

Figure 4. The general steps in the development of the land use/land cover change map
Figure 4. The general steps in the development of the land use/land cover change map from 2001 to 2013 (modified from Netzer et al. 2019).


Figure 5. 2001 land cover map (this paper) with subsequent forest loss between 2001–2013 (Netzer et al. 2019).
Figure 5. 2001 land cover map (this paper) with subsequent forest loss between 2001–2013 (Netzer et al. 2019).


To evaluate ecosystem services of forests related to water, such as freshwater supply and regulation, as well as groundwater recharge, the authors used the SWAT model. As the semi-distributive model with the best features for watershed modelling, it divides the watershed into smaller sub-watersheds and hydrologic response units (HRUs) with their unique attributes. It works with annual, monthly, or daily intervals (Janjić and Tadić, 2023). To implement a SWAT model, three datasets are required: topography data (digital elevation model, DEM), land use cover, and soil data (Janjić and Tadić, 2023). Figure 7 shows the percentage changes in several hydrological indicators (surface runoff and stream discharge, sedimentation yield and flow) caused by forest loss. The results indicate that as deforestation increases, runoff and sediment production also increase, which affects the flow of rivers and sediment flow.

Figure 6. Percent change in several hydrological indicators due to forest loss in the lower Mekong watershed from 2001–2013 (Netzer et al. 2019). The upper figure shows change in surface runoff and stream discharge, while the lower figure shows sediment yield and flow.
Figure 6. Percent change in several hydrological indicators due to forest loss in the lower Mekong watershed from 2001–2013 (Netzer et al. 2019). The upper figure shows the change in surface runoff and stream discharge, while the lower figure shows sediment yield and flow.



Utilizing space technology to observe the impact of forests on watersheds is a crucial step in understanding and protecting natural ecosystems. The hydrological consequences of forest changes are never straightforward. By harnessing the power of remote sensing and geographic information systems, we have gained a deeper understanding of current forest conditions and the relationships between forests and watersheds, enabling us to make informed decisions for sustainable management and conservation efforts. As we continue to enhance our understanding and application of spatial technology, let us remain steadfast in our commitment to working towards the goal set by the United Nations Forest Strategy Plan of increasing global forest area by 3% by 2030.


Agency of Natural Resources. 2023. “Keeping Vermont's Watersheds Healthy” https://storymaps.arcgis.com/stories/dfc9e2c6cbfb4a8d80bdaffad761de3a (accessed on 22 Jan 2024)

Aljahdali, M. O., Munawar, S., & Khan, W. R. 2021. “Monitoring mangrove forest degradation and regeneration: Landsat time series analysis of moisture and vegetation indices at Rabigh Lagoon, Red Sea”. Forests12(1), 52. https://doi.org/10.3390/f12010052

Archard, F., de Oliveira, Y. M. M., & Mollicone, D. 2017. Monitoring forest cover and deforestation. https://www.alice.cnptia.embrapa.br/alice/bitstream/doc/1083312/1/2017C…

Chakravarty, S., Ghosh, S. K., Suresh, C. P., Dey, A. N., & Shukla, G. 2012. Deforestation: causes, effects and control strategies. Global perspectives on sustainable forest management1, 1-26. https://cdn.intechopen.com/pdfs/36125/InTechDeforestation_causes_effect….

Creed, I. F., & van Noordwijk, M. 2018. “Forest and water on a changing planet: vulnerability, adaptation and governance opportunities”. A global assessment report. IUFRO world series38. https://www.cabdirect.org/cabdirect/abstract/20193124755

da Silva, V. S., Salami, G., da Silva, M. I. O., Silva, E. A., Monteiro Junior, J. J., & Alba, E. 2020. “Methodological evaluation of vegetation indexes in land use and land cover (LULC) classification.” Geology, Ecology, and Landscapes4(2), 159-169. https://doi.org/10.1080/24749508.2019.1608409

Earth Data. n.d.a “ERS-1”. https://asf.alaska.edu/datasets/daac/ers-1/ (accessed on 14 Feb 2024).

Earth Data. n.d.b “ERS-2”. https://asf.alaska.edu/datasets/daac/ers-2/ (accessed on 14 Feb 2024).

Earth Online. n.d.a "JERS-1”. https://earth.esa.int/eogateway/missions/radarsat#instruments-section (accessed on 14 Feb 2024).

Earth Resources Observation and Science (EROS) Center. 2018. “USGS EROS Archive - Radar - ALOS PALSAR Radar Processing System.” USGS website. https://www.usgs.gov/centers/eros/science/usgs-eros-archive-radar-alos-….

Earth Online. n.d.b “ERS-1/2 SAR IM Medium Resolution L1 [SAR_IMM_1P]”. https://earth.esa.int/eogateway/catalog/ers-1-2-sar-im-medium-resolutio…- (accessed on 14 Feb 2024).

eoPortal. 2012a. “ALOS-2 (Advanced Land Observing Satellite-2) / Daichi-2”. https://www.eoportal.org/satellite-missions/alos-2#performance-specific… (accessed on 15 Feb 2024).

eoPortal. 2012b. “EnviSat (Environmental Satellite)”. https://www.eoportal.org/satellite-missions/envisat#sciamachy-scanning-… (accessed on 15 Feb 2024).

European Space Agency (ESA). n.d. “Sentinel-2: Resolution and Swath”.  https://sentinel.esa.int/web/sentinel/missions/sentinel-2/instrument-pa… (accessed on 29 January 2024).

FAO and UNEP. 2020. The State of the World’s Forests 2020. Forests, biodiversity and people. Rome. https://doi.org/10.4060/ca8642en

Government of Canada. n.d. "Understanding Watersheds". https://agriculture.canada.ca/en/environment/managing-water-sustainably… (accessed on 22 Jan 2024).

Janjić, J., & Tadić, L. 2023. “Fields of Application of SWAT Hydrological Model—A Review”. Earth4(2), 331-344. https://doi.org/10.3390/earth4020018

Joshi, N., Baumann, M., Ehammer, A., Fensholt, R., Grogan, K., Hostert, P. et al. 2016. “A review of the application of optical and radar remote sensing data fusion to land use mapping and monitoring”. Remote Sensing8(1), 70. https://doi.org/10.3390/rs8010070

National Aeronautics and Space Administration (NASA). N.d. “Landsat 8”. https://landsat.gsfc.nasa.gov/preliminary-spectral-response-of-the-operational-land-imager-in-band-band-average-relative-spectral-response/ (accessed on 29 January 2024).

Netzer, M. S., Sidman, G., Pearson, T. R., Walker, S. M., & Srinivasan, R. 2019. “Combining global remote sensing products with hydrological modeling to measure the impact of tropical forest loss on water-based ecosystem services”. Forests10(5), 413. https://doi.org/10.3390/f10050413

Neuenschwander, A. L., K. L. Pitts, B. P. Jelley, J. Robbins, J. Markel, S. C. Popescu, R. F. Nelson, D. Harding, D. Pederson, B. Klotz, and R. Sheridan. 2023. ATLAS/ICESat-2 L3A Land and Vegetation Height, Version 6. [Indicate subset used]. Boulder, Colorado USA. NASA National Snow and Ice Data Center Distributed Active Archive Center. https://doi.org/10.5067/ATLAS/ATL08.006.

Routley, Nick. 2021. "Mapped: The Drainage Basins of the World’s Longest Rivers" https://www.visualcapitalist.com/mapped-the-drainage-basins-of-the-worl…

Sashikkumar, M. C., Selvam, S., Karthikeyan, N., Ramanamurthy, J., Venkatramanan, S., & Singaraja, C. 2017. “Remote sensing for recognition and monitoring of vegetation affected by soil properties”. Journal of the Geological Society of India90, 609-615. https://doi.org/10.1007/s12594-017-0759-8

Schumacher, J., & Christiansen, J. R. 2020. LiDAR Applications to Forest-Water Interactions. Forest-Water Interactions, 87-112. https://link.springer.com/chapter/10.1007/978-3-030-26086-6_4

Xue, J., & Su, B. 2017. “Significant remote sensing vegetation indices: A review of developments and applications”. Journal of sensors2017. https://doi.org/10.1155/2017/1353691

Zhang, M., & Wei, X. 2021. “Deforestation, forestation, and water supply.” Science371(6533), 990-991. https://doi.org/10.1126/science.abe7821.