Irrigation illustrates a major dilemma of agriculture: On the one hand, a growing world population demands more food and biomass (for example for energy production). On the other hand, natural resources such as water are only available in limited quantities and excessive use often leads to the degradation of ecosystems, which in turn has adverse effects on agricultural production and local livelihoods. Since agriculture without irrigation will hardly be possible today or in the future (De Wrachien, Goli, and others 2015), two measures are necessary to enable sustainable development in this area:

  1. Water demand and availability at regional level (e.g. river basins) must be known in order to identify possible overuse and adjust water allocation rights (Fabre et al. 2015).
  2. Irrigation technology must keep water losses to a minimum and intelligent management must provide plants with the right amount of water at the right time (Singh 2014).

In both aspects, space technology can make important contributions, as will be shown in the following two sections.

Mapping Irrigated Agriculture

For the estimation of agricultural water demand, there is usually insufficient data available from in-situ surveys, which also do not sufficiently capture the temporal dynamics of agricultural production cycles and meteorological processes.

Using satellite remote sensing techniques, agricultural crop types can be classified over wider areas, which allows an estimation of the potential water demand. For these methods, classification algorithms are mostly employed, which are either based on expert knowledge (rule-based systems), or are trained to differentiate between certain crop types using training data in a formalized learning process (learning systems). As input data mainly optical remote sensing data are used, which are partly combined with radar data (Joshi et al. 2016; Orynbaikyzy, Gessner, and Conrad 2019). Once data on the spatial distribution of crop types are known, the next step is to identify irrigated agriculture: This is relatively simple in arid regions where, due to climatic conditions, little or no plant growth would be possible without irrigation. If agricultural land with green vegetation is found in such regions, irrigation can be assumed with a high degree of probability.

Figure 2 shows the false colour representation of a Sentinel-2 image of a part of the Spanish Duero area in May 2018; reddish hues indicate green vegetation, while other colours indicate dry vegetation or open ground. The course of the Duero river is visible in black colour. Time series of such images can be used to enable the detection of irrigated agricultural land as described above. Furthermore, the image shows several centre pivots, which are circular structures along the river. Due to the dry climate in the area, the reddish coloured areas are most likely irrigated.

False-colour Sentinel-2 image over a part of the Spanish Duero basin – a region characterised by irrigation. Reddish colours indicate green vegetation which is most likely irrigated. Author prepared figure using ESA Sentinel-2 data.
Figure 2: False-colour Sentinel-2 image over a part of the Spanish Duero basin – a region characterised by irrigation. Reddish colours indicate green vegetation which is most likely irrigated. Author prepared figure using ESA Sentinel-2 data.


In contrast, the situation is more difficult in regions where rain-fed agriculture and irrigated agriculture coexist, or where irrigation is only temporary and irregular - for example during longer dry periods. Here, irrigation can often only be identified from optical data using machine learning approaches (Ferrant et al. 2017) or on the basis of soil moisture signals derived from e.g. NASA's SMAP mission  (Lawston, Santanello Jr, and Kumar 2017). Deriving soil moisture from space, however, is an ongoing field of research and currently only passive microwave sensor are capable to deliver soil moisture conditions for all types of weather conditions while optical sensors are only able to provide information under (nearly) cloud-free conditions (Kerr u. al. 2016).

Furthermore, differences in the peak of spectral indices, e.g., the NDVI, have been proposed as a discriminatory feature (Meier, Zabel, and Mauser 2018)⁠ highlighting differences in the phenological development of irrigated and non-irrigated crops (Pervez and Brown 2010) but the authors concluded that the more humid the region the lower the accuracy of their classification attempt.

Another approach is to identify characteristic irrigation systems such as centre pivots. Centre pivots are irrigation systems in which an arm equipped with sprinkler glands moves around a fix point in circular paths. It is thus possible to irrigate the area of an entire circle (Heermann and Hein 1968). Many centre pivot systems provide the ability to adjust irrigation rates sector by sector, which makes it possible to grow different crops within one centre pivot. Some newer systems also offer the option of controlling the individual sprinklers separately. This allows sub-area specific irrigation rates. These systems can be mapped using,  optical satellite data, as has been done for several decades in the Nebraska Centre Pivot Inventory. However, such approaches are only applicable in areas where centre pivots are the dominant irrigation practice (such as the Interior Plains of Northern America). Overall, Ozdogan et al. (2010) emphasized the lack of studies aiming at separating irrigated from rain-fed agriculture which coincidences well with the findings of a review by Thenkabail et al. (2010).

Concerning the estimation of water availability, meteorological data on the frequency and magnitude of precipitation events, temperature and solar radiation alongside with information about soil conditions (e.g., permeability and retention capacity) provide important information about surface water resources and groundwater recharge (Rushton and Ward 1979). Additionally, by measuring the Earth's gravitational field, which is made possible by satellites of the GRACE mission (, groundwater bodies and their changes over time can be measured (Rodell and Famiglietti 2002). This is particularly important for so-called fossil aquifers, which were formed in geological periods when the climate was wetter than today and therefore can only be recharged to a limited extent under current climatic conditions.

Using both quantities, water demand and availability, the water balance can be calculated. If the demand exceeds the availability, a deficit occurs. If irrigation is in use, an attempt is made to counteract this deficit, whereby the artificially added water must be supplied from other reservoirs (groundwater, surface water, desalination of seawater). By means of hydrological models, which can calculate the water balance for river basins on the basis of meteorological measurements and the information on crop types and extent of the irrigated area obtained from satellite data, it is possible to make statements on the (over)use of water resources and the strain on ecosystems as well as the access to water resources. This is an important element to support decision and policy making and evaluate current agricultural practices with regards to ecologic and socio-economic sustainability. Coupling these models with climate change projections allows to estimate water balances under future climatic conditions (Fowler, Blenkinsop, and Tebaldi 2007) which is an asset for adapting water usage and agricultural practices to global warming.

Implications for Irrigated Agriculture

For the individual farmer, irrigation poses a central question: Since water resources, as shown above, are often limited and cannot be used in any quantity at any time, they should be used as efficiently as possible to avoid water stress and to stabilize or even increase yields. Thus, an optimization problem arises to generate a high or temporally stable output (yield) from a limited amount of input (water). According to numbers presented by Thenkabail et al. (2010) irrigation efficiency rates range between 40 to 62%. This leaves much room for improvement and shows the urgency of the issue. Space technologies can also be helpful here:

As shown in the previous section, water stress can be detected by optical remote sensing and physical modelling. Modelling allows for an assessment of the current status, and furthermore for projections into the future, so that predictions about the occurrence of water stress in the near future are possible.

Ideally, irrigation takes place before water stress occurs, so that the question of when a field should be irrigated can be narrowed down. Based on hydrological modelling, plant growth modelling and remote sensing, a recommendation for the irrigation quantity can be issued based on the determined water deficit (D’Urso 2001). If the utilised irrigation system allows for sub-area specific irrigation (i.e. the irrigation quantity is not constant for the entire field but spatially variable), irrigation recommendations can also be made spatially variable, whereby the limiting factor is the spatial resolution of the satellite used and technical constraints of the irrigation system (Sadler et al. 2000).

An example application called “Observing the Earth – Tackling Drought Issues for Food Security” which was funded by the European Space Agency (ESA) via its Food Security Thematic Exploitation Platform (FS-TEP),  uses optical earth observation data to determine specifically the actual irrigation needs of centre pivot systems in Zambia, where several successive dry periods led to water shortages. (read more here). By using remote sensing data, plant physiological and hydrological modelling, 1750m³ of water could be saved per irrigation pivot and day, which meant that an above-ground storage pond emptied less quickly, as could also be precisely determined using hydrological modelling. An example of a resulting irrigation recommendation is shown in Figure 3. Due to the high spatial resolution of the satellite Sentinel-2 (10 to 20m), sub-area specific calculations of the crop water demand could be carried out, which in the example amount to values smaller than 6 and 7 mm per day.

Crop water demand in mm per day determined using optical Sentinel-2 data and integrated plant-physiological and hydrological modelling for an irrigation centre pivot in Zambia. Courtesy Vista GmbH, 2018. Image Source: ESA
Figure 3: Crop water demand in mm per day determined using optical Sentinel-2 data and integrated plant-physiological and hydrological modelling for an irrigation centre pivot in Zambia. Courtesy Vista GmbH, 2018. Image Source: ESA)

The above provided examples show that irrigation can be made more efficient by means of remote sensing, and thus, reduce agricultural water demand, because a more targeted application of water with low loss rates is possible (Toureiro et al. 2017). However, there are also some challenges that can be summarized in three points:

  1. Increasing efficiency alone does not make irrigated agriculture sustainable. If, despite optimization, water abstraction rates are higher than the recharge rate, or if any conflicts over water use rights remain unresolved, the farmer may be able to operate in an economically viable way, but the social and economic side of the sustainability triangle remains unaffected.
  2. Previous recommendations on irrigation have mainly concerned large-scale farms that operate correspondingly large, mechanized irrigation systems. Small-scale farmers, on the other hand, hardly benefit at all, since their fields are either too small to be adequately detected by remote sensing or lack funds to allow investment in and optimization of irrigation systems. As Meier et al. (2020)⁠ showed, depending on the world region, a spatial resolution of 5m is necessary to capture agricultural areas adequately without spectral mixing of adjoining fields – but this spatial resolution is currently not achieved by most of the freely available earth observation satellites. Additionally, high risks associated with the required investment in technology might also play a role (George 2014) at the farm level.
  3. Although the approaches presented here are already being used operationally in some cases, there is room for improvement and further development: for example, information on soil properties and meteorological events is not available worldwide in the same quality (or not at all). Furthermore, the determination of evapo-transpiration rates is an essentially important factor in water balance modelling of agricultural canopies but is still subject to uncertainties and inaccuracies (Cammalleri et al. 2014; Velpuri et al. 2013). Despite these challenges first products with continental coverage were available to the public like the FAO's WaPOR portal (read more: Besides information on water productivity, layers on actual evapotranspiration are available on a monthly and annual basis for the time since 2009. All information of the portal was generated by means of freely available remote sensing data.

These three aspects are undoubtedly a great challenge. Due to the complexity of the subject matter and the existing geographical differences (climate, soil, water availability) and differences in the economic equipment (available irrigation technologies and possibilities of water treatment and extraction) as well as different cultivation methods, there is probably no one-fits-all solution.
Regardless of these local specifics, remote sensing can provide important inputs which can be used in appropriate integrated frameworks to explicitly address conflicting goals between different actors. Thus, space technologies play an essential role to more sustainable and resilient agricultural practices.


Irrigation is essential for global agricultural production and food security. As water resources are limited, their use must be made more sustainable. Being the largest consumer of fresh water, agriculture, should definitely contribute to developing (best) practices of sustainable water use. Remote sensing technologies and physical models offer ways to assess the agricultural water balance - i.e., water demand and availability. On the one hand, the derived information can be applied to map irrigated areas, despite the difficulties that currently exist especially in climatically wetter regions. On the other hand, water consumption can be adapted to the actual demand and thus, at best, reduced at the field level.

To conclude, this article aimed at highlighting that space technologies can be used to generate insights into agricultural irrigation practices, which in turn can lead to a more efficient use of irrigation at the field level. However, incorporating such technologically sophisticated information into decision-making and legislation requires a solid understanding of the fundamental issues involved, as well as a more detailed study of the multiple social, environmental and economic interactions that exist in agricultural practices.



Cammalleri, C, M C Anderson, F Gao, C R Hain, and W P Kustas. 2014. “Mapping Daily Evapotranspiration at Field Scales over Rainfed and Irrigated Agricultural Areas Using Remote Sensing Data Fusion.” Agricultural and Forest Meteorology 186: 1–11.

D’Urso, Guido. 2001. Simulation and Management of On-Demand Irrigation Systems: A Combined Agrohydrological and Remote Sensing Approach. Wageningen: WU thesis 2961 Proefschrift Wageningen.

Fabre, J, D Ruelland, Alain Dezetter, and B Grouillet. 2015. “Simulating Past Changes in the Balance between Water Demand and Availability and Assessing Their Main Drivers at the River Basin Scale.” Hydrology and Earth System Sciences 19 (3): 1263.

Ferrant, Sylvain, Adrien Selles, Michel Le Page, Pierre-Alexis Herrault, Charlotte Pelletier, Ahmad Al-Bitar, Stéphane Mermoz, et al. 2017. “Detection of Irrigated Crops from Sentinel-1 and Sentinel-2 Data to Estimate Seasonal Groundwater Use in South India.” Remote Sensing 9 (11): 1119.

Flörke, Martina, Christof Schneider, and Robert I McDonald. 2018. “Water Competition between Cities and Agriculture Driven by Climate Change and Urban Growth.” Nature Sustainability 1 (1): 51–58.

Fowler, Hayley J, Stephen Blenkinsop, and Claudia Tebaldi. 2007. “Linking Climate Change Modelling to Impacts Studies: Recent Advances in Downscaling Techniques for Hydrological Modelling.” International Journal of Climatology: A Journal of the Royal Meteorological Society 27 (12): 1547–78.

Geerts, Sam, and Dirk Raes. 2009. “Deficit Irrigation as an On-Farm Strategy to Maximize Crop Water Productivity in Dry Areas.” Agricultural Water Management 96 (9): 1275–84.

George, Thomas. 2014. “Why Crop Yields in Developing Countries Have Not Kept Pace with Advances in Agronomy.” Global Food Security 3 (1): 49–58.

Heermann, D. F., und P. Hein. 1968. „Performance Characteristics of Self-Propelled Center-Pivot Sprinkler Irrigation System“. Transactions of the ASABE 11: 11–0015.

Joshi, Neha, Matthias Baumann, Andrea Ehammer, Rasmus Fensholt, Kenneth Grogan, Patrick Hostert, Martin Rudbeck Jepsen, et al. 2016. “A Review of the Application of Optical and Radar Remote Sensing Data Fusion to Land Use Mapping and Monitoring.” Remote Sensing 8 (1): 70.

Kerr, Y. H., J.-P. Wigneron, A. Al Bitar, A. Mialon, und P. K. Srivastava. 2016. „Chapter 1 - Soil Moisture from Space: Techniques and Limitations“. In Satellite Soil Moisture Retrieval, herausgegeben von Prashant K. Srivastava, George P. Petropoulos, und Yann H. Kerr, 3–27. Elsevier.

Lawston, Patricia M, Joseph A Santanello Jr, and Sujay V Kumar. 2017. “Irrigation Signals Detected from SMAP Soil Moisture Retrievals.” Geophysical Research Letters 44 (23): 11–860.

Meier, Jonas, Wolfram Mauser, Tobias Hank, and Heike Bach. 2020. “Assessments on the Impact of High-Resolution-Sensor Pixel Sizes for Common Agricultural Policy and Smart Farming Services in European Regions.” Computers and Electronics in Agriculture 169 (October 2019): 105205.

Meier, Jonas, Florian Zabel, and Wolfram Mauser. 2018. “A Global Approach to Estimate Irrigated Areas – a Comparison between Different Data and Statistics.” Hydrol. Earth Syst. Sci. 22: 1119–33.

Mueller, Nathaniel D, James S Gerber, Matt Johnston, Deepak K Ray, Navin Ramankutty, and Jonathan A Foley. 2012. “Closing Yield Gaps through Nutrient and Water Management.” Nature 490 (7419): 254–57.

Orynbaikyzy, Aiym, Ursula Gessner, and Christopher Conrad. 2019. “Crop Type Classification Using a Combination of Optical and Radar Remote Sensing Data: A Review.” International Journal of Remote Sensing 40 (17): 6553–95.

Oweis, Theib, Mustafa Pala, and John Ryan. 1998. “Stabilizing Rainfed Wheat Yields with Supplemental Irrigation and Nitrogen in a Mediterranean Climate.” Agronomy Journal 90 (5): 672–81.

Ozdogan, Mutlu, Yang Yang, George Allez, and Chelsea Cervantes. 2010. “Remote Sensing of Irrigated Agriculture : Opportunities And.” Remote Sensing 2: 2274–2304.

Pervez, Md Shahriar, and Jesslyn F Brown. 2010. “Mapping Irrigated Lands at 250-m Scale by Merging MODIS Data and National Agricultural Statistics.” Remote Sensing 2 (10): 2388–2412.

Rodell, M, and J S Famiglietti. 2002. “The Potential for Satellite-Based Monitoring of Groundwater Storage Changes Using GRACE: The High Plains Aquifer, Central US.” Journal of Hydrology 263 (1–4): 245–56.

Rushton, K R, and Catherine Ward. 1979. “The Estimation of Groundwater Recharge.” Journal of Hydrology 41 (3–4): 345–61.

Sadler, E J, R G Evans, G W Buchleiter, B A King, C R Camp, and others. 2000. “Design Considerations for Site Specific Irrigation.” In National Irrigation Symposium. Proceedings of the 4th Decennial Symposium, Phoenix, Arizona, USA, November 14-16, 2000., 304–15.

Singh, Ajay. 2014. “Irrigation Planning and Management through Optimization Modelling.” Water Resources Management 28 (1): 1–14.

Thenkabail, Prasad S, Munir A Hanjra, Venkateswarlu Dheeravath, and Muralikrishna Gumma. 2010. “A Holistic View of Global Croplands and Their Water Use for Ensuring Global Food Security in the 21st Century through Advanced Remote Sensing and Non-Remote Sensing Approaches.” Remote Sensing 2 (1): 211–61.

Toureiro, Célia, Ricardo Serralheiro, Shakib Shahidian, and Adélia Sousa. 2017. “Irrigation Management with Remote Sensing: Evaluating Irrigation Requirement for Maize under Mediterranean Climate Condition.” Agricultural Water Management 184: 211–20.

Velpuri, N M, G B Senay, R K Singh, S Bohms, and J P Verdin. 2013. “Remote Sensing of Environment A Comprehensive Evaluation of Two MODIS Evapotranspiration Products over the Conterminous United States : Using Point and Gridded FLUXNET and Water Balance ET.” Remote Sensing of Environment 139: 35–49.

Wisser, Dominik, Steve Frolking, Ellen M Douglas, Balazs M Fekete, Charles J Vörösmarty, and Andreas H Schumann. 2008. “Global Irrigation Water Demand: Variability and Uncertainties Arising from Agricultural and Climate Data Sets.” Geophysical Research Letters 35 (24).

Wrachien, Daniele De, Mudlagiri B Goli, and others. 2015. “Global Warming Effects on Irrigation Development and Crop Production: A World-Wide View.” Agricultural Sciences 6 (07): 734.….