Applications of Remote Sensing in Agriculture

Introduction

Increasing interest in Remote Sensing and advanced technologies such as ArcGIS, QGIS, ENVI, ERDAS, and others have opened gateways to various research—agriculture being one of them. The easy viability of hyperspectral images (thanks to the USGS EROS Center) and the enthusiasm of the scientists in using their expertise in interpreting and analyzing these images to produce robust, sensible, and understandable products has not only benefited the scientific community but also the local stakeholders, decision-makers, and mankind as a whole.

Various attempts have been made regarding the study of agriculture using remote sensing. The basic idea is to use and interpret these remotely sensed images such as MODIS, Landsat, Sentinel, AVHRR, ASTER, and VIRRS and produces user-friendly datasets that have the capacity to distinguish between agricultural land and others. The National Land Cover Datasets (NLCD) produced by USGS, for example, have few classes that distinguish cropland from others. The Cropland Data Layer (CDL) produced by the U.S. Department of Agriculture goes even further to classify different crop types such as corn, rice, wheat, and others. The NASA-funded Global Food Security-Support Analysis Data (GFSAD30 ) provide high-resolution global cropland data and their water use that contributes towards global food security in the twenty-first century. The GFSAD30 products are derived through multi-sensor remote sensing data (e.g., Landsat, MODIS, AVHRR), secondary data, and field-plot data and aim at documenting cropland dynamics from 1990 to 2017. There are various other datasets produced at the global, regional, and local scales that focus on providing detailed, comprehensive, and consistent geospatial datasets at different temporal and spatial scales.

Remote Sensing in Agriculture

The application of remote sensing in agriculture ranges from simply identifying the patches of cropland to sophisticated applications like precision agriculture. The easy (free) assess to remotely sensed data (via USGS) and the advancement of geospatial analysis tools have triggered the studies in a vigorous way. However, due to some technical limitations, budget, and time constraints, the studies are lagging behind. Let's give a quick look at how remote sensing has helped in agriculture:

1. Land Cover Mapping

Land cover mapping is one of the most popular applications of remote sensing. Land cover mapping focuses on distinguishing different land cover types on the earth's surface. This land cover includes cropland, grassland, forest, water, urban area, and others. While mapping the land cover, the remotely sensed images such as Landsat are processed by integrating with multi-source ancillary datasets such as temperature, elevation, and precipitation using some robust classifiers such as decision-tree or support vector machine, then a hierarchical theme-based post-classification is done for generating land cover products at different temporal, spectral, and spatial scales.

Land cover mapping has helped researchers as well as stakeholders in decision/policy making, planning the agro-based economy, management of food supply, water resources management, and so on and so forth. The identification of crop types, on the other hand, helps in best management practices, deciding the crop types to cultivate, and forecasting the crop yields. The integration of crop types with current and historical weather and climate, crop yield models, soil characteristics, and market condition hell build a decision support system which in return helps in crop management including selecting crops based on field and soil type, developing the treatment plans to improve crop yields and reduce the risk of diseases or pest damage.

2. Precision Agriculture

Precision agriculture also called Precision farming and site-specific crop management refers to a group of techniques, technologies, and management strategies designed to optimize plant growth and farm profitability by adjusting treatments to suit variable biophysical conditions that occur within an agricultural field instead of applying the same treatment uniformly across the entire area. Precision agriculture uses new technologies like remote sensing, GIS, and GPS to increase crop yields and profitability while lowering the levels of traditional inputs needed to grow crops (land, water, fertilizer, herbicides, and insecticides). In other words, a controlled way of farming where farmers can decide what crops to plant, what nutrient (and what amount) to use, and when to farm based on the models and algorithms developed using advanced technologies like GPS and GIS tools.?

The use of remote sensing to monitor the crop condition during the growing season and GIS technology to analyze the results has made it possible to identify the problems and map their location. In addition, the use of GPS to collect field data of soil samples and integrate these results with current and historical weather and climate data helps make the best decisions in terms of choosing the best crop types, supplying the nutrients, and using the proper treatments.

Due to the advancement in remote sensing and added functionalities in GIS, the characterization, modeling, and mapping of almost any crop have been possible—which is to say, the future of precision agriculture heavily relies upon GIS and Remote Sensing.

3. Regional Crop Condition Monitoring

The remotely sensed data used in conjunction with historical and current crop data, weather data, and field reports provide an overall assessment of the crop and food supply situation--and integration of these data with digital maps of administrative boundaries, recent price and market conditions on food stocks and consumption rates can be used to predict the prospects of current crops. Various interactive Web-based tools are developed to maintain this information.

Some of the sources for regional crop condition monitoring to look at are:

1. Global Information and Early Warning System (GIEWS) of the United Nations Food and Agriculture Organization: It provides a detailed analysis and frequent reports on crop conditions and food supply situation for all countries.
2. Crop Condition Assessment Program of Statistics Canada: It provides maps, statistical data, and updated NDVI curves for the entire region weekly. The users can interactively choose regions of interest by name or by selecting them from maps at several scales.

4. Land use land cover change

The land cover indicates the physical land type such as forest or open water whereas land use documents how people are using the land. By comparing land cover data and maps over a period of time, coastal managers can document land use trends and changes. Land use land cover change simply refers to the conversion of a piece of land's use by humans, from one purpose to another. For example, land may be converted from cropland to grassland, or from wildland (e.g. tropical forests) to human-specific land uses (e.g. palm oil plantations).

Remote sensing has an important contribution to make in documenting the actual change in land use/land cover on regional and global scales. Using the time-series remotely sensed images, helps to detect the change in land cover, for an instance, grassland conversion to cropland and/or cropland conversion to grassland. Remote sensing allows an easy and quick way to look at different land-cover types (grass, shrubs, trees, barren, water, and man-made features) at different time intervals--and this snapshot of land cover at different time interval give a broader picture of how the land cover has changed over a period of time. The rate of change can be abrupt, such as the changes caused by logging, hurricanes, and fire, or subtle and gradual, such as the regeneration of forests and damage caused by insects. Using remote sensing techniques, we can keep track of the long-term natural changes in climate conditions, geomorphological and ecological processes, human-induced alterations of vegetation cover and landscapes, interannual climate variability, and the human-induced greenhouse effect and make the right decisions at times.

Knowing the land use and land cover trend, one can have a better understanding of how and where to plan agricultural practices and get benefits likewise.

5. Irrigated Land Cover Mapping

Another important application of remote sensing in agriculture is Irrigated Land Cover Mapping. Satellite observations provide reliable, economical, and synoptic data on the Earth’s surface. These data contribute to mapping land cover, including agricultural lands. Existing methods for agrarian land cover characterization have often been derived through image classification techniques. However, the variety of irrigated crops and the spatial patterns of their phenology require multi-temporal, consistent, composite vegetation growth information with sufficient spatial detail, along with a rich library of field reference training and ancillary data (e.g., climate and topography), to classify irrigated lands using satellite observations. Obtaining these parameters consistently at the national scale is a significant challenge.

Remotely sensed images with various spatial and temporal characteristics have been extensively used to map cropland extent and its dynamics. Specifically, the Moderate Resolution Imaging Spectroradiometer (MODIS) imagery provides a unique capability to map cropland extent at resolutions of 250–1000m. Landsat data provide less frequent coverage but enable cropland mapping at the much finer spatial resolution of 30 m, as exemplified by the crop-specific classes of the U.S. Department of Agriculture (USDA) National Agricultural Statistics Service (NASS) cropland data layer (CDL), the planted/cultivated classes of the U.S. Geological Survey (USGS) National Land Cover Dataset (NLCD), and Global Food Security-Support Analysis Data (GFSAD30).

Detailed and current information about where irrigated agriculture is now located and how its distribution and scope change over space and time can contribute to solutions to this challenge. Such information is critical if we want to fully understand the impact of agriculture on water use and formulate effective management policies for this limited resource.

6. Crop health monitoring

Remote sensing can be used to monitor the health and growth of crops by analyzing spectral data obtained from satellites, airborne sensors, or ground-based instruments. This information can help farmers identify areas of their fields that may need additional attention or water, fertilizer, or pest management.

7. Yield estimation:

Remote sensing can also be used to estimate crop yields by analyzing factors such as plant height, biomass, and chlorophyll content. This information can help farmers plan their harvests and manage their crops more effectively.

8. Soil mapping:

Remote sensing can be used to map soil properties such as moisture content, texture, and fertility. This information can help farmers optimize their fertilizer and nutrient management practices.

Overall, remote sensing has the potential to improve agricultural efficiency and sustainability and help farmers meet the growing demand for food production in a changing climate.