Although optical data have shown potential in the identification of crop types, the data is not reliable under cloudy conditions. SAR utilizes the temporal backscatter physical response of a crop and, along with machine learning techniques, can be effectively used for crop mapping and monitoring Sonobe et al.
Recent studies utilize coarse to fine resolution satellite imagery for crop type mapping Wardlow and Egbert, Some of the well-known approaches for crop type mapping using different sensors and resolutions are listed in Table 1.
In this study, we developed a system for in-season wheat sown area mapping by harnessing the power of multisensory remote sensing imagery optical and SAR and cloud computing GEE techniques Dong et al.
The system is designed keeping in mind the challenges in Afghanistan and provides the capacity for operationalization. The system can provide independent and evidence-based information on the status of annual crops at the province level. Ingesting field data at regular intervals for different seasons in the system will lead to higher accuracy in crop area estimates at the province level. The Hindu Kush mountain range divides the country into three very different geographic regions: a The central highlands, characterized by dry hot summers and very cold winters; b the southern plateau consist of sandy deserts with arable lands along the rivers; c the northern plains, which are highly fertile and include most of the land under agriculture.
The total area of the country is , sq km with a population of Total arable land is 6. Wheat, rice, barley, and maize are the main cereal crops grown in the country, with wheat accounting for Thus, wheat is the most important crop for the food security of the country Ahmad, However, other than cereals, fruits, vegetables, and opium poppy are also important crops.
This study used Sentinel 1 SAR and Sentinel 2 multispectral optical satellite images as the main data sources Table 2. High-resolution images from Digitalglobe and Airbus Company Pleiades were also used for collecting reference data for training and validation. Apart from satellite datasets, agriculture mask irrigated and rainfed from the Afghanistan land cover FAO, was used to aid in crop mapping. The land cover maps Figure 1 have eleven land cover classes viz.
Reference data were collected from various sources for training and validation of the classification model. The field survey was conducted by professionals from MAIL to collect samples from the crop field. During the collection of samples, the location of the crop field together with field multi-direction photographs for different crops were collected. The second set of reference data were collected by MAIL during a crop cutting survey that covered 17 provinces.
Most of the reference data collected through field and crop cut survey were mainly from wheat fields. Very few samples were collected from non-wheat crops, such as vegetable farms, orchards, and vineyards. Samples for other crops were generated through visual interpretation by relevant experts and using earlier land cover maps and time-series images. The reference data obtained through field and crop cutting survey covered only 25 out of 34 provinces.
For the remaining 10 provinces, samples were generated from high-resolution images through visual interpretation and analysis of NDVI time-series for the current and previous years.
Google Earth images and Digital Globe high-resolution images acquired during October —June were used for the interpretation. Overall 16, reference? Additional reference data were collected through visual interpretation of Airbus Pleiades images. The distribution of reference data and sources are shown in Figure 2.
The goal of the classification algorithm was to distinguish the phenology of wheat from other crop types and land cover. The study of LSP is important to understand vegetation-growth pattern changes Myneni et al.
Satellite-based analysis of LSP addresses the development patterns in photosynthetic biomass by way of derived vegetation indices Ahl et al. Phenology is measured commonly by i onset of greening, ii onset of senescence, iii peak development during the growing period, and iv the length of the growing season Hudson and Keatley, Various methods have been used for the assessment of phenology including threshold, derivative, smoothing, and model-based methods Hudson and Keatley, Among these, the threshold-based method is the simplest and is used by many researchers.
In the threshold-based method, the values of VI are plotted against time of year and single values are chosen to define different stages of phenology Karlsen et al. Some authors use single arbitrary thresholds, e. Afghanistan has diverse topographic and climatic conditions resulting in wide variability in growing seasons across the entire landscape. Knowledge of the growing season is important for the acquisition of satellite data. The crop calendar is a tool that provides information on the sowing, growing and harvesting stages of crops in our case, wheat.
The crop calendar information can also be used for crop type mapping using the satellite data. Broad crop calendars at a province-level were provided by MAIL; these were compiled in The calendars Figure 3 were utilized as a starting point to characterize the timing of phenological stages of wheat.
In this study, optical and SAR data were utilized in two steps in the process of mapping wheat areas. The flowchart of the detailed methodology is shown in Figure 4. The description of the methodology is given in the following sections. In the first step, reference data collected from wheat and other crops for 34 provinces of Afghanistan were subjected to quality check.
This is because some of the sample points collected by the field staff were not always inside the crop fields. Accordingly, adjustments were made to correct the location based on three criteria: a direction and orientation of the field photographs; b phenological characteristics of the crop; and c visual interpretation through high-resolution Google Earth images.
For each province, the reference points were merged and divided randomly into two categories, i. The wheat mapping was done at the provincial level. The cloud masking utilizes Sentinel-2 Band QA60, a quality flag band, to identify and mask out flagged cloud and cirrus pixels.
After that, the median-compositing function was used on the cloud-masked Sentinel-2 images to generate a per-pixel median composite of each of the multi-spectral bands and the indices for every province Hird et al. NDVI is an effective means to characterize these growth patterns during the crop cycle Menenti et al.
Using randomly collected training samples and the seasonal composite of sentinel 2 images, NDVI thresholds were identified to separate the wheat from other crops during sowing, peak and harvest time at the provincial level. NDVI thresholds were identified for different seasons sowing, peak and harvest and were different for each province. The difference in the NDVI thresholds for different provinces is mainly because of the shift in the phenological cycle early and late sowing of wheat and other crops.
To define the thresholds for separating the wheat from other crops, the minimum and maximum values of NDVI were calculated for wheat using the training samples. The separation was done as below:. The NDVI threshold derived for — were specifically derived on the basis of collected ground sample points from the field.
The NDVI threshold values depend on various factors, such as: i whether it is a dry year or wet year; ii whether there has been early or late sowing of the crop; and iii atmospheric conditions at the time of data acquisition. Hence, the NDVI thresholds are specific to the — growing season. However, Equations 3 — 5 can be utilized for deriving the NDVI thresholds for other years if field sample points for that particular year are available.
Compared to the other crops, wheat has a different cropping calendar and growth patterns Figure 3 , except for barley, some vegetables and opium poppy have growth patterns similar to wheat. Due to the similarity in cropping season and growth pattern, it was difficult to accurately separate barley, opium poppy and some vegetables using Sentinel 2 data despite having a 5-days temporal resolution. This is because of the limited availability of cloud-free images which makes it difficult to utilize the image of a specific time where wheat can be separated from other overlapping crops.
To overcome this limitation, the wheat area map obtained from the optical image analysis was further refined using Sentinel-1 SAR data. SAR sensors have all-weather capability to acquire images and are sensitive to plant structure; however, to use the SAR S1 based classification alone would require much more sample data for all the crops.
The initial separation of crops using optical data enabled the use of SAR for only separating the wheat from crops with similar crop calendar and phenology. The S1 data has a consistent time-series in terms of incidence angle and has a wide scope in mapping different crops Inglada et al. Pre-processing includes orbital file correction, thermal noise removal and terrain correction.
Monthly median composites were developed for the entire wheat crop cycle i. The analysis of S1 data shows the difference in response patterns from different crops. However, the variability of responses shows overlap and makes it difficult for the threshold-based separation Figure 5. Thus, a Random Forest RF classification technique was utilized using time series S1 data and training points to separate the wheat from other crops.
The Random Forest RF randomly selects a subset of training sample through replacement to build a single tree, i. There are two important user- define parameters in RF, i. The generalization error always converges as the number of trees increases Breiman, Therefore, RF classifier doesn't have any issue of overfitting which can also be attributed to the Strong Law of Large Numbers Bercovici and Pata, There is no well-defined rule for selection of the number of trees.
However, Guan et al. Also, the increased number of trees would require high computation. In this study, we have used the number of trees as The selection was based on the hit and trial method. Secondly, the number of variables highly affects the performance of RF classifier, which is usually set to the square root of the number of input variables.
The application of Random Forest was applied within the classified mask generated from the optical image analysis. Pest control becomes a less serious issue when noticed on time.
Field monitoring, and satellite monitoring, in particular, is an efficient tool successfully applied by advanced agrarians. Crop Monitoring is an online platform that covers all the above-mentioned aspects to enhance field performance wisely.
Reduction of personal inputs, preservation of natural resources, and eliminated atmosphere pollution are eloquent advocates to opt for online software with multiple beneficial features of precise and sustainable agriculture. The recent innovations allow you to know exactly, where, what, when and how you should sow to get maximum profit with the least efforts and expenses.
With Crop Monitoring, it is very convenient to keep all records in order. What is more, farmers are notified about the scheduled activities or upcoming hazards in advance, so that they would be able to battle them decently.
Different growth stages have different indices. Alongside that, biomasses of certain crops tend to correlate with their own specific indices more accurately than with others. Crop Monitoring provides detailed information on the development of each plant species based on their vegetation indices, crop rotation, and growth stages. Furthermore, not only farmers win from this software product. It is beneficial for other players of the agricultural process as well.
Thus, insurance agents can easily check what was grown in the previous years, and to estimate possible risks of poor years if crops are rotated wrong. Whatever party you represent, Crop Monitoring gives your own advantage to be in the know. EOS Data Analytics partners up with Cambridge Technology to introduce precision agriculture and forestry technologies to growers of palm oil in Malaysia, modernizing production and decelerating climate change.
Fertilizer burn reduces yields and can kill the crops. Proper prevention allows mitigating potential risks. By knowing how to deal with fertilizer burn, farmers can mitigate their losses. EOS Data Analytics, a data-driven satellite analytics platform for farm management, announces partnership with Complete Farmer, a digital agriculture platform.
Areas with a warm and damp climate are not suited for wheat growing. During the heading and flowering stages, excessively high or low temperatures and drought are harmful to wheat.
Cloudy weather, with high humidity and low temperatures is conducive for rust attack. Wheat plant requires about c optimum average temperature at the time of ripening. The temperature conditions at the time of grain filling and development are very crucial for yield.
Temperatures above c during this period tend to depress grain weight. When temperatures are high, too much energy I lost through the process of transpiration by the plants and the reduced residual energy results in poorer grain formation and lower yields. Wheat is mainly a rabi winter season crop in India. Wheat is grown in a variety of soils of India. Soils with a clay loam or loam texture, good structure and moderate water holding capacity are ideal for wheat cultivation.
Care should be taken to avoid very porous and excessively drained oils. Soil should be neutral in its reaction. Heavy soil with good drainage are suitable for wheat cultivation under dry conditions. Soy 3. Hay 4. Wheat 5. Acids can corrode some natural materials. Acids have pH levels lower than 7. Also called industrial agriculture. Native American usually does not include Eskimo or Hawaiian people. Pesticides can be fungicides which kill harmful fungi , insecticides which kill harmful insects , herbicides which kill harmful plants , or rodenticides which kill harmful rodents.
Curds are used to make rubber. Sustainable agriculture aims to cultivate the land so it may be used by future generations. The audio, illustrations, photos, and videos are credited beneath the media asset, except for promotional images, which generally link to another page that contains the media credit.
The Rights Holder for media is the person or group credited. Caryl-Sue, National Geographic Society. Dunn, Margery G. For information on user permissions, please read our Terms of Service. If you have questions about how to cite anything on our website in your project or classroom presentation, please contact your teacher. They will best know the preferred format. When you reach out to them, you will need the page title, URL, and the date you accessed the resource.
If a media asset is downloadable, a download button appears in the corner of the media viewer. If no button appears, you cannot download or save the media. Text on this page is printable and can be used according to our Terms of Service. Any interactives on this page can only be played while you are visiting our website. You cannot download interactives.
Agricultural communities developed approximately 10, years ago when humans began to domesticate plants and animals. By establishing domesticity, families and larger groups were able to build communities and transition from a nomadic hunter-gatherer lifestyle dependent on foraging and hunting for survival.
Select from these resources to teach your students about agricultural communities. Technological innovations have greatly shaped agriculture throughout time.
From the creation of the plow to the global positioning system GPS driven precision farming equipment, humans have developed new ways to make farming more efficient and grow more food. We are constantly working to find new ways to irrigate crops or breed more disease resistant varieties.
These iterations are key to feeding the ever-expanding global population with the decreasing freshwater supply. Explore developments in agricultural technology and its impacts on civilization with this curated collection of classroom resources. In the United Nations General Assembly adopted 17 sustainable development goals designed to transform our world by The second goal is to end hunger, achieve food security, improve nutrition, and promote sustainable agriculture.
This initiative strives to help us rethink our global food infrastructure, from how we grow and harvest food to how we consume it. With a growing global population, we will need to implement innovative, sustainable practices to ensure equitable food access and eliminate fears of going hungry. Use these resources in your classroom to teach your students about sustainable agriculture and tactics that could potentially end world hunger. Hundreds of years before the arrival of European explorers, the ancient civilizations of South America developed rich and innovative cultures that grew in and amongst the geographic features of their landscape.
The most famous of these civilizations is the Incan Empire. Emerging in C. The Inca relied on the Pacific Ocean and major rivers originating in the Amazon Basin for fishing and trade, as well as rich plant and animal life that they supported.
The Inca constructed inns, signal towers, roads, and massive forts such as the famous Machu Picchu, the ruins of which continue to teach archaeologists about the Incan Empire. Learn more about the history and rich culture of the Inca and the ancient civilizations of South America with this curated resource collection. Made up of a wide variety of plants grown for consumption or for profit, crops can be used for food, to feed livestock, for textiles and paper, for decoration, or for fuel.
Our MapMaker Interactive layers show how many tons of cassava, maize, plantains, potatoes, rice, sorghum, soybeans, sweet potatoes, wheat, and yams were produced per country as an average from to Join our community of educators and receive the latest information on National Geographic's resources for you and your students. Skip to content. Twitter Facebook Pinterest Google Classroom. Encyclopedic Entry Vocabulary. Grains, such as the sorghum these farmers are carrying, are the most popular food crop.
Photograph by W. Robert Moore.
0コメント