Wednesday, 27 March 2013

Pervious & Impervious

High Resolution Imagery based Classification – Pervious & Impervious

Many management and policy decisions of local, county and regional agencies require timely, accurate data on urban land growth. For example, urban land information is critical for planning of urban infrastructure such as roads, water and sewers. Other policy decisions relating to public services also rely on land information, such as, siting new schools, retail development, public parks and landfills. In addition, urban growth information is needed to identify natural resource areas in need of protection.

 One important aspect of urban land information is the extent of impervious surfaces such as roads, parking lots and rooftops which lead to decreased infiltration of rainfall and increased storm water runoff. To address land use information needs many cities, counties and other governmental units routinely acquire high-resolution digital imagery. However, processing this information has historically been labor intensive and costly. A number of recent efforts have been directed at reducing the effort and cost of classifying digital imagery by using automated and semi-automated classification methods for monitoring and mapping urban land cover and imperviousness.

Although features can always be extracted from high resolution imagery through manual means, the fact that it is collected in digital format and is multispectral makes it a good candidate for an automated approach. The standard automated mapping approach to date has been to use unsupervised or supervised classification techniques. These traditional methods are so-called “per-pixel” classifications, relying entirely upon the spectral information in an image, while neglecting the spatial arrangement of the pixels. If we were trying to detect features on a high resolution image, such as Quickbird2 and were to use an unsupervised classification to detect these features, we would get class values that represent information at a finer scale than the features in which we are interested.

Area Covered:
The LU/LC was tested for below listed areas:
·         Miami Dade
·         San Antonio

Scope of Work:
Current case study required to carry out satellite based study using High Resolution Satellite Image to find out the different LU/LC classes. Hybrid classification techniques were used to extract the required information in raster format into:
  • Non woody
  • Woody
  • Impervious
  • Water
  • Bare
  • Swimming Pools

Inputs Used:
The inputs that were used for this case study were:
·         Image type  - High Resolution Satellite Images
·         Images Resolution                  -           2.0 Ft
·         No. of bands used       - 03
Software Used:
The following software was used:
l  Erdas Imagine 9.1

The methodology followed in this project is
·         Image preparation (Sub-setting image as per area of interest)
·         Image classification and compilation
·         Assessment (Quality Analysis/Error removing)
·         Final shipment

1.      Image Preparation
 The imagery was delivered from the contractor as high-resolution satellite image. Any image processing software can used which is a pixel-based classifier. With its help first of all we generated our area of interest by subsetting the image. Later, we classified that image subset as per the client’s requirement.

1.      Image Classification and Compilation

In Digital image classification the analyst uses the spectral information represented by the digital numbers in one or more spectral bands and attempts to classify each individual pixel based on this spectral information. This type of classification is termed as ‘spectral pattern recognition’. The objective is to assign all pixels in the imagery to particular classes or themes. The resulting classified image comprises a mosaic of pixels, each of which belongs to a particular theme and is essentially a thematic map of the original image.
The commonly used classification methods are Supervised and Unsupervised classification. Normally, multispectral data are used to perform the classification. The objective of image classification is to identify and portray, as a unique gray level (or color), the features occurring in an image in terms of the object or type of land cover these features actually represent on the ground.
Luminous ETS has adopted an automated process of classification followed by manual cleaning. These automated processes include:
2.1 Hybrid Classification

Hybrid classification is the use of both supervised and unsupervised techniques to classify an image. Both methods, when taken singularly, have their drawbacks. Supervised classification requires knowledge of the area and/or detailed field data. Unsupervised classification requires minimum initial input from the analyst, but the output takes a significant amount of time to assign the computer-generated clusters to a known land cover.  Hybrid combines the benefits of both techniques.

2.2 Land Cover Mapping

The term land cover relates to the type of features present on the surface of the earth whereas the term land use relates to the human activity or economic function associated with the specific piece of land. Depending upon the level of mapping details, its land use can be described as in urban use, residential use or single-family residential use.
The total number of classes which were classified are:
·         Non-woody: Grass, Turf, Non-woody wetland (Cattails, rushes, etc.)
·         Woody: Trees and Shrubs
·         Impervious: Roads, buildings, driveways, sidewalks, gravel and hard-packed dirt roads
·         Water: Open water or wetland where the land is substantially inundated. Eg. Open tanks.
·         Bare: Man-made and natural barren: construction sites, mines, beaches, etc.
·         Swimming pool

1)      Non-Woody: Non-woody plants (also called herbaceous plants or herbs) are plant with a relatively short-lived shoot system (compared to woody plants).  Most angiosperms lack a vascular cambium, i.e., they are non-woody herbs or herbaceous.
a) Miss classification: During the automated classification process it occurred that pixels are classified wrongly. Some impervious and mostly woody pixels are randomly scattered in Non woody classes.
b) Shadow: In automated classification the shadow of the trees were classified in Water due to the Black signature.
Solution: The non woody class is masked and extracted separately to pay attention on this class. It is further classified and picked accurately. When automated process and picking of the pixels completed then it is manually recoded with the help of AOI creation to fulfill the quality requirement of the client.

2)      Woody: A woody plant is a vascular plant that has a perennial stem that is above ground and covered by a layer of thickened bark. Woody plants are adapted to survive from one year to the next; the stem supports continued vegetative growth above ground from one year to next.

a) Miss classification: The main problem in this class was misclassification by the automatic classification process. Though the maximum area was covered but in some of the area to be enhanced. The non woody pixels were found in the woody classes.
Solution: The pixels of this class are clumped and then recoded in the defined class. Further viewing the quality of the images it was thought that the image should be masked and then recoded.

3)      Impervious: The impervious objects are the features by which water can’t be percolated.
Shape: This class is classified nearer to the desired quality but problem was in the shape of the feature. The impervious features did not pick the exact shape.
Solution: The general classification and processing techniques are not feasible to extract exactly the features in their proper shape. To overcome with these problems a semi automated classification technique was used which shows good results in a very less manual efforts and time.

4)      Water &Swimming Pool: To extract these classes manual recoding technique was preferred because it can be identified easily and are found less. Manually AOI was created and then it is recoded in the main classified image.

5)      Bare: These are the land which remains as such which is not in use for anyone.
Miss classification: This class is misclassified between impervious and bare.
Solution: Manual recoding technique was adopted to overcome with this problem.

2.3 Generation of Land Cover Map

Geometrically and radiometrically corrected satellite images were used for classification. Classification was carried out using the unsupervised classification in which the number of classes, maximum iterations and convergence threshold were given first. Further, the classification was carried on by analyzing the spectral values of the pixels as well as its tone. Finally, the classes will be finalized as mentioned by the client.
      At many times it becomes difficult to classify certain pixels especially the ones which fall in water and shadow areas. Therefore, to avoid such confusion neighborhood function was run in order to remove water and shadows. For it kernel filter of 3*3 (function: majority) was used. This was saved in a particular file for further use. Masking is an automated process which is performed to overall enhance the image. During the image classification process its obvious that many pixels can be left without having a proper class. This process enables us to extract a particular class in which pixels are left unclassified. This was performed for the woody and non-woody classes.
After which very small pixels were clumped and eliminated. Clump performs a contiguity analysis on a single layer RASTER. Each separate raster region, or clump, is recoded to a separate class. The output is a single layer raster in which the contiguous areas are numbered sequentially. Then mosaicking of the extracted classes from woody were done with the earlier saved file on which neighborhood function was run. Further, the woody class was masked again so that more information could be extracted about the woody and impervious classes respectively. Once again the above mentioned steps were carried out.  Finally, both the images were mosaicked again for which neighborhood function was run. This time it was done in order to remove line feature of woody class. Finally, it was mosaicked with water body, which was separately extracted.

2.      Assessment (Quality Analysis/Error removing)
After the classification of the image is done, the final assessment was carried out which comprised of the manual editing along with the Quality Check/Quality Analysis, so as to remove the errors.

3.     Project Shipment:
      The following outputs were prepared-
i)                    Classified LU/LC Map
ii)                  Thematic Maps in Img/TIFF format

For more information education or Business queries contact me at

Agrometeorology – Current Demands and Future Challenges

Agrometeorology has advanced over decades from a descriptive to a quantitative science using physical and biological principles. Now, the challenge is to balance the continuing need for increased productivity with new and growing concerns about climate change, climate variability and the associated environmental impacts. The farming community in India especially is becoming more and more aware of the weather and its impact on the crop at different phenological stages. During this decade farmers’ awareness also increased substantiallyabout the increasing and efficient agrometeorological services mainly weather-based agroadvisory services.Business community is now gearing up its activities based on monthly weather forecast along with the agroadvisories and it needs to be addressed immediately by the Agrometeorologists. Decision Support System with major focus on pest and diseases based on simple thumb rules for various export potential crops including some cash crops and horticultural crops needs to be developed. Another demand and future challenge is to help the Agricultural insurance sector which plays a pivotal role in settling the farmers’ claims.It becomes absolutely necessary to provide research, monitoring and advisory services to the farming community to miminize the weather related or influenced losses and maximize the production through early fore-warning systems regarding possible pests/diseases incidence, their remedial actions etc. Another important area viz., weather based commodity trading is gaining fast momentum which needs specific attention.It is time that the climate and weather should be looked as a resource but not as a hazard only.Public Private Participatory approach is fast emerging asan efficient and effective way for some of the out-reach programs in agriculture sector. Probably, it is the holistic approach integrating soil, weather and physiological aspects of crops with remote sensing/GIS techniques that may address these emerging issues in the next two decades which stands a great challenge to the Agrometeorologists. As an attempt to meet the future challenges, one step forward could be introduction of some of the weather based agribusiness topics in the curriculum of Agrometeorology.

India is an agrarian based country and hence agriculture plays a significant role in the overall socio-economic fabric of India. Weather becomes more significant in crop management and an expert knowledge of past, present and future weather will help to solve a host of other problems.  It is a known fact that the success of farming is intimately related to the prevailing weather conditions and quantitative information much in advance would become necessary to predict/assess the yields to a great accuracy. Agriculture remains one of the few areas for which accurate short-term and extended-period forecasts can create a material benefit. There is an urgent need to integrate the weather forecast with a real time decision support system leading directly to solutions/services to the farmers which eventually help the farmers in saving their crop from all possible losses.

It is a well-known fact that most of the farmers are adversely affected by climate risks in their farming, which include increase in temperature, decrease in rainy days, increase in precipitation intensity and amount, shorter winter periods, decreased ground water availability, increased occurrence of drought and floods, increased duration of water logging etc. which in turn result in yield losses.Farmers, often with the support of other machinery have been trying to adapt to these climate risks in different ways while only few of them are aware of the crop insurance and its benefits though the awareness is increasing very fast. Most of these adaptations are technological often pushed or promoted by research and extension agencies involved in agriculture like changing the date of planting, cultivating new paddy varieties that can better tolerate water-logging, introducing new crops/diversification to vegetables and adopting SRI (Systems of Rice Intensification) in paddy etc.  The present paper gives a brief note about some of the services and agencies that arelooking for agrometeorologists for holistic approach. The current demands as well as future challenges in each of them are emphasized. 

 Weather influences agriculture in a profound way. Despite the technological advances, Indian farmers are mostly dependent on seasonal rains which are highly variable in time and space. If farmers have advance information about the probable occurrence of events such as depressions, drought, storms, floods and heat waves in their geographical locations, the impact of these events on farmers’ livelihood can be reduced to a great extent. Thus weather forecasts are of great importance to agricultural activities Much research has gone into characterization of various crop environments, quantification of crop-weather interactions in relation to crop yield, crop weather modeling and crop-pest-weather dynamics. It is time to consolidate these findings in different locations, make use of this information and reach the farmers through agroadvisories on a larger scale in a more realistic way. At present, the India Meteorological Department in collaboration with the State Agricultural Universities is giving the Integrated Agromet Advisory Services which include weather forecast coupled with expert advises on crop planning, disease and pest incidences as well as different farm operations like fertilizer, pesticide applications etc. in developing timely weather based agro-advisories. These weather based agro-advisories can be used to take up prophylactic plant protection measures, fertilizer application, irrigation scheduling etc. These can also be used to take up appropriate measures in day to day field operations to minimize the risk involved in agricultural production (Rathore and ParvinderMaini, 2008).
A lot needs to be done to make these advisories more and more farmer-friendly. It is time that the Agrometeorologists respond to the urgent/pressing needs/demands of the farmers who insist on receiving a more accurate forecast along with actionable agroadvisories. Block level forecast is the need of the hour and the farmers are looking for location specific accurate short, medium and long term forecasts which will eventually help them in proper planning of the crops, varieties, sowing schedules and other farm operations for minimizing the yield losses leading to a higher grain/seed yields.
For real-time forewarnings in agrometeorology, the reliability of regular, specialized information along with the real-time crop status is critical. Agrometeorological decision-making in agricultural operations for healthy crops or crops endangered by pests, diseases and/or other environmental disasters needs weather forecasting and climate prediction, where that is possible, to the required accuracies(Blench, 1999).Use of advisories by progressiveand medium size farmers  related to accurate predictions of sowing date, timing of irrigation and fertilizer use strategies, is slowly on the rise.Crop-weather models that are mainly used for operational yield forecasting and prediction of phenological development have been generated for a large number of crops in our country (Aggarwal et al, 1997, 2005and2006).They have different degrees of complexities and many of these models need to be further refined and tested before widespread practical application may be expected. Current research is being focused on detailed soil–water–crop relationships, determining the adjusted crop genetic coefficients, bridging simulation model outputs with user needs for applications, and developing practical decision support systems. The results of these findings may be of immense utility and form an important component of the decision support system which when coupled with the crop status on a regional scale would go a long way to alleviate the farmers’ problems to a great extent. 
Recent field survey in some parts of Karnataka, Andhra Pradesh, Orissa and Chhattisgarh  revealed that there is presently still a considerable gap between the information needed by poor, small-scale farmers and that is presently available (Skymet,2012-personal communication).
For a vigorous plant growth and rich harvest, satellite based remote sensing-aided weather based agro-advisories may be a promising way which would enable the farmers to take the most appropriate actions on real time basis, at a regional scale. Satellite remote sensing technology is increasingly gaining recognition as an important source of many agricultural applications as it is superior to the traditional methods in terms of accuracy and saving of time (Fig.1).  In addition, Geographic Information Systems (GIS) technology is becoming an essential tool for combining various maps and satellite information sources in models that simulate the interactions of complex natural systems. 

This will not only help in planning, advising and monitoring the status of the crops but also will help in responding quickly for taking immediate planning or remedial actions.  Planning for seeds distribution, fertilizer supply/requirements, supplying/relocating of sowing/harvesting equipment, procurement of crop from mandies/markets, etc., can be tackled effectively through information derived using these technologies.

Apart from benefitting the farmers, these services would help the Tractor manufacturers and farm machinery manufacturers to make tactical decisions on movement of their farming machinery (like tractor-driven combined harvesters etc.). Also water pump motor companies in deciding about need based supply or deploying the appropriate Horse Power motors to different locations. It is needless to emphasis the underlying fact that all these are challenges that the Agrometeorologists are required to meet right now. 

Fig 1: Flow diagram for integration of remote sensing aided weather-based agroadvisories

Crop insurance is one of the main non-structural mechanisms used to reduce risk in farming. A farmer who insures his crop is guaranteed a certain level of crop yield or income, which is equivalent, for instance, to 60 or 70 per cent of the long-term average. If, for reasons beyond the farmer’s control and in spite of adequate management decisions, the yield drops below the guarantee, the farmer is paid by the insurer a sum equivalent to his loss, at a price agreed before planting.
Crop insurance schemes can be implemented relatively easily when there is sufficient spatial variability of an environmental stress (such as with hail), but they remain extremely difficult to implement for some of the major damaging factors, such as drought, which typically affect large areas.
One of the basic tools for insurance companies is risk analysis (Abbaspour, 1994; Decker, 1997). Crop forecasting models play a central part, when run with historical data, they provide insight into the variability patterns of yield (WMO, 2010) Technical Bulletin 134.
The first crop insurance program in the country was introduced in 1972-73 by the ‘General Insurance’ Department of Life Insurance Corporation of India on H-4 cotton in Gujarat. Later, the then newly set up General Insurance Corporation of India took over the experimental scheme and subsequently included Groundnut, Wheat and Potato and implemented in the states of Gujarat, Maharashtra, Tamil Nadu, Andhra Pradesh, Karnataka and West Bengal.Professor V. M. Dandekar, often referred to as the “Father of Crop Insurance in India”, suggested an alternate “Homogeneous Area approach” for crop insurance in the mid-seventies.                                                               
After continuous evolution, the Comprehensive Crop Insurance Scheme (CCIS) was introduced with effect from 1stApril 1985 by the Government of India with the active participation of State Governments.The implementation and administration of crop insurance schemes, which were being done by General Insurance Corporation of India (GIC), was taken over by Agriculture Insurance Company of India Ltd. (AIC) since its commencement of business from 1st April 2003.
It is imperative that crop yield hence the crop insurance are closely related and dependent on prevailing weather conditions during the entire crop growth cycle. The Agrometeorologists not only make use of the weather forecast but also should be able to estimate the likelihood of unusual weather events and their potential impact on every single farmer’s fields/crops (quantitative analysis), which will be a great challenge.  This creates a great hope that insurance policies based on clever weather based analytics will one day also protect Indian farmers against the vagaries of the weather.
In many rural areas, disaster often strikes poor farmers hard, forcing them to make choices that drag their families deeper into poverty. To survive, they might have to sell their tools for cash to buy food, or take their children out of schools to save on fees. With weather insurance, farmers can protect the investment they make in their crops, and feel confident in taking out loans for fertilizer and better seeds to improve their harvests. (Brian Kahn,IRI,2012).
Now the Agri-insurance companies including some private banks are looking forward for weather based agroadvisories to be disseminated through SMS or Voice mail to the end users (farmers) apart from assessing the exact yield at individual Grampanchayat / village level for their judicious payouts. 
Agricultural commodities are those which are living things grown by farmers or ranchers, or, in some cases, such things which have been minimally processed. Agricultural commodities are often referred to as soft commodities to distinguish them from metals, energy and other non-agricultural commodities.There are many factors that can impact the supply of commodities like weather, acreage covered, production strikes, crop pests/diseases and technology. Given a particular locality or region, the produce is totally dependent on the weather more so when they are rainfed crops. There are a handful of grains and meat which make up the core of agricultural commodity trading while other agricultural commodities include rape seed, milk, cocoa, coffee, sugar, frozen orange juice concentrate, and cotton. Livestock include hogs, pork bellies and cattle. These commodities are traded in a variety of different grades and types, and there are other exchange-traded agricultural commodities.

A farmer would like a guaranteed minimum return and would prefer money now over money later. A purchaser would like to plan on a maximum price for an agricultural product now and so has an incentive to mitigate the risk of a price rise at harvest time. Commodities markets were established in the ancient world in rice and other grains. Supply and demand and unpredictable market conditions have always added price volatility to agricultural commodities markets, and trading and hedging techniques have been developed over long periods of time to help everyone in the market manage their risk.While spot trading, with physical inspection and delivery does take place, most trading in financial markets is done through futures contracts. A futures contract is an agreement between a producer and purchaser that a transaction for a certain quantity of a specific commodity will take place at a future date and at a particular price. This smooths out the volatility for both parties and provides liquidity to the market. Weather forecast thus has an important role to play in the trading. The trading community is looking forward to get the medium, long range and seasonal forecast for this speculation of output of the commodities.
Advance knowledge of the likely volume of future harvests is a crucial factor in the market. Prices fluc­tuate as a function of the expected production with a large psychologi­cal component.
In fact, prices depend more on the production that the traders anticipate than on actual production. Accurate forecasts are, therefore, a useful planning tool. They can also often act as a mechanism to reduce speculation and the associated price fluctua­tions, an essential factor in the availability of food to many poor people.
In India, prior to the introduction of commodity futures market, the commodity prices were found to have experienced high volatility. With the introduction of the commodity futures market in India in 2005, it was expected that weather shocks should have had smooth transmission on the general price levels.
Economic wellbeing of farmers is going to become better if Agro-market advice facility is provided and thereby the standard of living of the farmers shall increase. This facility will actually help the farmers to know the prices for their products in and around them so that they can take their products to these places for their better market price.
Lot of research has been done in various aspects of agrometeorology and it is time to consolidate the results and translate them to actionable weather based agroadvisories. The Agrometeorologists not only make use of the weather prediction but also should be able to estimate the likelihood of unusual weather events and their potential impact on every single farmer’s fields/crops (quantitative analysis), which will be a great challenge.  This creates a great hope that insurance policies based on clever weather based analytics will one day also protect Indian farmers against the vagaries of the weather. The farming as well as the trading community is looking forward for the weather-based crop produce status coupled with market information which is need of the hour and more and more such value added information to the agroadvisories will definitely improve the farmers’ financial status leading to their ultimate prosperity. Satellite based remote sensing-aided weather based agro-advisories may be a promising way which would enable the farmers to take the most appropriate actions on real time basis, at a regional scale. The agrometeorologists need to gear up to meet the future demands of the country. 

For more information education or Business queries contact me at

Popular Posts