UNIT 3

LAND EVALUATION AND MANAGEMENT

 Land use/Land cover
• Land use - Areas are
classified based on the
purpose the land serves:
such as recreation, wildlife
habitat or agriculture.
• Land cover - Areas are
classified based on the
surface cover on the
ground whether
vegetation, urban
infrastructure, water, bare
soil or others.
Why Land Cover?
• Identification, delineation and mapping land cover is important for
monitoring studies, resource management, and planning activities.
• Identification of land cover establishes the baseline from which monitoring
activities (change detection) can be performed.
• Remote sensing methods can be employed to classify the types of land use in
a practical, economical and repetitive fashion, over large areas.

Why Land Use?

• Land use applications involve both baseline mapping and subsequent
monitoring, since timely information is required to know what current
quantity of land is in what type of use and to identify the land use changes
from year to year.
• This knowledge will help develop strategies to balance conservation,
conflicting uses, and developmental pressures.
Important considerations during LULC mapping
• Purpose- scientific studies, policy, planning or management purposes.
• Thematic content - needed for few cover types or for all cover types.
• Scale - locally, regional scales, or continental to global scales.
• Data- RS data limit type and accuracy of information that may be
extracted.
• Methodology – visual or digital or automatic or semi-automatic.

The purpose and thematic content help determine the classes that must be
differentiated in the land cover product, i.e. the mapping legend. The scale, together
with the legend, determines the remote sensing data source appropriate to the
mapping problem.
SOURCES OF LAND USE/COVER INFORMATION
1. CONVENTIONAL
• Revenue records compiled by the Directorate/Bureau of Economic
and Statistics (DES/BES) which is mainly tabular.
• Topographical maps from Survey of India represent very broad land
cover categories. These maps mainly provide topographical
information.
• Land use Atlas from NATMO are mainly small scale and are secondary
compilations.
• Soil Survey organizations (NBSS&LUP and AIS&LUS) generate soil and
land capability maps for specific project/ areas.
SOURCES OF LAND USE/COVER INFORMATION
2. REMOTE SENSING BASED
• LULC Maps at different scales from NRSC using satellite data / Aerial
Photos
*Ref: LAND USE LAND COVER
MAPPING Dr. T. RAVISANKAR Group
Director, LRUMG / RSAA National
Remote Sensing Centre (NRSC) / ISRO
Hyderabad – 500625 INDIA
Five factors in LULC Mapping
1. Objectives
2. Classification Scheme
3. Mapping Scale
4. Data Source
5. Analysis technique
CHARACTERISTICS OF CLASSIFICATION SYSTEM
• 85 % or greater interpretation accuracy.
• Repeatable results among interpreters and from one time of sensing to
another.
• Geographically extensible.
• Suitable for use with data from different seasons.
• Effective use of subcategories to permit use of data from ground surveys
and large scale imagery.
• Aggregation of categories must be possible.
• Comparison with future land use data should be possible.
• Multiple uses of land should be recognized i.e. segments having multiple
activities, each activity should be included
Classification adopted for land-utilization
statistics(DES)
Methodology
HYPERSPECTRAL DATA APPLICATION IN LULC
STUDY
• The accuracy of the thematic map produced from automatic digital
classification depends on many factors like classification algorithm,
parameters used in the classifier, size of the training samples etc.
• The space borne hyperspectral(HS) datasets have a medium to coarse spatial
resolution with high spectral information leading to many mixed pixels in the
image.
• Hence, finding too many pure pixels from the image for training the classifier
is a challenging task for HS datasets.
• The conventional multispectral image classifiers like Maximum Likelihood
require at least 10*n(n – number of bands) training pixels for classification.
• Hence, advanced classifiers like Spectral Angle Mapper, Support Vector
Machines, Artificial Neural Networks, Random Forests etc. were designed for
specifically classifying the Hsdatasets.
Hypothesis adopted in Microwave Remote
Sensing data
• A multi polarised SAR data with high spatial resolution provides better
land cover identification and mapping compared with single SAR
channels.
• Inclusion of derived features such as image texture provides improved
land cover information extraction compared to processing of original
multi-polarised bands or Landsat TM data only.
• The fusion of the high textured and high resolution SAR data with
high spectral but medium spatial resolution Landsat TM provides
better results compared to those obtained using either SAR or
Landsat TM alone.
SAR data in classification
SAR Classification
• Decomposing the scattering powers into single, double and double bounce
scattering components and feed the same as the inputs for the
classification algorithms (Machine and Deep Learning).
LULC classification
Unsupervised classification
• This type of classification is based on the software analysis of an image without
the user provided sample classes.
• This involves grouping of pixels with common characteristics. The computer uses
techniques to determine which pixels are related and groups them into classes.
• The user can specify which algorithm the software will use and the desired
number of output classes but otherwise does not aid in the classification
process.
• However, the user must have knowledge of the area being classified (such as
wetlands, developed areas, coniferous forests, etc.).
LULC classification
Supervised classification
• This is based on the idea that a user can select sample pixels in an image
that are representative of specific classes and then direct the image
processing software to use these training sites as references for the
classification of all other pixels in the image.
• Training sites (also known as testing sets or input classes) are selected
based on the knowledge of the user.
• The user also sets the bounds for how similar other pixels must be to group
them together. These bounds are often set based on the spectral
characteristics of the training area, plus or minus a certain increment
(often based on "brightness" or strength of reflection in specific spectral
bands). The user also designates the number of classes that the image is
classified into.
Supervised classification
LULC CHANGE MODELING
• Which factors (biophysical, socio-economic) or processes drive the
LULC changes and why?
• What is the spatio-temporal distribution of LULC changes?
• What will be the future LULC patterns?
• Ultimately to help decision makers in sustainable land-use planning
Types of LULC Models
• Empirical – Statistical models: (univariate/multivariate/logistic
regression)
• Stochastic models: (Transitional probability models; e.g. CA-Markov
Models)
• Optimization models (generally based on economic theory, e.g.
maximizing profit, minimising loss; Agent Based Models)
• Dynamic (process-based) models (simulate spatio-temporal patterns
of LULC through interaction of biophysical and human processes)
• Hybrid models
Confusion matrix
SVM
Waterlogged Omission Producers
Waterbodies Soil Vegetation vegetation Settlements CCP error Accuracy Recall
Waterbodies 22 0 0 0 2 24 0.083

Soil 1 25 1 0 0 27 0.074
Vegetation 0 0 30 1 0 31 0.032
Field Waterlogged
vegetation 0 0 6 19 0 25 0.24

Settlements 0 0 0 10 16 26 0.384
CCP 23 25 37 30 18 112
Commission
error 0.043 0 0.189 0.366 0.111
User Accuracy
Precision

CC- Correctly Classified Pixels

Confusion matrix
• Overall Accuracy = Correctly Classified/Total No of pixels
• Commission error= {(no. of diagonal cell of error matrix) / (Total no. of
pixels in the column)}
• User Accuracy= (1- Commission error) * 100
• Omission error= {(no. of diagonal cell of error matrix) / (Total no. of pixels
in the row)}
• Producers Accuracy= (1- Omission error) * 100
• Kappa coefficient = ((N×A)-B)/ N2 –B
N= 133 A= 112
B= (23*24) + (25*27) + (37*31) + (30*25) + (18*26)
Difference between Overall Accuracy and
Kappa coefficient
• kappa does not quantifies the level of agreement between two
datasets. It represents the level of agreement of two dataset
corrected by chance.
• The Kappa Coefficient can range from -1 t0 1.
Recall and Precision Measures
The results of two-dimensional classification schema were
validated with ground-truth observations using AUCPR measure
arrived from standard accuracy matrices, namely Recall, Precision,
and F1 score.
Method Evaluation metrics Waterbodies Wet soil Vegetation Waterlogged Settlements Overall F1 AUC
(resultant classes) Vegetation score (Precision-
Recall Graph)

2D Classification Omission (%) 8 7 3 24 38 0.84 0.74

plane
(SVM) Recall 0.92 0.93 0.97 0.76 0.62

Commission (%) 4 0 19 37 11

Precision 0.96 1 0.81 0.63 0.89

F1 score 0.94 0.96 0.88 0.69 0.73

Area (%) 3.5 34.2 30.7 2.1 1.5

Area Under the Curve (AUC) – Precision-Recall
Curve
• From the error (confusion) matrix, it was
found that the SVM classifier has shown
good performance for all categories with
an overall success rate of 84%.
• Though the overall accuracy indicated
substantial agreement with the ground
data, the classifier’s performance for
every individual predicated category
was studied using precision and recall
metrics which further offers an insight
into the fact that whether the classifier
reflects the actual characteristics of the
ground data population and overall
accuracy rate is balanced or unbalanced
between each mapped category.
Levels of Land use planning
• (i) National-level land-use planning is applied to planning at national
government level which deals with the country's land, water or other
resources as a whole. Small map scales are used. In large countries,
planning of the major administrative divisions has some of the
characteristics of national-level planning.
• (ii) District-level land-use planning is a level of land-use planning between
the national and local levels, typically but not necessarily that of the
administrative district. Intermediate map scales such as 1:10000 to
1:50.000 are used. Land-use development projects are often at the district
level.
• (iii) Local-level land-use planning is a level of land-use planning based on a
village or other local community. Large map scales are used, such as
1:10000. This level is also referred to as village level land-use planning.
Land Use Sustainability Assessment
• In order to be meaningful, planning for the resources should be sustainable
i.e. the land use option(s) selected for implementation should meet the
needs of the present without compromising the ability of the future
generations to meet their needs.
• Selected social and economic indicators of land use sustainability were
combined into a Multi-Criteria Evaluation (MCE) by a Decision Support
System to obtain a ranking of the alternatives based on their performance in
the evaluation criteria.
• DSS is a management information system that supports decision makers
such as planners, analysts, managers in the decision making process.
• They enable the decision maker to explore the problem environment by
provision of data and models appropriate to the decision and also generate
and evaluate alternative solutions.
Decision support and DSS
Why DSS in Agriculture?
DSS in Agriculture
Definition of Decision support System
A decision support system for land use planning

• Land use analysis through Remote Sensing (RS) and Geographic

Information System (GIS) is very useful in generating scientifically-
based statistical data for understanding ecosystem characterization
and ecological diversity.
• GIS-based commercial and open source software's aids in decisions
related to land use and land cover changes. This spatial decision
support system aids the policy makers and planners to visualize the
decision outcome.
General DSS
Systems
GIS Support in Decision-Making in Planning
 National Academy of Agricultural Research Management (NAARM) was
Level- 1 established by the Indian Council of Agricultural Research (ICAR) at Hyderabad,
Level- 1 Use: NAARM Geospatial Library
Level 2 use: Agro ecological sub regions, Climate
classes, Length of the growing period, Potential
evaporation, and Mean Temperature
Level 2 use: Soil properties:
Level 3 Use: GIS based DSS – Case studies
Process
Case Study 2
Process
Case Study 3: Real time water demand
estimation in Patna Canal System, Bihar
Process
Optimum land use planning for sustainable
agriculture
• Sustainability, refers to a development in land use planning that
meets the. needs of the present while conserving resources for future
generations. This requires a combination of production and
conservation: the production of the goods.
Precision agriculture (PA)
Precision agriculture (PA) is a farming management concept based on
observing, measuring and responding to inter- and intra-field variability
in crops. PA is also sometimes referred to as precision farming, satellite
agriculture, as-needed farming and site-specific crop management
(SSCM).
 UNIT 4

DAMAGE ASSESSMENT
Pests and Diseases and their impact and
Agriculture
What is crop disease?
• Generally, a plant gets diseased when it is continually disrupted by a
certain causal agent, resulting in a physiological process anomaly that
disrupts the normal structure of the plant’s function, and growth, among
other activities.
• Pathological conditions and symptoms result from the disruption of one or
more of a plant’s critical biochemical and physiological systems.
• The occurrence and prevalence of crop diseases vary seasonally, depending
on the prevalence of a pathogen, conditions of the environment, and the
crops and varieties grown. Some plant varieties are more prone to
outbreaks of plant diseases than others.
 Classification of Plant Diseases
• Crop disease has traditionally been classified into two types: abiotic (also
known as non-infectious) and biotic (infectious).
• Unfavorable environmental conditions frequently result in
noncommunicable diseases. Low or high temperature and excess or lack of
moisture are a few examples.
Abiotic Agents
Non-living environmental conditions or poor
1. Extreme temperatures. farm management are examples of abiotic,
2. Wind. or noninfectious, disease agents. They are
3. Drought or flood. not passed on to other plants.
4. Moisture.
5. Frequent and heavy rain.
6. Soil compaction.
7. Excess or deficiency of nutrients.
8. Improper water management.
9. Chemical injury caused by pesticides or salts.
Classification of Plant Diseases
Biotic disease agents, also known as infectious
disease agents, are living organism pathogens
that can spread from one host to another and
transmit disease.
1. Fungi: The most common agricultural
problem is pathogenic fungi. According to
studies, this plant disease type destroys
roughly one-third of all food crops each year.
In this regard, the problem is severe from
both a humanitarian and an economic
standpoint. These infections, like bacterial
crop diseases, primarily affect plants through
wounds, water pores, and stomata.
Furthermore, fungal spores are frequently
carried by wind gusts.
Classification of Plant Diseases
2. Viruses: Viroids and viruses are the most minor but critical plant enemies
(subviral contagious agents). It is close to impossible to save a plant
following an infection of the plant. In most situations, the infection
spreads by contact between healthy and diseased plants. Viruses can
also spread by vegetative reproduction in the form of seeds, pollen, and
insects. However, viruses are most commonly transmitted through the
soil.
3. Bacteria: Bacterial crop diseases, usually caused by bacteria are among
the most common infections in agriculture. In this regard, prevention and
control of this type of disease are difficult. The causal agent must enter
the culture’s tissue in order to infect it. It primarily occurs as a result of
damaged areas, such as those caused by agricultural tools, insects (such
as fleas), or simply unfavorable weather conditions (like dust, heavy rain,
and wind). Bacteria, on the other hand, can infect plants through natural
holes or glands (for example, which secrete nectar).
Classification of Plant Diseases
Classification of Plant Diseases

4. Nematodes: Nematodes are plant-parasitic roundworms that

cannot be seen without specialized equipment. Because they live in
the soil, they primarily affect roots, bulbs, and tubers. More than
4100 dangerous nematode species have been identified.
5. Parasitic plants: feed on crops and get their chlorophyll from the
host plant because they lack it. Dwarf mistletoe, for example, grows
on other plants and obtains nutrients from them.
Algae; Theoretically, they do not cause significant damage; however,
under certain conditions, they may be problematic.
 Remote Sensing for Pests and Diseases
• At the field scale, remote sensing monitoring of pests and diseases mainly
relies on manned or unmanned aerial systems, such as unmanned-owned
systems, or high-resolution satellite remote sensing systems to acquire field-
scale remote sensing images.
Remote Sensing for Pests and Diseases
• Reduction of biomass and decrease of LAI. This type of destruction usually
occurs in some pest attacks. Pests (e.g., armyworm in maize) can eat plant
parts (e.g., leaf, stalk), and thus cause a significant loss of leaf area and
biomass. However, given that this destruction lacks spectral specificity, its
monitoring always encounters the issue of high uncertainty.
• Lesions or pustules due to infection. Lesions or pustules, which are sori or
necrotic tissues caused by diseases and pests, are the most common
symptoms. The lesions and pustules tend to vary among diseases and pests
in their color and shape. It should be noted that the in-canopy distribution
and abundance of these lesions and pustules (e.g., uniformly distributed
within the canopy or localized at the bottom) can have a great influence on
their detectability.
Remote Sensing for Pests and Diseases
• Destruction of pigment systems. In many cases, the disease infection
and pest attack can cause destruction of chloroplast or
other organelles, leading to variations in pigments contents (e.g.,
chlorophyll (Chl), carotenoid (Car) and anthocyanin). Hyperspectral RS
observations are usually required to detect this type of response.
• Wilting. The loss of rigidity due to dehydration is not a very common
symptom for plant diseases and pests and it can be easily confused
with drought stress. The piercing and sucking behavior of some pests
(e.g., beetles or aphid) will cause wilting of the plants. Besides, in
some severely infected situations, the damaged vascular system will
block the water flow in plants and thus cause dehydration to the
entire plants.
Remote Sensing for Pests and Diseases at
temporal sequence
• The remotely sensed monitoring should be able to capture the accumulation of
symptoms.
• Given that the disease infection or pest attack usually behaves as a temporal
process, different forms of symptoms may superimpose or interact with each other
and will show up with varied severities over different stages.
• For example, the yellow rust in winter wheat will initially cause lesions on leaves and
decrease of chlorophyll content at the beginning stage. Without proper control, the
damaged photosynthetic system will lead to decrease of biomass and will impede
the water metabolism in plant.
• As a result, the severely infected plants will show a wilting symptom at a later stage.
• Such a temporal pattern of the symptom changes will thus complicate the detection
and monitoring of some diseases and pests. Therefore, it is necessary to identify
specific RS features at different development stages of the disease or pest attack.
Crop damage by floods
• Flooding often causes significant crop loss over large agricultural
areas.

Drought damage assessment Hailstorms damage assessment Floods damage assessment Wildfires damage assessment
Crop loss assessment Crop loss assessment Flood extent & duration Fire extent & severity

MODIS Sentinel-2 Sentinel-2, Sentinel-1 Sentinel-2

composite NDVI data sets Change Detection Change Detection Change Detection
NDVI Anomaly pre- post-damage imagery pre- post-damage imagery pre- post-damage imagery

Drought Hailstorms Floods Wildfires

 Types of floods and associated losses
1. Flash floods
Flash floods are floods of short duration with a
relatively high peak discharge.
Flash floods are short term events that occur
6-12 hours or less after the associated rainfall.
Mountainous or hilly areas with steep terrain
gradients and high surface runoff are
particularly susceptible.
Flash floods can also be triggered by the
failure of hydraulic infrastructure such as
dams or levees, outburst of glacier lakes, ice-
jams or logjams.
Flash flood causes severe damage to crops, particularly
it occurs just before harvesting
Types of floods and associated losses
Types of floods and associated losses
2. Riverine flooding
• Riverine Flooding is when streams and rivers
exceed the capacity of their natural or
constructed channels to accommodate
water flow and water overflows the banks,
spilling out into adjacent low-lying, dry land.
• It can occur seasonally with spring rains or
melting snow. River waters overflow the
banks and fill the adjoining low, flat
floodplains. This type of seasonal flooding is
often gradual and lasts for weeks or months.
• Riverine flooding can affect croplands and
livestock farmers.
Types of floods and associated losses
3. Coastal flooding
• Coastal flooding occurs when dry and low-lying
land is submerged (flooded) by seawater. The
range of a coastal flooding is a result of the
elevation of floodwater that penetrates the
inland which is controlled by the topography of
the coastal land exposed to flooding.
• Coastal flooding of farmland can lead
to immediate, as well as long term, crop losses.
Even after flood waters recede, salt deposition
from sea water establishes a legacy of soil
salinity
Types of floods and associated losses
4. Groundwater flooding
Groundwater flooding occurs as a result of water either rising up from
underlying rocks or flowing from abnormal springs. It is often characterized
by long-lasting flooding that can cause prolonged disruption and relatively
high levels of damage to the crops.
Groundwater flooding tends to occur after sustained periods of high rainfall.
Higher rainfall means more water seeps into the ground, causing the water
table to rise above normal levels.
Groundwater flooding is most likely to occur in low-lying areas underlain by
permeable rocks (aquifers). These may be extensive, regional aquifers, such
as chalk or sandstone, or localised sands or river gravel at the bottom of a
valley underlain by less permeable rocks.
Remote Sensing in assessing flood damage to
the crops
• Accurate crop-specific damage assessment immediately after flood
events is crucial for grain pricing, food policy, and agricultural trade.
 NDVI (Normalized Difference
Optical Remote Sensing Vegetation Index), EVI (Enhanced
Vegetation Index), RVI (Ratio
Vegetation Index), SAVI (Soil Adjusted
Vegetation Index), OSAVI (Optimized
Soil Adjusted Vegetation Index), VCI
(Vegetation Condition Index), MVCI
(Mean Vegetation Condition Index),
RMVCI (Ratio to Median Vegetation
Condition Index), and LAI (Leaf Area
Index).

Radar Vegetation Index (RVI), Dual

Polarization SAR Vegetation Index
(DPSVI), Radar Forest Degradation
Index (RFDI), Microwave Polarization
Difference Temperatures (MPDT),
Microwave Polarization Difference
Index (MPDI), Microwave Emissivity
Microwave Remote Sensing Difference Vegetation Index (EDVI),
and Microwave Vegetation Index
(MVI).
 • Both passive and active sensing are useful
Microwave Remote Sensing for this purpose.
• Passive- most suitable for analyses over
larger-scale water features
• Active (SAR)- Local-scale application.

Modified Normalized Difference Water Index (MNDWI)

Optical Remote Sensing (When the cloud
cover is less)
Flood – Remote Sensing Simulation
• Earth observation satellites are also used
extensively in the phases of
preparedness/warning and
response/monitoring.
• The use of optical sensors for flood mapping
is seriously limited by the extensive cloud
cover.
• Synthetic Aperture Radar (SAR) from ERS
and RADARSAT have been proven very
useful for mapping flood inundation areas,
due to their all weather capability.
• In India, ERS -SAR has been used successfully
in flood monitoring since 1993, and Radarsat
since 1998.
• Colour composites are generated using SAR
data during floods and pre-flood SAR
Sentinel-1 – considered to be effective
images.
Why SAR is a performing tool for water bodies and
flood mapping ? • Near all weather capability
• Day & night capabilities.
• Relative large swath
• Relative good revisit
• On SAR data water surfaces
have low values of Backscattering
values
Water backscattering in function of surface
roughness
 Water
backscattering in
function of water
surface roughness
Drought
• Drought is considered to the most complex but least understood of all
natural hazards, affecting more people than any other hazard.
• Drought, lack or insufficiency of rain for an extended period that causes a
considerable hydrologic (water) imbalance and, consequently, water
shortages, crop damage, stream flow reduction, and depletion
of groundwater and soil moisture. It occurs
when evaporation and transpiration (the movement of water in the soil
through plants into the air) exceed precipitation for a considerable period.
Drought is the most serious physical hazard to agriculture in nearly every
part of the world.
• Drought is caused by not only lack of precipitation and high temperatures
but by overuse and overpopulation.
Types of Drought
• Meteorological drought is specific to different regions, depending on the amount
of yearly precipitation that's average for that area. For example, the southwest
portion of the United States averages less than 3 inches (7.6 centimeters) of
precipitation per year, while the Northwest gets more than 150 inches (381 cm)
per year, according to the U.S. Department of Interior. A decrease in precipitation
compared to the historical average for that area would qualify as a
meteorological drought.
• Agricultural drought accounts for the water needs of crops during different
growing stages. For instance, not enough moisture at planting time may hinder
germination, leading to low plant populations and a reduction in yield.
• Hydrological drought refers to persistently low water volumes in streams, rivers
and reservoirs. Human activities, such as drawdown of reservoirs, can worsen
hydrological droughts. Hydrological drought is often linked with meteorological
droughts.
• Socioeconomic drought occurs when the demand for water exceeds the supply.
Examples of this kind of drought include too much irrigation or when low river
flow forces hydroelectric power plant operators to reduce energy production.
Quiz
Remote sensing capabilities and contributions for
drought management
Remote Sensing for Drought Assessment
• Optical (AVHRR, Landsat, SPOT, and MODIS)
• Thermal (Landsat, AVHRR, MODIS)
• Microwave remote sensing (Sentinel-1, ALOS PALSAR, RADARSAT,
NISAR)

The components of the agro-hydrologic system relevant to

drought monitoring that can be estimated or modeled from
satellite remote sensing data are: (1) rainfall; (2) vegetation
condition; (3) soil moisture; (4) groundwater; and (5)
evapotranspiration (ET).
Remote Sensing for Drought Assessment
• Satellite IR and VIS images of cloud top temperature can be converted into a
precipitation rate using empirical statistical relationships.
• Passive microwave (MW) sensors offer a more physically based approach to
instantaneous precipitation estimation.
• Studies suggest that combining both satellite MW and IR information leads to
better precipitation estimates, especially of diurnal patterns, by combining
the strengths of both sensors.
• Currently, several satellite precipitation data sets are available including the
Climate Predicting Center (CPC) Morphing Technique (CMORPH), Tropical
Rainfall Measuring Mission (TRMM), Multi-satellite Precipitation Analysis
(TMPA), Precipitation Estimation from Remotely Sensed Information using
Artificial Neural Networks (PERSIANN), POWER (NASA) and the Global
Precipitation Climatology Project (GPCP) for a comprehensive review of
precipitation algorithms.
• These data sets have been extensively intercompared and validated against
ground-based observations
Remote Sensing for Drought Assessment- Soil
Moisture
• Soil moisture is a fundamental component of
the water cycle and plays a key role in drought
monitoring and prediction, especially in water-
limited ecosystems.
• Soil moisture is often used as a measure of
agricultural drought since it affects plant
growth and productivity.
• A number of soil moisture-based indices have
been developed and used for drought
monitoring, including the Standardized Soil
Moisture Index (SSI) and the soil moisture
percentile.
• The soil moisture input to these drought
indicators can be obtained from land surface
model simulations or from satellite estimates.
Remote Sensing for Drought Assessment- Soil
Moisture
• Optical remote sensing- The normalized difference vegetation index (NDVI),
standardized soil moisture index (SSMI), soil moisture percentile (SMP),
normalized soil moisture (NSM), standardized precipitation index (SPI), vegetation
health index (VHI), and soil moisture deficit index (SMDI) are the most used
indices by researchers for drought detection and its assessment.
• Microwave Remote Sensing- The principal of MW-based soil moisture retrieval
relies on the relationship between soil permittivity and liquid water content.
• There are empirical relationships that link passive MW brightness temperature
and active MW backscattering to volumetric water content of soil. MW soil
moisture observations typically represent the top 2–5 cm of soil depth.
• For root-zone soil surface moisture estimates, MW soil moisture observations can
be coupled to an appropriate land surface model.
SMAP (Soil Moisture Active/Passive)

• Launched in January 2015, the Soil Moisture

Active Passive (SMAP) mission consists of an
imaging radar satellite whose operations are
led by the National Aeronautical and Space
Administration (NASA), with assistance from
the Canadian Space Agency (CSA).
• SMAP uses radio waves to map soil moisture
and determine the freeze or thaw state of
the same area.
SMAP (Soil Moisture Active/Passive)

Radar: L-band Radiometer: L-band

Frequency: 1.2 to 1.3 GHz
Polarizations: VV, HH, HV (not fully Frequency: 1.41 GHz
polarimetric). Polarizations: H, V, 3rd and 4th
Relative accuracy (3 km grid): 1 dB Stokes.
(HH and VV), 1.5 dB (HV).
Relative accuracy (30 km grid): 1.3 K
Data acquisition:
• High-resolution (SAR) data acquired Data collection:
over land •High-rate (sub-band) data acquired
• Low-resolution data acquired over land
globally •Low-rate data acquired globally
Land degradation due to waterlogging and salinity
• Waterlogging refers to the accumulation
of excess water in soil, which negatively
impacts plant growth and soil
productivity.
• Soils degraded due to excess salt
accumulation are termed as salt
affected soils which include saline, sodic
and saline-sodic soils.

It occurs when the soil is unable to absorb

or drain water adequately. Waterlogging
leads to oxygen deprivation in the root
zone, causing root decay and reduced
nutrient uptake.
Remote Sensing for Mapping Waterlogging/Soil
Salinity
• The essence of remote sensing is the measuring and recording of the
electromagnetic radiation emitted or reflected by the earth's surface.
• For waterlogging and soil salinity investigation, this may be useful
where ponded water, high water table areas, saline water, salty soil,
and salt-affected vegetation give contrasting reflectance with other
landscape features so that they can be unambiguously distinguished.
• Excess soil moisture can cause a change in soil colour and a change in
soil reflectance properties, which can be easily detected by remote
sensing
Remote Sensing for Mapping Waterlogging/Soil
Salinity
• For the visible part of the spectrum, the soil reflectance of salt cover areas
was found to be prominent.
• Bands in the middle infrared gave information of moisture content which
was often associated with salt content differences and some information
on type of salts.
• The crusted saline soil surfaces are generally smoother than non-saline
surfaces and causes higher reflectance values in the visible and near
infrared bands.
• The lack of vegetation or scattered vegetation and highly salt-affected salt
surface makes it possible to directly detect salt on the surface.
• Many researches described salinity detection through use of vegetation on
the basis of the fact that reflectance from single leaf depends on their
chemical composition (salt) and morphology.
Remote Sensing for Mapping Waterlogging/Soil
Salinity
• Brightness index proved to be a more useful spectral parameter if
surface soil properties are to be extracted from satellite data, but
ratio of the values in red and infrared band seems to be a better
technique to employ when subsurface soil properties are of interest.
• Further, strongly saline-sodic soils were found to have higher spectral
response as compared to moderately saline-sodic soils. The
vegetation cover modifies the overall spectral response pattern of
salt-affected soils especially in the green and red spectral bands.
• Digital classification techniques help in improving the identification
and mapping of salt-affected soils or crops.
• Microwave remote sensing has also shown encouraging results for its
relationship with dielectric constant and electrical conductivity.
Case Study- Identification of saline landscapes from an
integrated SVM approach from a novel 3-D classification
schema using Sentinel-1 dual-polarized SAR data
• The saline affected areas, the broad level of LULC
categories (vegetation and soil) could be further
classified into detailed classes such as i) healthy,
waterlogged, and salt-tolerant vegetation and ii)
non-saline and saline soil, according to the
magnitude of salinity level in the region.
• It was reported to be relatively challenging to
perform LULC classification for such highly dynamic
landscapes compared to the homogenous
landscapes using traditional remote sensing
methods.
As a geologically active semi-arid zone with uncertain weather conditions, this area is often
categorized by various heterogeneous landforms.
This saline landscape is often found with flooded vegetation and waterlogging soil after every
rainfall and irrigation activity.
Crucial Methods
Results
• The proposed two-dimensional classification plane with SVM classification strategy
resulted in five LULC categories such as water bodies, wet soils (ranges from semi-
saturated to waterlogged condition), healthy vegetation, waterlogged vegetation,
and settlements whose spatial coverage was estimated to be 3.5% (222 Km2), 34.2%
(2142 Km2), 30.7% (1922 Km2), 2.1 % (133 Km2), and 1.5% (96 Km2) of the total
geographical area (6265 Km2) respectively.
• The proposed integrated SVM method on the three-dimensional classification
schema has resulted in twelve classes, and the area occupied by each category is i)
1.2 Km2 (Non-saline water body), ii) 61 Km2 (Non-saline soil), iii) 177 Km2 (Healthy
Vegetation), iv) 31 Km2 (Waterlogged Vegetation under non-saline condition), v) 8
Km2 (Slightly saline water body), vi) 234 Km2 (Slightly saline soil), vii) 261 Km2 (Salt-
tolerant plants under slightly saline conditions), viii) 50 Km2 (Waterlogged vegetation
under slightly saline conditions), 213 Km2 (Moderately saline water body), 1849 Km2
(Moderately saline soil), 1483 Km2 (Salt-tolerant plants under moderately saline
conditions), and 49 Km2 (Waterlogged Vegetation under moderately saline
conditions)
TYPES OF STRESS IN PLANTS
• TEMPERATURE STRESS.
• STRESS DUE TO A LACK OR EXCESS OF
LIGHT.
• STRESS PRODUCED BY INFECTIONS
FROM PATHOGENS, PESTS AND
VIRUSES.
• STRESS DUE TO ABSENCE OR EXCESS
OF NUTRIENTS.
• SALINITY STRESS.
• WATER STRESS.
• MECHANICAL STRESS.
 TEMPERATURE STRESS

• Most plants are sensitive to

temperature stress and suffer when
temperatures rise above their
defined thresholds. High
temperatures, for example, cause
physiological and metabolic
damage. With low temperatures,
plants stop growing or freeze and
eventually die.
STRESS DUE TO A LACK OR EXCESS OF LIGHT

• Light is the source of energy for

photosynthesis, which is directly
related to the production of
carbohydrates and oxygen. Each
plant has its own needs for light and
sun, and the lack or excess of them
also influences growth and
development.
STRESS PRODUCED BY INFECTIONS FROM PATHOGENS,
PESTS AND VIRUSES.

• Plants may suffer attacks from other

harmful organisms that limit or affect
their development, and furthermore,
the application of synthetic treatments
to combat them may also be stressful
depending on the conditions in which
they are undertaken.
STRESS DUE TO ABSENCE OR EXCESS OF NUTRIENTS.

• Of course, the absence of nutrients affects the growth of plants and

their fruits, but an excess also produces negative effects, as is the
case with nitrogen and phosphorous.
SALINITY STRESS
• This type of stress may be
produced by the excess of salt
in the soil or in the irrigation
water.
• Poor irrigation control may also
have an impact in these salinity
problems.
WATER STRESS
• Water stress is one of the most frequent
stresses and one of the primary causes of
plant death.
• Without water, there is an imbalance
between transpiration and water
absorption, and plants cannot transport
nutrients from the roots to the leaves,
affecting crop growth and production.
• On the other hand, too much water in the
medium may provoke a series of shocks
that limit the growth of the plant because
of the lack of oxygen.
 CROP MECHANICAL STRESS.
• This is stress caused by damage from
agricultural machinery and other tools
used in agricultural processes, such as
pruning, transplanting, or thinning. This
also includes stresses provoked by
adverse weather conditions like hail,
wind, etc.
Fertilizer Burn

• Fertilizers are essential for crop

development, yet their misuse may
result in fertilizer burns and lead to
significant yield losses.
• This is why farmers should know
how chemical burn from fertilizer
looks like and what causes it, trying
to avoid misapplications.
• By understanding how to treat
fertilizer burns, you can save your
crops if the problem is timely
spotted.
What Is Fertilizer Burn And What Are Its Causes?
Fertilizer plant burn is an abnormal vegetation state manifesting yellow or
brownish curled and shedding leaves. The cause of fertilizer burn on crops is
excessive fertilization at any crop phenological stage. Besides, the risk of
plant damage is high when fertilization is performed:
• close to seeds while planting (in-furrow);
• to dry soil;
• to wet leaves;
• at direct sunlight;
• in hot weather;
• to crops under heat stress;
• in incompatible combinations.

Fertilizer burn on crops is an ultimate effect of nutrient toxicity, commonly due to excessive
nitrogen. However, nitrogen fertilizer salt burn may be confused with nitrogen deficiency, an
opposite vegetation state with similar signs.
Fertilizer Burn
How To Spot Fertilizer Burn?

• Excessive fertilization causes the following changes in crops, affecting

their leaves and roots:
• impaired growth;
• brown leaf margins and tips;
• leaf curling and scorching;
• defoliation;
• root discoloration (from brown to black);
• root rots;
• salt crust on the soil due to evaporation of fertilizing solutions.
What Makes The Situation Worse?
• Soil salinization. Excessive soil salinity enhances osmotic pressure, which reverses water
flows between the soil and the crop. Furthermore, due to salinity, crops cannot absorb
water even if it is present in the soil and get dehydrated.

• Insufficient irrigation. Lacking soil moisture cannot fully support plant water needs. In
other words, crops cannot replenish the lost moisture after evaporation and transpiration
while soluble salts sit in the foliage, causing fertilizer burn on a leaf. The risk decreases in
cool cloudy weather when the moisture loss is not as fast as on hot dry days.

• Droughts. Hot dry weather increases the salt content in the soil, which can burn crop
roots. Since roots are underground, their damage is not visible directly, but it can be
understood through the symptoms of leaf fertilizer burn.

• Plant sensitivity. The impact is high if plant tolerance to salts is low. Besides, some crops
are vulnerable to in-furrow fertilization, which refers to soybeans, sorghum, corn, and
some others.
How To Treat Fertilizer Burn?

• Fix fertilizer burn by flushing excessive nutrients out with water for about a
week. Excessive nutrients will leach below the root zone, making the soil
environment safer for crops. Remember to avoid waterlogging as it is also
critical to yields. Thus, this technique works only in the case of adequate
soil infiltration.
• Help plants recover from fertilizer burn faster by removing damaged and
dead leaves to save the plant’s energy. This way, the affected plant will be
able to produce new foliage quicker.
• Check the results in a couple of weeks after treating burn spots from
fertilizer. If rehabilitation has been successful, crops will get green again.
• Remove the old plants and resow the field if the attempts to fix burn spots
from fertilizer have been in vain. However, mind to flush out excessive
nutrients first.
Identification of Crop Stress
Crop Stress assessment from various remote
sensing techniques
Agricultural Insurance in India
Comprehensive Crop Insurance Scheme (CCIS) – 1985
• The Comprehensive Crop Insurance Scheme (CCIS) was introduced with
effect from 1st April 1985 by the Government of India with the active
participation of State Governments.
• The Scheme was optional for the State Governments.
• The CCIS was implemented on Homogeneous Area approach and was
linked to short-term crop credit, that is, all crop loans given for notified
crops in notified areas were compulsorily covered under the CCIS.
The salient features of the Scheme
• It covered farmers availing crop loans from Financial Institutions for growing
food crops & oilseeds on compulsory basis. The coverage was restricted to
100% of crop loan subject to a maximum of ` 10,000/- per farmer.
• The Premium rates were 2% for Cereals and Millets and 1% for Pulses and Oil
seeds. 50% of the Premium payable by Small & Marginal farmers was
subsidized by Central and State Governments in equal proportion.
• Premium & Claims were shared by Central & State Government in 2:1 ratio.
• The Scheme was optional to State Governments.
• The maximum Sum Insured was 100% of the crop loan, which was later
increased to 150%.
• CCIS was a multi-agency scheme, involving Government of India,
Departments of State Governments, Banking Institutions and GIC.
The salient features of the Scheme
• 15 States and 2 UTs had participated in the CCIS during its tenure from
Kharif 1985 to Kharif 1999.
• These were Andhra Pradesh, Assam, Bihar, Goa, Gujarat, Himachal
Pradesh, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Meghalaya,
Orissa, Tamil Nadu, Tripura, West Bengal, Andaman & Nicobar Islands and
Pondicherry.
• The States of Rajasthan, Uttar Pradesh, Jammu & Kashmir, Manipur and
Delhi had initially joined the Scheme but opted out after few year.
• CCIS was eventually discontinued after Kharif 1999, to be replaced by the
improved and expanded “National Agriculture Insurance Scheme” (NAIS),
which is being continued till date.
Experimental Crop Insurance Scheme (ECIS) -
1997
• While the CCIS was being implemented, attempts were made to
modify the existing CCIS from time to time as demanded by the
States. During the Rabi 1997-98 season, a new scheme, viz.
Experimental Crop Insurance Scheme (ECIS) was introduced in 14
districts of 5 States.
• The Scheme was similar to CCIS, except that it was meant only for all
small / marginal farmers with 100% subsidy on Premium.
• The Premium subsidy and Claims were shared by the Central and
respective State Governments in the ratio of 4 : 1.
• The Scheme was discontinued after one season due to its many
administrative and financial difficulties.
Farm Income Insurance Scheme (FIIS) - 2003
• NAIS (National Agriculture Insurance Scheme) protects the farmers only
against the yield fluctuations. The price fluctuations are outside
the purview of this scheme.
• Farmers' income is a cumulative function of yield and market prices. In
other words, a bumper harvest tends to bring down the market prices of
grains and vice versa.
• Therefore, despite normal production, farmers often fail to maintain their
income level due to fluctuations in market prices. To take care of variability
in both the yield and market price, the government introduced a pilot
project, viz. Farm Income Insurance Scheme (FIIS) during Rabi 2003-04
season.
• The objective of the scheme was to protect not only the income of the
farmer, but also to reduce the government expenditure on procurement at
Minimum Support Price (MSP).
Farm Income Insurance Scheme (FIIS) - 2003
• FIIS was implemented on the basis of 'homogeneous area' approach
in respect of rice and wheat crops only. The scheme was compulsory
for loanee farmers and voluntary for non-loanee farme` The premium
rates were actuarial, determined for each State at the District level, to
be subsidised by the Govt. of India.
• Claims would arise if the actual income (current yield X current
market price) was lower than the guaranteed income (7 years'
average yield X level of indemnity [80% or 90%] X MSP).
NATIONAL AGRICULTURAL INSURANCE
SCHEME (NAIS)
OBJECTIVES:
The objectives of the NAIS are as under:-
1. To provide insurance coverage and financial support to the farmers
in the event of failure of any of the notified crop as a result of
natural calamities, pests & diseases.
2. To encourage the farmers to adopt progressive farming practices,
high value in-puts and higher technology in Agriculture.
3. To help stabilise farm incomes, particularly in disaster years.
SALIENT FEATURES OF THE SCHEME
• CROPS COVERED:
• Food crops (Cereals, Millets & Pulses) b. Oilseeds c. Sugarcane, Cotton
& Potato (Annual Commercial / annual Horticultural crops)
• STATES AND AREAS TO BE COVERED:
• The Scheme extends to all States and Union Territories. The States /
UTs opting for the Scheme would be required to take up all the crops
identified for coverage in a given year.