Upload
others
View
10
Download
0
Embed Size (px)
Citation preview
2
Environmental Geography
Spatial distribution of human-environmental interactions
Land Change Science(Land Use Land Cover Change)
Environmental/Societal Effects of Climate Change/Hazards
Biodiversity Conservation / Natural Resource Management
Overall Research Focus
Pench, Maharashtra
Pench, Madhya Pradesh
National Park: 1975
Tiger Reserve: 1999
Area: 257 km2
NPench – the Forest of Jungle Book
4
Specific Research Question
Are parks working as an ecologically sustainable management regime?
Has Pench Tiger Reserve, Maharashtra been effective to conserve forest, promote regeneration, and maintain connectivity with the surroundings?
5
How can we evaluate
conservation interventions?
• Mixed methods used: –Field work
–Remote sensing analysis• Land cover classification
• Change detection analysis
–Landscape fragmentation analysis
–Land cover modeling• Cellular-automata Markov
7
Land Covers in Park and Buffer –Actual vs. Predicted
Χ2 test at 95% confidence level suggests statistically significant difference between actual and predicted land covers
40
50
60
70
80
90
1970 1980 1990 2000 2010
Park_forest - actual
Park_forest - predicted
Buffer_forest - actual
Buffer_forest - predicted
(a)
0
10
20
30
40
50
60
1970 1980 1990 2000 2010
Park_bare soil - actual
Park_bare soil - predicted
Buffer_bare soil - actual
Buffer_bare soil - predicted
(b)
Are
a
(%)
Are
a (
%)
Year
8
Main Findings
• The park suffered deforestation until 1989, despite its protected status, due to regular forest management activities conducted by the state government.
• The Revised Forest Policy in 1988 has been effective for forest conservation within the park, e.g. in the simulated scenario the park could have suffered 20%-26% more deforestation.
• State- and World bank-funded forestry projects have been effective to establish a reforesting trend in the buffers. While the buffer is mainly managed for commercial logging, increasing amount of stable forest is encouraging.
9[Source: Mondal & Southworth, 2010a, 2010b; Mondal, 2011]
Private forest, housing pressure, and
alternate scenarios
10
• Examining changing spatial distribution of harvesting activities on private forests across the urban–rural gradients.
• Comparing projected changes in the amount of impervious surface within private forests under different scenarios between 2010 and 2060.
• Quantifying overall pressure on private forests in the US.
[Source: Mondal et al. 2013]
Data Sources:Forest Inventory and Analysis (FIA) dataNational Woodland Owner Survey (NWOS) dataNational Land Cover Database (NLCD)Environment Protection Agency (EPA) housing density data
11
Environmental Geography
Spatial distribution of human-environmental interactions
Land Change Science(Land Use Land Cover Change)
Environmental/Societal Effects of Climate Change/Hazards
Biodiversity Conservation / Natural Resource Management
Overall Research Focus
12
Smallholder Agricultural System
Spatial distribution of human-environmental interactions(GIS and big data)
• Land Change Science (Land Use/Land Cover Change)• Environmental/Societal Effects of Climate Change/Extreme Events• Sustainability
Current Research Focus
Weather variability
Socio-economic
Well-being (food & nutrition
security)
Adaptation strategies
Samberg et al., 2016
Mean agricultural area in three global regions.
Research Context
>380 million farming households70% of food calories produced>50% staple crops worldwide
Samberg, Gerber, Ramankutty, Herrero & West, 2016. Subnational distribution of average farm size and smallholder contributions to global food production. Environmental Research Letters 11: 12401013
Current Research Context
14
• 78% of Indian farmers are smallholders
• 263 million (10% of global total) depend on agriculture
• 210 million are hungryCNN: April, 2016
H. Times: September, 2017 Photo: P. Mondal
IND
IA
***2017 US population: 326 million***
15
Research Context• 37% of agricultural land
is irrigated
• 60% of irrigation is provided by groundwater
• 29% less winter cropped area in regions with low groundwater availability
Map produced using India Water Tool
16
Rapidly changing weather patterns is one of the biggest challenges facing smallholder farmers
Modified after Google Image
17
Questions
• How does weather variability affect smallholder agricultural systems?
• What are the potential adaptation strategies under future scenarios?
• Can these strategies lead us to achieve food and nutrition security?
Ph
oto
cou
rtesy: Go
ogle Im
age
18
Questions
• How does weather variability affect smallholder agricultural systems?
Location-specific component
Crop-specific component
Socio-economic component
19
Satellite-derived Crop Phenology
Climate Parameters Crop Types Irrigation Demography
Methods: PIXEL to PEOPLE
20
Methods: Satellite data
Mondal, Jain, DeFries, Galford & Small, 2015. Sensitivity of crop cover to climate variability: Insights from two Indian agro-ecoregions. Journal of Environmental Management 148: 21-30
Enhanced Vegetation Index (EVI)
Source: MODIS
Spatial resolution: 250m
Temporal resolution: 16 days
(a) Single crop
(b) Double crop
21
Methods: Satellite data
Double crop
Single crop Mixed crop
MODIS pixel
250 m
Landsat pixel
30 m
Ph
oto
: P. M
on
da
l
22
Methods: Satellite data
Precipitation
• Monsoon start date• Monsoon end date• Seasonal total• Season length• Monsoon dry days• Days with low rain• Days with heavy rain
Mondal, Jain, Singh, Galford, & DeFries. Relative importance of climatic and non-climatic factors in Indian winter crop. In prep.
Total monsoon rainfall for 2015
23
Methods: Satellite data
Temperature
• Monthly average of daily Tmax
• Monthly average of daily Tmin
• Daily temperature range (DTR)
Monthly average of daily Tmax for August 2015
Mondal, Jain, Singh, Galford, & DeFries. Relative importance of climatic and non-climatic factors in Indian winter crop. In prep.
24
Methods: Space time cube
Image courtesy: ArcGIS Pro
Time: 16 years (2001 - 2016)
Space:
• 3,660 pixels (NOAA CPC)• 14,157 pixels (TRMM)• 355,116 pixels (CHIRPS)• 1,787,433 pixels (MODIS)
Variables:
• Response: winter cropped area
• Predictors: precip., temp.
25
Methods: Census data
Time: 2 years (2001, 2011)
Space:
• 554 districts
Variables:
• Response: winter cropped area
• Predictors: density of cultivators and agricultural laborers
26Mondal, Jain, DeFries, Galford & Small, 2015. Sensitivity of crop cover to climate variability: Insights from two Indian agro-ecoregions. Journal of Environmental Management 148: 21-30
Gujarat, Western India
Madhya Pradesh,Central India
Location-specific Vulnerability
27
• Sensitivity of crop productivity to climate variability is location specific – mostly due to different cropping practices and irrigation access
• Temperature is critically important for winter crops
• Sensitivity of crop productivity to precipitation depends on irrigation source
Location-specific Vulnerability
28
Crop-specific Vulnerability
Mondal, Jain, Robertson, Galford, Small & DeFries, 2014. Winter crop sensitivity to inter-annual climate variability in central India. Climatic Change 126: 61-76
PULSES (chickpea, pigeon pea,
lentil, moong)
WHEAT
Photo: P. Mondal
Irrigated field wheat Non-irrigated field pulses
29
Crop-specific Vulnerability
• Ground-truth points collected in field for wheat and pulse fields
• Other data collected in field include access to irrigation and type of irrigation
30
Crop-specific Vulnerability
PulsesWheat
Mondal, Jain, Robertson, Galford, Small & DeFries, 2014. Winter crop sensitivity to inter-annual climate variability in central India. Climatic Change 126: 61-76
31
Crop-specific Vulnerability
• A longer wet season followed by higher winter temperatures OR a late and dry monsoon limited water availability through surface irrigation
• Pulses can be grown on residual moisture in rainfed rice fallowlands potential candidate for alternate winter crop
• Some possible adaptation strategies: • Switching to crops less sensitive to heat• shifting planting date• new early maturing crop varieties
DeFries, Mondal, Singh, Agrawal, Fanzo, Remans & Wood, 2016. Synergies and trade-offs for sustainable agriculture: Nutritional yields and climate-resilience for cereal crops in Central India. Global Food Security 11: 44-53
32
• Findings indicate a fluctuating landscape – 2.11 million ha to 3.73 million ha of winter cropped area in central India.
• Seasonal labor migration to nearby towns was found to be associated with less winter crop.
• Increasing irrigation coverage will eventually result in more agricultural intensification.
Fluctuations – where & why?
33
Overall findings
• More irrigation accessibility more winter crop
• Current winter crops might not be climate resilient
• Potential for coarse cereals, along with pulses, needs to be examined under projected conditions
• Ongoing and planned collaborative work with crop-climate modelers
34
Machine Learning for Crop Phenology
Issues:
• Dense time-series data required
• Lack of cloud-free data
• Fine spatial resolution suitable for small farms
Synthetic Aperture Radar (SAR):
• Freely available SAR data
• Advancement of machine learning algorithms
Mondal. Drought and rice intensification in Vietnam: Mapping small farms using time-series ofSentinel-1radar data. In prep.
35
Food and Nutrition Security
0
0.2
0.4
0.6
0.8
1Grain/tuber
Pulse
Nut/seed
Dairy
Meat/fish
Egg
Dark greenleafy veg
Other Vit-Arich veg/fruit
Other veg
Other fruit
Summer
0
0.2
0.4
0.6
0.8
1Grain/tuber
Pulse
Nut/seed
Dairy
Meat/fish
Egg
Dark greenleafy veg
Other Vit-Arich veg/fruit
Other veg
Other fruitWinter
Mondal, DeFries, Harou, Downs, Md. Arif, Gallant, & Fanzo. Implications of Agricultural Intensification for Diet and Nutrition in Rural India. In prep.
Photo: P. Mondal
36
Food and Nutrition Security
Goal: identifying problem nutrients in smallholder farmer diet using Linear Programing Tool
Collaborators: Rutgers, Johns Hopkins, McGill
Photo: P. Mondal
37
Summary
Research focus:
• Using satellite data and GIS to understand smallholder agricultural system
• Human-environment interaction
• Integrated framework Holistic approach
• Developing countries transferable to other regions
Potential collaboration beyond Geography:
• Computer and Information Sciences
• Applied Economics and Statistics
• Plant and soil sciences