Intensification of tropical agriculture as seen by satellite
Abstract
We present case studies from Latin America and Africa on intensification of tropical agriculture. The Brazilian Amazon of the early 2000s experienced intensification and extensification. We use time series analysis of MODIS vegetation indices to track changes in cropping intensity and crop types over time. The state of Mato Grosso is Brazil's leading producer of soy, corn and cotton. Using 250 m MODIS EVI data and a new decision-tree algorithm tuned to phenological patterns characteristic of Mato Grosso's major natural vegetation and crop rotations, we mapped land-cover across the state over 11 growing seasons (2001-2011). Between 2000 and 2011, a majority of the cultivated land in Mato Grosso transitioned from the cultivation of one commercial crop per growing season (soy or cotton) to two commercial crops (a soy crop followed by a corn or cotton crop). Over our study period, the cultivated area of double cropped land in Mato Grosso steadily increased over 6-fold from .46 million hectares to 2.9 million hectares, 92% of which was in a soy-corn double cropping rotation. In the sub-Saharan country of Malawi, 70% of the land is dedicated to food production yet yields of the primary staple crop, maize, have stagnated around 1 ton ha-1 (developed nations' maize yields are 12-16 tons ha-1). Due to the limited land area, improving yields through intensification is a necessary objective of development. Poverty and food insecurity were widespread and persistent for smallholder farmers cultivating less than 1 hectare of land until the implementation of a government intervention, funded through foreign aid, subsidized allocations of fertilizer and improved seed to small farmers. Since implementation of the policy, the number of food insecure, or people in need of food aid, has decreased from 5 million to half a million people. We present indicators that levels of poverty have decreased since the subsidy. National yields have doubled. Applying modified methods from Brazil, we are able to detect cropland intensification through remote sensing. We present remote sensing analysis of social and economic correlates to changes in yields and build an empirical model of sustainable intensification. Together, these case studies demonstrate that remote sensing techniques can be easily adapted across very different crop types, field sizes and environments.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.B53E..01G
- Keywords:
-
- 0480 BIOGEOSCIENCES Remote sensing;
- 0402 BIOGEOSCIENCES Agricultural systems;
- 1632 GLOBAL CHANGE Land cover change;
- 1640 GLOBAL CHANGE Remote sensing