widespread areas (Wilhite 1997, 2016; Western Governors Association 2004; Cutter
and Emrich 2005; Villarini and Smith 2010; Peterson et al. 2013). For instance, the
Mississippi River floods in April and May 2011 were among the largest and most
damaging recorded along the US waterway in the past century; they affected
Missouri, Illinois, Tennessee, Arkansas, Mississippi, and Louisiana (Wikipedia
2017). The crop loss was $60 million from the Birds Point-New Madrid Foodway
levee breach (Brown et al. 2011; Olson and Morton 2012). All of the aforementioned
events caused crop losses in huge agricultural areas, and the belated and incomplete
information becomes the main constraint for efficient decision-making. In order to
address this problem and facilitate the post-disaster response, a number of
agriculture-related geospatial data centers and systems are supposed to provide
relevant information (Lu and Campbell 2009; EDO 2017; GDM 2017). For example,
the US Drought Monitor (USDM) has been developed to provide large-scale drought
information to the general public (USDM 2017).
Despite the great efforts, the current agricultural data systems, either agricultural
disaster monitoring systems or agricultural condition long-term management systems, have been limited in some aspects. The spatial resolution is coarse and limits
the utilization in a more localized scale; the temporal resolution is low and limits the
near-real-time response to the emergency situation. For instance, previously, USDA
NASS used the AVHRR 17 (dead) and AVHRR 18 (aging, and not consistent with
AVHRR 17) NDVI data for monitoring the US crop conditions; the data is low
spatial resolution (1 km) and low temporal resolution (biweekly). Seldom systems
are able to support the on-demand agricultural data customization and information
generation, continuous historical data dissemination and analytics, near-real-time
data and information visualization, and interoperable communication with other
systems. In most agricultural application cases, for the sake of well-balanced spatial
and temporal resolutions, it’s hard to find a perfect Earth observation (EO) data
through a single data source; therefore, multisource EO data, along with the traditional instrumental ground truth data, presents a massive and heterogeneous data
infrastructure for agricultural data and information systems. Above all, current
agricultural information systems are hard to fulfill the increasing demands from
different agencies and communities for sufficient and timely agricultural drought
information.
A successful Web-based agricultural decision support system should flexible,
scalable, reusable, high-efficiency, and user-friendly and provide sufficient and
meaningful capabilities for agricultural decision support. Furthermore, the Web
services provided by the system can be dynamically repurposed and utilized by
any other standard compliable Web clients and applications.
12 Spatial and Temporal Monitoring System for Agriculture
223
and Emrich 2005; Villarini and Smith 2010; Peterson et al. 2013). For instance, the
Mississippi River floods in April and May 2011 were among the largest and most
damaging recorded along the US waterway in the past century; they affected
Missouri, Illinois, Tennessee, Arkansas, Mississippi, and Louisiana (Wikipedia
2017). The crop loss was $60 million from the Birds Point-New Madrid Foodway
levee breach (Brown et al. 2011; Olson and Morton 2012). All of the aforementioned
events caused crop losses in huge agricultural areas, and the belated and incomplete
information becomes the main constraint for efficient decision-making. In order to
address this problem and facilitate the post-disaster response, a number of
agriculture-related geospatial data centers and systems are supposed to provide
relevant information (Lu and Campbell 2009; EDO 2017; GDM 2017). For example,
the US Drought Monitor (USDM) has been developed to provide large-scale drought
information to the general public (USDM 2017).
Despite the great efforts, the current agricultural data systems, either agricultural
disaster monitoring systems or agricultural condition long-term management systems, have been limited in some aspects. The spatial resolution is coarse and limits
the utilization in a more localized scale; the temporal resolution is low and limits the
near-real-time response to the emergency situation. For instance, previously, USDA
NASS used the AVHRR 17 (dead) and AVHRR 18 (aging, and not consistent with
AVHRR 17) NDVI data for monitoring the US crop conditions; the data is low
spatial resolution (1 km) and low temporal resolution (biweekly). Seldom systems
are able to support the on-demand agricultural data customization and information
generation, continuous historical data dissemination and analytics, near-real-time
data and information visualization, and interoperable communication with other
systems. In most agricultural application cases, for the sake of well-balanced spatial
and temporal resolutions, it’s hard to find a perfect Earth observation (EO) data
through a single data source; therefore, multisource EO data, along with the traditional instrumental ground truth data, presents a massive and heterogeneous data
infrastructure for agricultural data and information systems. Above all, current
agricultural information systems are hard to fulfill the increasing demands from
different agencies and communities for sufficient and timely agricultural drought
information.
A successful Web-based agricultural decision support system should flexible,
scalable, reusable, high-efficiency, and user-friendly and provide sufficient and
meaningful capabilities for agricultural decision support. Furthermore, the Web
services provided by the system can be dynamically repurposed and utilized by
any other standard compliable Web clients and applications.
12 Spatial and Temporal Monitoring System for Agriculture
223