Information

Dear user, the application need JavaScript support. Please enable JavaScript in your browser.

You are browsing as a GUEST
Title of the item:

Incorporation of Satellite Data and Uncertainty in a Nationwide Groundwater Recharge Model in New Zealand

Title :
Incorporation of Satellite Data and Uncertainty in a Nationwide Groundwater Recharge Model in New Zealand
Authors :
Rogier Westerhoff
Paul White
Zara Rawlinson
Show more
Subject Terms :
groundwater
recharge
uncertainty
satellite
large-scale
nationwide
evapotranspiration
vegetation
model
Science
Source :
Remote Sensing, Vol 10, Iss 1, p 58 (2018)
Publisher :
MDPI AG, 2018.
Publication Year :
2018
Collection :
LCC:Science
Document Type :
article
File Description :
electronic resource
Language :
English
ISSN :
2072-4292
Relation :
http://www.mdpi.com/2072-4292/10/1/58; https://doaj.org/toc/2072-4292
DOI :
10.3390/rs10010058
Access URL :
https://doaj.org/article/4aa96968385644f286e72a974da5ab1a
Rights :
Journal Licence: CC BY
Accession Number :
edsdoj.4aa96968385644f286e72a974da5ab1a
Academic Journal
A nationwide model of groundwater recharge for New Zealand (NGRM), as described in this paper, demonstrated the benefits of satellite data and global models to improve the spatial definition of recharge and the estimation of recharge uncertainty. NGRM was inspired by the global-scale WaterGAP model but with the key development of rainfall recharge calculation on scales relevant to national- and catchment-scale studies (i.e., a 1 km × 1 km cell size and a monthly timestep in the period 2000–2014) provided by satellite data (i.e., MODIS-derived evapotranspiration, AET and vegetation) in combination with national datasets of rainfall, elevation, soil and geology. The resulting nationwide model calculates groundwater recharge estimates, including their uncertainty, consistent across the country, which makes the model unique compared to all other New Zealand estimates targeted towards groundwater recharge. At the national scale, NGRM estimated an average recharge of 2500 m 3 /s, or 298 mm/year, with a model uncertainty of 17%. Those results were similar to the WaterGAP model, but the improved input data resulted in better spatial characteristics of recharge estimates. Multiple uncertainty analyses led to these main conclusions: the NGRM model could give valuable initial estimates in data-sparse areas, since it compared well to most ground-observed lysimeter data and local recharge models; and the nationwide input data of rainfall and geology caused the largest uncertainty in the model equation, which revealed that the satellite data could improve spatial characteristics without significantly increasing the uncertainty. Clearly the increasing volume and availability of large-scale satellite data is creating more opportunities for the application of national-scale models at the catchment, and smaller, scales. This should result in improved utility of these models including provision of initial estimates in data-sparse areas. Topics for future collaborative research associated with the NGRM model include: improvement of rainfall-runoff models, establishment of snowmelt and river recharge modules, further improvement of estimates of rainfall and AET, and satellite-derived AET in irrigated areas. Importantly, the quantification of uncertainty, which should be associated with all future models, should give further impetus to field measurements of rainfall recharge for the purpose of model calibration.
Please log in to access the full text.

We use cookies to help identify your computer so we can tailor your user experience, track shopping basket contents and remember where you are in the order process.