Question - Dynamic analysis of yield changes over time
145 views
Skip to first unread message
francesco semeria
May 20, 2021, 5:16:53 PM5/20/21
to MapSPAM
Dear MapSPAM team,
I am a Research Fellow at Fondazione Eni Enrico Mattei (FEEM), a research center based in Italy. I am writing you as, together with my collegues, we would like to make use of the MapSPAM database that your team developed. Our work contributes to RE4AFAGRI project, part of EU-funded LEAP-RE program.
We recently started to use data from MapSPAM to analyse changes in distribution of agricultural practices in Sub-Saharan Africa. We would like to compare outputs of your model at different years (e.g.: 2000 vs 2017) as a first step to capture how changes in production methods (e.g.: from rainfed low-inputs to irrigated high-inputs) would produce changes in yield. This would allow us to better understand the distribution of the yield gap in Sub-Saharan Africa, to thus inform a modelling framework to evaluate areas suitable for production method upgrade, based on potential revenues, costs and environmental sustainability.
Going through the methodology your model is based on, we didn’t find any remark that would discourage such analysis using your outputs at different years as panel data, to analyse changes in yield and production methods. Anyway, we decided to ask you directly, if you could please give us some advice in this sense.
Thank you for your great work and I look forward to hearing from you.
Best regards,
Francesco
Francesco Semeria
Junior Research Fellow
Fondazione Eni Enrico Mattei
Corso Magenta, 63
20123 Milano (MI), Italy
I am a Research Fellow at Fondazione Eni Enrico Mattei (FEEM), a research center based in Italy. I am writing you as, together with my collegues, we would like to make use of the MapSPAM database that your team developed. Our work contributes to RE4AFAGRI project, part of EU-funded LEAP-RE program.
We recently started to use data from MapSPAM to analyse changes in distribution of agricultural practices in Sub-Saharan Africa. We would like to compare outputs of your model at different years (e.g.: 2000 vs 2017) as a first step to capture how changes in production methods (e.g.: from rainfed low-inputs to irrigated high-inputs) would produce changes in yield. This would allow us to better understand the distribution of the yield gap in Sub-Saharan Africa, to thus inform a modelling framework to evaluate areas suitable for production method upgrade, based on potential revenues, costs and environmental sustainability.
Going through the methodology your model is based on, we didn’t find any remark that would discourage such analysis using your outputs at different years as panel data, to analyse changes in yield and production methods. Anyway, we decided to ask you directly, if you could please give us some advice in this sense.
Thank you for your great work and I look forward to hearing from you.
Best regards,
Francesco
Francesco Semeria
Junior Research Fellow
Fondazione Eni Enrico Mattei
Corso Magenta, 63
20123 Milano (MI), Italy
Amir Dadrasi
May 21, 2021, 2:05:24 PM5/21/21
to map...@googlegroups.com
Hi dear Francesco
i study yield gap of potato based on GYGA protocol in Iran condition
for this aim we want to use the mapspam results but this maps have not
high quality than we made the revised maps based on real crop area in
Iran. You can find this maps in this email.
best regards,
Amir
i study yield gap of potato based on GYGA protocol in Iran condition
for this aim we want to use the mapspam results but this maps have not
high quality than we made the revised maps based on real crop area in
Iran. You can find this maps in this email.
best regards,
Amir
Liang
May 21, 2021, 2:06:01 PM5/21/21
to MapSPAM
Dear Francesco -
Thank you for your interest in SPAM. Indeed we did not explicitly say that you could not compare the SPAM results over years. However, the SPAM model is built with the idea of getting the best SPATIAL distribution around the specific year (e.g 2010 is a average between 2009-2011). We did not consider the consistency over the years. For example, we will take the best satellite-derived cropland around e.g. 2010 without considering the methodological consistency between 2005 cropland and 2010 cropland. In this sense, the difference between SPAM 2005 and SPAM 2010 includes (1) the true change in the field (which is what you wanted) (2) the uncertainty of cropland 2005 and 2010 (3) any model improvement of SPAM itself between SPAM 2005 and 2010.
If you are OK with the last two errors/uncertainties (e.g. treating them as a noise), you could indeed compare crop distributions among different SPAM results. I listed my contact below. Let me know if further questions.
All the best,
Liangzhi
Reply all
Reply to author
Forward
0 new messages