Somewhat off-topic comment - For farming, no-till practices allow for successful crops during drought periods. No till seed drills plant seeds directly through grass and weeds. The grass and weeds are then killed with selective herbicides that will not kill the GMO seeded crop. GMO has gotten a bad name by anti-GMO movements, but there are no studies that prove GMO crops or associated herbicides are harmful to people or animals. Of course that doesn't stop the 1-800-BAD-DRUG lawyers from going after deep pockets.
In #10 (Into to Drought), you wrote that there are three kinds of droughts:
1) [reduced] rainfall,
2) [reduced] soil moisture [due to decreased precipitation and increased transevaporation]
3) hydrological (which may be less water in reservoirs, rivers and ground water)
We can measure any reduction in precipitation, and there has probably been a trivial increase in precipitation because warmer air can hold more water. However, I believe you said we had no systematic way of monitoring changes in soil moisture. Above, Spinoni et al 2019 report the change drought in terms in terms of "drought events" per decade, which is not a very robust way to quantify something. More comprehensive monitoring alone can cause an increase in detected events. With limited ability to systematically measure soil moisture, our ability to count the number of times and places that cross the threshold for being counted as a "drought event" can't be very robust. Yield of crops per acre might be a more meaningful measure of the IMPACT of increasing drought AFTER MITIGATION. So might falling levels of ground water in wells. Does the IPCC discuss other metrics for drought.
We have the same problems with in homogeneity in the record of TCs "events". As our surveillance has improved, our count of the number of TCs qualifying as Category 1 hurricanes and Category 3 or 4 ("major" hurricanes) has improved. In Kossin's paper, homogenizing the hurricane record by downgrading the resolution of images to a common low-resolution at the beginning of the satellite period reduced the count of TCs and major TCs in the past two decades by about 30%. Crossing a threshold to be counted as a TC or major TC is less reliable method of quantifying TCs (IMO) than more continuous monitoring such as Accumulated Cyclone Energy. I suspect the same applies to quantifying droughts by "events".
Steve: The above comment reminds me that my understanding of the importance of warming to drought would be improved by information about how much transevaporation increases with temperature. 7%/degC like evaporation at constant relative humidity and windspeed? Less because stomata close when water is short?
All good questions. I'll give brief answers but try and explain more in the next few articles.
1. AR6 does discuss hydrological droughts, I probably should cover what they say at some stage but basically data is limited. Uncertainties are high.
2. In the case of passing a threshold for drought.. of course, any threshold is arbitrary, but in this case it's not "better monitoring means we can detect more droughts" because it's a calculation of trends in "rainfall minus evaporation".
- There are different ways of calculating evaporation, but as you make the calculation more sophisticated you are bringing in "inferred data" from other meteorological variables, and the uncertainties rise a lot. Not sure whether I want to dive into the detail on that.
3. Trends in relative humidity over land have dropped over the last couple of decades. This is a big clue. Temperature rises. Water evaporates. Relative humidity decreasing indicates that there's a bigger atmospheric deficit - the land can't keep up.
4. The expectation is that precipitation over land will rise at less than 7%/degC, something like 3%. Whereas evaporation over land - or in reality, atmospheric demand for evaporation (AED) will rise at 7%. It all gets involved, so more in a later article.
5. Is the ground more arid? Are there less plants? Are plants dying? We will look at this last one in the next article.
Another comment on this, because one of the many open papers on my computer that I'm rereading is "Unraveling the influence of atmospheric evaporative demand on drought and its response to climate change", Sergio M. Vicente-Serrano et al, WIREs Climate Change (2020).
This is a paper cited by AR6 on droughts, so their introduction is interesting:
"There is intense scientific debate on past drought trends and future drought projections.
The increased availability of meteorological records in the last decades has not allowed consensus to emerge regarding long-term drought dynamics and their main drivers.
This is mainly due to the: (a) complexity and impractical definition of drought; (b) different (meteorological, hydrological, agricultural/ecological, and socioeconomic) drought types; and (c) difficulty providing an absolute assessment of the drought severity given the complex implications of the phenomenon.
The debate is particularly intense for future drought projections as uncertainties depend on model representation of biophysical processes, including land–atmosphere feedbacks, and the metrics used for drought assessment."
I'm not sure how anyone can say soil moisture decreases are "bad news" and say they are related to warming, when it's been cooling for the last 8 years.
Yes, models produce very different results especially for precipitation, everyone gets their chance to hazard a guess. eg. Carbon Brief states that Australia year 2100, modeled guesses range from 30% more precipitation to 50% less precipitation. No one guess is better than any other guess. Models aren't independent of their algorithms and biases, that's what makes guessing with models so intriguing.
Somewhat off-topic comment - For farming, no-till practices allow for successful crops during drought periods. No till seed drills plant seeds directly through grass and weeds. The grass and weeds are then killed with selective herbicides that will not kill the GMO seeded crop. GMO has gotten a bad name by anti-GMO movements, but there are no studies that prove GMO crops or associated herbicides are harmful to people or animals. Of course that doesn't stop the 1-800-BAD-DRUG lawyers from going after deep pockets.
In #10 (Into to Drought), you wrote that there are three kinds of droughts:
1) [reduced] rainfall,
2) [reduced] soil moisture [due to decreased precipitation and increased transevaporation]
3) hydrological (which may be less water in reservoirs, rivers and ground water)
We can measure any reduction in precipitation, and there has probably been a trivial increase in precipitation because warmer air can hold more water. However, I believe you said we had no systematic way of monitoring changes in soil moisture. Above, Spinoni et al 2019 report the change drought in terms in terms of "drought events" per decade, which is not a very robust way to quantify something. More comprehensive monitoring alone can cause an increase in detected events. With limited ability to systematically measure soil moisture, our ability to count the number of times and places that cross the threshold for being counted as a "drought event" can't be very robust. Yield of crops per acre might be a more meaningful measure of the IMPACT of increasing drought AFTER MITIGATION. So might falling levels of ground water in wells. Does the IPCC discuss other metrics for drought.
We have the same problems with in homogeneity in the record of TCs "events". As our surveillance has improved, our count of the number of TCs qualifying as Category 1 hurricanes and Category 3 or 4 ("major" hurricanes) has improved. In Kossin's paper, homogenizing the hurricane record by downgrading the resolution of images to a common low-resolution at the beginning of the satellite period reduced the count of TCs and major TCs in the past two decades by about 30%. Crossing a threshold to be counted as a TC or major TC is less reliable method of quantifying TCs (IMO) than more continuous monitoring such as Accumulated Cyclone Energy. I suspect the same applies to quantifying droughts by "events".
Steve: The above comment reminds me that my understanding of the importance of warming to drought would be improved by information about how much transevaporation increases with temperature. 7%/degC like evaporation at constant relative humidity and windspeed? Less because stomata close when water is short?
Frank,
All good questions. I'll give brief answers but try and explain more in the next few articles.
1. AR6 does discuss hydrological droughts, I probably should cover what they say at some stage but basically data is limited. Uncertainties are high.
2. In the case of passing a threshold for drought.. of course, any threshold is arbitrary, but in this case it's not "better monitoring means we can detect more droughts" because it's a calculation of trends in "rainfall minus evaporation".
- There are different ways of calculating evaporation, but as you make the calculation more sophisticated you are bringing in "inferred data" from other meteorological variables, and the uncertainties rise a lot. Not sure whether I want to dive into the detail on that.
3. Trends in relative humidity over land have dropped over the last couple of decades. This is a big clue. Temperature rises. Water evaporates. Relative humidity decreasing indicates that there's a bigger atmospheric deficit - the land can't keep up.
4. The expectation is that precipitation over land will rise at less than 7%/degC, something like 3%. Whereas evaporation over land - or in reality, atmospheric demand for evaporation (AED) will rise at 7%. It all gets involved, so more in a later article.
5. Is the ground more arid? Are there less plants? Are plants dying? We will look at this last one in the next article.
Another comment on this, because one of the many open papers on my computer that I'm rereading is "Unraveling the influence of atmospheric evaporative demand on drought and its response to climate change", Sergio M. Vicente-Serrano et al, WIREs Climate Change (2020).
This is a paper cited by AR6 on droughts, so their introduction is interesting:
"There is intense scientific debate on past drought trends and future drought projections.
The increased availability of meteorological records in the last decades has not allowed consensus to emerge regarding long-term drought dynamics and their main drivers.
This is mainly due to the: (a) complexity and impractical definition of drought; (b) different (meteorological, hydrological, agricultural/ecological, and socioeconomic) drought types; and (c) difficulty providing an absolute assessment of the drought severity given the complex implications of the phenomenon.
The debate is particularly intense for future drought projections as uncertainties depend on model representation of biophysical processes, including land–atmosphere feedbacks, and the metrics used for drought assessment."
I'm not sure how anyone can say soil moisture decreases are "bad news" and say they are related to warming, when it's been cooling for the last 8 years.
The Spinoni 2019 paper is looking at a 70-year trend.
Yes, models produce very different results especially for precipitation, everyone gets their chance to hazard a guess. eg. Carbon Brief states that Australia year 2100, modeled guesses range from 30% more precipitation to 50% less precipitation. No one guess is better than any other guess. Models aren't independent of their algorithms and biases, that's what makes guessing with models so intriguing.